<%BANNER%>

Being the Best Bearer of Bad Tidings

Permanent Link: http://ufdc.ufl.edu/UFE0021952/00001

Material Information

Title: Being the Best Bearer of Bad Tidings The Bad News Response Model
Physical Description: 1 online resource (140 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bad, communication, coping, decisions, news, responses
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Giving bad news is an unpleasant task, and the medical literature provides numerous guidelines for giving bad news well. However, what people mean by 'giving bad news well' is less clear. What should be the goal when communicating bad news? I suggest that the goal of news-givers should be to guide recipients toward a desired response. I propose a theoretical framework, the Bad News Response Model, for delivering bad news that draws from research in health and social psychology. The model is applicable to all forms of bad news and specifies that three characteristics of the news (controllability, likelihood, and severity) influence which response (watchful waiting, active change, or acceptance) will most often lead to the best quality of life for the recipient. Our studies examined three questions related to the Bad News Response Model: * How does the type of bad news affect response choices? (Studies 1 and 2) * How do people feel about responses to bad news? (Study 3) * How do people view the effectiveness of responses to bad news? (Study 4) Results suggest that response choices are predictable based on the controllability, likelihood, and severity of news outcomes, as suggested by the model (Studies 1 and 2). In addition, results revealed that people may also make response choices based on the beliefs they have about response options (Study 3) and on the perceived effectiveness of each response (Study 4).
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Shepperd, James A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021952:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021952/00001

Material Information

Title: Being the Best Bearer of Bad Tidings The Bad News Response Model
Physical Description: 1 online resource (140 p.)
Language: english
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2008

Subjects

Subjects / Keywords: bad, communication, coping, decisions, news, responses
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Giving bad news is an unpleasant task, and the medical literature provides numerous guidelines for giving bad news well. However, what people mean by 'giving bad news well' is less clear. What should be the goal when communicating bad news? I suggest that the goal of news-givers should be to guide recipients toward a desired response. I propose a theoretical framework, the Bad News Response Model, for delivering bad news that draws from research in health and social psychology. The model is applicable to all forms of bad news and specifies that three characteristics of the news (controllability, likelihood, and severity) influence which response (watchful waiting, active change, or acceptance) will most often lead to the best quality of life for the recipient. Our studies examined three questions related to the Bad News Response Model: * How does the type of bad news affect response choices? (Studies 1 and 2) * How do people feel about responses to bad news? (Study 3) * How do people view the effectiveness of responses to bad news? (Study 4) Results suggest that response choices are predictable based on the controllability, likelihood, and severity of news outcomes, as suggested by the model (Studies 1 and 2). In addition, results revealed that people may also make response choices based on the beliefs they have about response options (Study 3) and on the perceived effectiveness of each response (Study 4).
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis: Thesis (Ph.D.)--University of Florida, 2008.
Local: Adviser: Shepperd, James A.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2010-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2008
System ID: UFE0021952:00001


This item has the following downloads:


Full Text

PAGE 1

BEING THE BEST BEARER OF BAD TIDINGS: THE BAD NEWS RESPONSE MODEL By KATHARINE M. SWEENY A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008 1

PAGE 2

2008 Katharine M. Sweeny 2

PAGE 3

ACKNOWLEDGMENTS I thank James Shepperd for his constant guidance and encouragement and for never allowing me to do less than my best work. I th ank my parents for their unconditional love and support and for being my favorite and most enth usiastic cheerleaders. Finally, I thank Ryan Johnson, who manages to fill even the most ar duous process with laughter and who makes each day worth waking up for. 3

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS ............................................................................................................... 3 LIST OF TABLES ...........................................................................................................................7 LIST OF FIGURES .........................................................................................................................8 ABSTRACT ...................................................................................................................... ...............9 CHAPTER 1 INTRODUCTION ................................................................................................................ ..10 The Bad News Response Model .............................................................................................11 How Can People Respond to Bad News? ........................................................................12 Watchful waiting ......................................................................................................12 Active change ...........................................................................................................14 Acceptance ...............................................................................................................15 Non-responding ........................................................................................................16 Which Responses Should News-Givers Suggest? ...........................................................18 Controllability ..........................................................................................................19 Likelihood ................................................................................................................20 Severity .....................................................................................................................2 1 Communicating Desired Responses ................................................................................21 Summary and Implications .....................................................................................................22 2 REVIEW OF THE LITERATURE ........................................................................................26 Giving Bad News Well .......................................................................................................... .28 Decreasing the News-givers Discomfort .......................................................................28 Providing Sufficient Information ....................................................................................30 Promoting Recipient Satisfaction ....................................................................................31 Improving Memory and Understanding ..........................................................................32 Reducing Recipients Distress .........................................................................................34 Promoting Hope ...............................................................................................................35 Guiding Recipients Towa rd Desired Responses .............................................................37 The Bad News Response Model .............................................................................................39 How Can People Respond to Bad News? ........................................................................40 Watchful waiting ......................................................................................................41 Active change ...........................................................................................................43 Acceptance ...............................................................................................................44 Non-responding ........................................................................................................47 How Can People Respond? Summary and Conclusions .................................................49 Which Responses Should News-Givers Suggest? ...........................................................51 4

PAGE 5

Controllability ..........................................................................................................52 Likelihood ................................................................................................................53 Severity .....................................................................................................................5 5 Communicating Desired Responses ................................................................................56 Summary, Critique, and Future Directions ......................................................................57 Strengths of the model ..............................................................................................58 Limitations of the model ..........................................................................................59 Future directions .......................................................................................................61 Coda .......................................................................................................................... ..............63 3 STUDY 1 ..................................................................................................................... ...........65 Methods ..................................................................................................................................65 Participants .................................................................................................................. ....65 Procedure ..................................................................................................................... ....65 Results .....................................................................................................................................66 Manipulation Checks .......................................................................................................66 Likert-Type Ratings .........................................................................................................67 Watchful waiting ......................................................................................................67 Active change ...........................................................................................................68 Acceptance ...............................................................................................................69 Forced-Choice Responses ...............................................................................................70 Discussion .................................................................................................................... ...........71 4 STUDY 2 ..................................................................................................................... ...........77 Methods ..................................................................................................................................78 Participants .................................................................................................................. ....78 Procedure ..................................................................................................................... ....79 Results .....................................................................................................................................80 Manipulations Checks .....................................................................................................80 Likert-Type Ratings .........................................................................................................81 Watchful waiting ......................................................................................................81 Active change ...........................................................................................................81 Acceptance ...............................................................................................................83 Forced-Choice Responses ...............................................................................................83 Does News-Giving Experience Matter? ..........................................................................84 Discussion .................................................................................................................... ...........86 5 STUDY 3 ..................................................................................................................... ...........92 Methods ..................................................................................................................................94 Participants .................................................................................................................. ....94 Procedure ..................................................................................................................... ....94 Results .....................................................................................................................................95 5

PAGE 6

Affect Items .................................................................................................................. ...95 Trait Items ................................................................................................................... ....96 Discussion .................................................................................................................... ...........97 6 STUDY 4 ..................................................................................................................... .........100 Methods ................................................................................................................................100 Participants .................................................................................................................. ..100 Procedure ..................................................................................................................... ..101 Results ...................................................................................................................................102 Situational Factors and Response Ratings .....................................................................102 Consequences of Responding ........................................................................................103 Likert-type ratings of responding ...........................................................................103 Forced-choice measures of responding ..................................................................104 Situational factors and c onsequences of responding ..............................................106 Mediation Analyses .......................................................................................................107 Discussion .................................................................................................................... .........108 7 GENERAL DISCUSSION ...................................................................................................111 Implications .................................................................................................................. ........113 Limitations and Future Directions ........................................................................................114 Conclusions ...........................................................................................................................117 APPENDIX A STUDY 1 SAMPLE QUESTIONNAIRE ............................................................................118 B STUDY 2 SAMPLE QUESTIONNAIRE (PATIENT CONDITION) ................................121 C STUDY 2 SAMPLE QUESTIONNAIRE (PHYSICIAN CONDITION) ............................123 D STUDY 3 SAMPLE QUESTIONNAIRE ............................................................................125 E STUDY 4 QUESTIONNAIRE .............................................................................................126 LIST OF REFERENCES .............................................................................................................130 BIOGRAPHICAL SKETCH .......................................................................................................140 6

PAGE 7

LIST OF TABLES Table page 1-1 Impact of situational fact ors on appropriate responding ....................................................25 2-1 Characteristics of the three response categories ................................................................64 3-1 Marginal means for 3-wa y interactions (Study 1) .............................................................75 3-2 Study 1 frequency analyses of response choices ...............................................................76 4-1 Study 2 frequency analyses of response choices ...............................................................89 5-1 Predicted results (Study 3) ............................................................................................... ..99 5-2 Affect category ratings across response types ...................................................................99 5-3 Impression category ratings across response types ............................................................99 6-1 Correlations between situa tional factors, responses, and response consequences ...........110 7

PAGE 8

LIST OF FIGURES Figure page 1-1 Bad news response model ..................................................................................................2 4 3-1 Responses of watchful waiting as a function of the in teraction between controllability and likelihood (Study 1). ............................................................................74 3-2 Responses of active change as a functi on of the interaction be tween controllability and severity (Study 1). ....................................................................................................... 74 3-3 Responses of acceptance as a function of the interaction between likelihood and severity (Study 1). ........................................................................................................... ...75 3-4 Responses of acceptance as a function of the interaction between controllability and severity (Study 1). ........................................................................................................... ...75 4-1 Responses of active change as a function of the interaction be tween controllability and severity (Study 2). ....................................................................................................... 88 4-2 Responses of active change as a function of the interaction between role and severity (Study 2).............................................................................................................................88 4-3 Suggestions of watchful waiting by part icipants reporting low, moderate, and high news-giving experience and for low a nd high controllability (Study 2). ..........................90 4-4 Suggestions of acceptance by participants reporting low, moderate, and high newsgiving experience and for low and high controllability (Study 2). ....................................90 4-5 Suggestions of acceptance by participants reporting low, moderate, and high newsgiving experience and for low and high severity (Study 2). ..............................................91 8

PAGE 9

9 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy BEING THE BEST BEARER OF BAD TIDINGS: THE BAD NEWS RESPONSE MODEL By Katharine M. Sweeny May 2008 Chair: James A. Shepperd Major: Psychology Giving bad news is an unpleasant task, a nd the medical literature provides numerous guidelines for giving bad news well. However, wh at people mean by giving bad news well is less clear. What should be the goal when communicating bad news ? I suggest that the goal of news-givers should be to guide recipients toward a desired re sponse. I propose a theoretical framework, the Bad News Response Model, for deliv ering bad news that draws from research in health and social psychology. The model is applicable to all forms of bad news and specifies that three characteristics of the news (controlla bility, likelihood, and seve rity) influence which response (watchful waiting, active change, or accept ance) will most often lead to the best quality of life for the recipient. Our studies examined three questions related to the Bad News Response Model: How does the type of bad news affect response choices? (S tudies 1 and 2) How do people feel about res ponses to bad news? (Study 3) How do people view the effectiveness of responses to bad news? (Study 4) Results suggest that response choices are predic table based on the cont rollability, likelihood, and severity of news outcomes, as suggested by th e model (Studies 1 and 2). In addition, results revealed that people may also make response choices based on the beliefs they have about response options (Study 3) and on the perceive d effectiveness of each response (Study 4).

PAGE 10

CHAPTER 1 INTRODUCTION Giving bad news is an unpleasant task (R osen & Tesser, 1970). Unfortunately, most people must transmit bad news at some point in life, and for many people (e.g., health care professionals, police officers, business manage rs) giving bad news is part of their job descriptions. Although giving bad news is uncomfo rtable for the giver, in many cases it is important that people give bad news well. A nu mber of researchers in the medical field have provided guidelines to help those who must give bad news, but what people mean by giving bad news well is less clear. Some articles focus on th e goal of providing hope to the recipient of the news, others focus on making the transmission eas ier for the news-giver, and others focus on increasing recipients sa tisfaction with the bad news transmission. I propose that news-givers shoul d focus on guiding recipients towards a desired response. For example, a physician giving news of cancer sh ould give the news in such a way that the patient understands the best course of action and is likely to e ngage in the desired response. Furthermore, I propose the Bad News Response Mode l as a framework to aid the givers of bad news in determining the best response. The model is applicable to all forms of bad news and specifies that three characterist ics of the outcomes of bad news (controllability, likelihood, and severity) predict that one of three responses (Watchful waiting, active change, or acceptance) will most often lead to the best quality of life for the recipient of the news (Figure 1-1). This paper presents four st udies that address the following questions related to the Bad News Response Model: How does the type of bad news affect response choices? Although other re searchers have made broad suggestions as to how peopl e should respond to bad news, no studies systematically examine how various situati onal factors influence how people give and respond to bad news. Studies 1 and 2 examined the relationship between aspects of bad news and response choices. 10

PAGE 11

How do people feel about responses to bad news? Study 3 examines why people might prefer certain responses over others, particularly addressi ng how people perceive various response options. We propose that although different responses may be best suited to different types of news, people are more lik ely to choose some responses than others, regardless of the circumstances. Study 3 examin ed the possibility that people may associate particular emotions with each response a nd anticipate making particular impressions by responding to bad news in each way. How do people view the effectiveness of responses to bad news? Study 4 examined yet another reason people may prefer some responses over others: some responses may be more effective than others for producing pos itive outcomes. Study 4 examined peoples perceptions of how their responses to bad news affected their outcomes. The Bad News Response Model As mentioned earlier, we propose that giving bad news well is defined as guiding newsrecipients toward desired responses responses th at news-givers believe will result in the best long-term outcomes for recipients. The goal of guiding recipients toward the most effective responses prompts two questions. First, what ar e the different ways people can respond to bad news? Second, which responses should bad news-givers suggest? I developed the Bad News Response Model to answer these tw o questions. I propose that all re sponses to bad news fall into one of four categories: Watc hful waiting, active change, acceptance, and non-responding. I further suggest that three fact ors of the outcomes of bad news (controllability, likelihood, and severity) indicate which response is like ly to be effectiv e (Figure 1-1). The Bad News Response Model suggests that giving bad news well involves guiding news-recipients towards a desired response. Thus the model is aimed both at the person who must give bad news and at the recipient of the news. Ultimately, the goal of the Bad News Response Model is to elicit a desired response from the recipient of bad news, but bad newsgivers must evaluate the characteristics of th e possible bad outcome that determine what the desired response should be. Bad news-givers can th en tailor their communica tion of bad news to 11

PAGE 12

encourage the desired response from the recipien t. In addition, recipients of bad news can individually evaluate their situ ation and determine the most effective response to the news. It is important to note that the Bad News Response Model does not attempt to precisely predict the response that will lead to the best quality of life. The model provides guidance for bad news-givers as to which responses may be best under different circumstances, but the model addresses a wide spectrum of bad news and must make generalizations based on situational factors. Bad news-givers and reci pients should choose the response th at is most likely to result in the best outcomes, given their assessment of the situational factors. In addition, the Bad News Response Model focuses on what lies in the future as a result of the bad news and not on the event that has passed. The model addresses responses to bad ne ws and the outcomes of those responses on the future. Thus, the situational fact ors in the model do not pe rtain to the event that is being disclosed, but rather to th e possible results of that event. How Can People Respond to Bad News? The Bad News Response Model suggests that people can respond to ba d news in one of four ways: watchful waiting, active change, acceptance, and non-responding. These response categories broadly apply to many kinds of bad news, although the specif ic nature of each response may differ across domains. For example, a patient who responds to a diagnosis of cancer with active change will engage in differe nt specific behaviors than will a student who responds to a failed exam with active change. Ho wever, I suggest that these two responses will be similar in fundamental ways. Watchful waiting The first category represents a relativel y passive form of responding. The medical literature has used the term watchful waiting as a specific cont rast to aggressive treatment options (de Haes & Koedoot, 2003). Here, Watchful waiting indicates a more general wait and 12

PAGE 13

see mentality regarding the bad news. The term watchful emphasizes that people engaged in this response are aware that they are facing a possible thre at and are vigilant to changes in their situation. However, they maintain the status quo rather than take action. To illustrate, consider a man diagnosed with prostate cance r. The man is in his late 80s, a widower, and has few financial responsibilities. Although this man registers a nd accepts his diagnosis of cancer, he may choose not to get a second opinion or unde rgo treatment but instead go on with his life largely as if nothing had changed. He may make annual appointme nts to reassess his response, but otherwise his life remains as it wa s prior to his diagnosis. Watchful waiting bears similarity to the ge neral conceptualizati on of emotion-focused coping (Folkman & Lazarus, 1980), although the sp ecific characterization of emotion-focused coping differs widely between studies (Carver, Scheier, & Weintraub, 1989). The similarity resides in the fact that both watchful waiting and emotion-focused coping focus on distraction and emotional regulation. Emotion-focused copi ng entails directing energy toward managing anxiety and other negative emoti ons arising from a stressful situ ation rather than engaging in active intervention. People in the watchful waiting category may engage in act ivities that distract them from the bad news. Behaviors that are design ed to take ones mind off of a threat may be beneficial if no actions will make a difference, or if action would be too costly or if dwelling on the threat is counterpr oductive (Lazarus, 1985). However, watchful waiting differs from em otion-focused coping in a fundamental way. Emotion-focused coping is not mutually exclus ive with other more ac tive forms of responding, and in fact people may engage in emotion-focuse d coping in all four re sponse categories of the Bad News Response Model. Emotion-focused coping complements all forms of responding by reducing the intensity of stressful emotions a nd allowing people to ga in perspective on their 13

PAGE 14

situation (Folkman & Lazarus, 1980). In contrast, watchful waiting involves a specific set of behaviors and emotions that represen t one way of responding to bad news. Active change Active change represents the most vigorous, engaged form of responding. Unlike the distraction or irrelevant activity characterizing watchful waiting, active change involves specific responses directed toward addressing the bad news. Active change aligns most clearly with traditional views of productive coping strategies, such as problem-focused coping, that directly address the negative situation. Pr oblem-focused coping in part i nvolves taking action to solve a problem or change a negative situation (Carve r et al., 1989; Folkman & Lazarus, 1980; Folkman & Lazarus, 1985). Active change includes three types of be havior: information-seeking, prevention, and treatment. Information-seeki ng serves two purposes. First, information-seeking provides recipients of bad news with the information th ey need to make decisions about how to respond. Second, information-seeking serves to connect recipi ents with others who have dealt with similar experiences and provides a networ k of support. Of note, other re searchers have discussed these roles of information-seeking as part of pr oblem-focused or active coping (Aldwin & Revenson, 1987; Lazarus, 1981; Lazarus & Launier, 1978). The terms prevention and treatment have medical connotations, but in this context they broadly refer to behaviors directed towa rd preventing the situation from deteriorating (maintenance) and treating an undesirable situ ation that has emerged (improvement). To illustrate, consider a different man diagnosed with prostate cancer. This man is in his late 40s, has a wife and several children, and is the primar y breadwinner for the family. Unlike the man in his 80s who chooses watchful waiting, the s econd man may be very willing to undergo chemotherapy and radiation in hopes that it will eradicate the cancer and allow him to live a full 14

PAGE 15

and long life with his family. He should activ ely investigate his condi tion, perhaps seeking a second opinion or researching pros tate cancer online or at the lib rary, and undergo preventative and/or aggressive measures to prolong his life. Acceptance Acceptance is the third and most complex fo rm of responding. I view acceptance as action towards acceptance rather than passive re signation. People who respond to bad news with acceptance do not necessarily collapse in a heap although this response may be unavoidable at first. Instead, they eventually direct their energy towards moving forward and addressing any consequences of the bad news. Acceptance involves looking beyond the negative outcomes to the possibility for hope that lies in the future. Even in the case of i mminent death, people can find hope in living life to the fu llest during their remaining time and dying with dignity (Dean, 2002). Acceptance is not a final, static state of resignation; instead, it involves an ongoing positive process of making the best of a bad situation. Acceptance involves two types of behavior : information-sharing and accommodation. Information-sharing involves telling others a bout the negative event, although the extent of sharing with others may vary depending on the news. For example, certain types of bad news, such as testing HIV-positive, may stigmatize th e individual, and people may want to limit their information-sharing to close friends and family. Furthermore, the effectiveness of informationsharing depends in part on the receptiveness of the listener (Harber & Pennebaker, 1992; Kelly & McKillop, 1996). Information-sharing serves three purposes. Firs t, information-sharing helps people accept the negative event by making the event part of th eir social reality. Second, information-sharing elicits social support from frie nds and family. Researchers have distinguished between seeking social support for emotional reasons vs. seeking social support for practical reasons (advice, 15

PAGE 16

assistance, etc.; Carver et al., 1989). Acceptance focuses more on the emotional side of social support, rather than the more ac tive, change-focused practical side. Third, information-sharing seems to serve an important function in an e nd unto itself (Pennebake r, 1988; Pennebaker, Colder, & Sharp, 1990; Pennebaker & OHeeron, 1984; Pennebaker, Zech, & Rim, 2001; Spera, Buhrfeind, & Pennebaker, 1994). Accommodation involves making changes, not to affect the news-specific outcome, but rather to incorporate the negative event into ones life. When a negative outcome is uncontrollable, accommodation focuses peoples en ergy on productive activity rather than futile efforts to change the outcome. For example, in most cases a woman who receives a rejection letter from her first choice graduate school should consider alte rnative schools or career plans rather than continuing to pursu e admission at the school that re jected her. Accommodation often involves behavioral changes su ch as cutting back on strenuou s activities in the case of a debilitating disease or putting away a lost loved ones personal item s in the case of a death in the family. It often also involves cognitive changes that entail looking for reasons why the tragedy occurred (sense-making) and focusing on positiv e changes resulting from the tragedy (benefitfinding) (Davis, Nolen-Hoeksema, & Larson, 1998; Rabow & McPhee, 1999). Non-responding The fourth category of responding captures a number of responses. Lubinsky (1994) distinguishes between four forms of non-responding: denial, disbel ief, deferral, and dismissal. Although the four may appear simila r, their sources differ. Denial is form of repression brought on as a defense mechanism. It involves vehement disagreement with any disliked information, even when evidence makes it clear that the info rmation is correct, and is a relatively rare response to bad news. Disbelief is marked by confusion rather than rejection of bad news and may result from a desire to maintain hope for longer than is warranted. Deferral is marked by 16

PAGE 17

avoidance of information about bad news as a re sult of inadequate resources to cope with the situation. People responding with deferral may accep t the basis for bad news (e.g., results of a medical test) but reject or ignore the implications of those findings (i.e., the necessity of lifestyle changes or treatment). Finally, dismissal is marked by anger at the bad news-giver and denial of the news-givers competence or legitimacy. These four reactions, though different in significant ways, all fall into the res ponse category of non-responding. Non-responding is distinct from watchful waiting. Non-responding is not an attempt to reduce anxiety about bad news wh ile acknowledging it, but rather an attempt to pretend nothing has happened or wish away the bad news. Furt hermore, non-responding ma y be most likely to occur in situations when acceptance is called for. Both watchful waiting and active change are somewhat attractive responses: one allows people to monitor the news and defer action until it is appropriate, and the other involve s taking action to change thi ngs for the better (de Haes & Koedoot, 2003). Acceptance, in contrast, requires people to face the news head-on and does not offer the hope that things will turn out well. Although acceptance is necessary when a very bad outcome is unavoidable, people may prefer to embrace non-responding instead. Non-responding can feel good for a short time because it allows people to pretend that nothing has changed for the worse, but people eventually must face negati ve outcomes, such as the death of a loved one or a terminal illness, and cope with the consequences. On the other hand, non-responding may be an acceptable response in the short-term. A number of researchers and physic ians note that denial is a ne cessary response for some people under certain circumstances, and a number of theo rists have argued that news-givers should not force recipients to face bad news before they ar e ready (Bor et al., 1993; Faulkner, 1998; Greer et al., 1979; Radziewicz & Baile, 2001). As indicated in Figure 1-1, the Bad News Response Model 17

PAGE 18

indicates that non-responding is a legitimate but generally undesired response. News-givers may recognize that recipients are likely to engage in non-respondin g at first, but the Bad News Response Model suggests that the go al of the news-giver is to gui de people towards the response that will lead to the best long-term outc omes. Although non-responding may be functional at first, people must eventually face bad news and choose a different response. Which Responses Should News-Givers Suggest? The Bad News Model suggests that bad news -givers should guide recipients towards desired responses; the model does no t attempt to predict with certain ty the best responses to bad news. The model ultimately relies on news-givers to determine the response that will lead to the best outcomes for recipients and then guide the recipients towards that response. However, bad news varies on a number of predictable dime nsions, and research suggests that certain dimensions may lead one response to be more effective than others, depending on the situation. Specifically, examination of the vast literature on risk perception, heal th behavior, and coping reveals three factors that repeatedly emerge as playing a particularly im portant role in peoples responses to the possibility of bad news and othe r stressful situations: the controlla bility of negative outcomes, the likelihood of negative outcomes, and the seve rity of negative outcomes. Table 1-1 presents a summary of the respons es that may be most effective for each combination of high and low controllabilit y, likelihood, and severity. These suggestions represent the responses that seem most likely to be effective under different circumstances, in light of existing research on both responses to ba d news and situational factors of the news. In general, we suggest that people should engage in active change when two or three of the situational factors are high (high control, hi gh likelihood, and/or high severity) and watchful waiting when two or three of the situational fact ors are low (low control, low likelihood, and/or low severity). The only exception occurs when likelihood and severity are high but control is 18

PAGE 19

low. Under these circumstances, when severe ne gative outcomes are highly likely and little or nothing can be done to change the outcomes, we suggest that acceptance is the best response. Controllability The first factor that may influence effective responding to bad news is the controllability of the negative outcomes that may result from ba d news. The ability to control the outcomes of bad news varies greatly across di fferent situations. For example, a student who discovers s/he is failing a course several weeks befo re the semesters end may be able to improve his or her grade by completing extra credit assignments, getting he lp from the professo r, or studying long and hard for the final exam. However, as the semester draws to a close, cont rol over the co urse grade diminishes, and once final course grades are tu rned in, there may be no remaining avenues to affect the outcome of the course. Controllability plays a significant role in pred icting peoples responses to threat (Aspinwall & Taylor, 1997; Becker, 1974; Floyd, Prenti ce-Dunn, & Rogers, 2000; Janz & Becker, 1984; Kirscht, 1988; Maddux & Rogers, 1983; Rogers 1983). Research on coping shows that the controllability of a stressful s ituation affects the strategies pe ople choose to adopt when coping with stressful situations (Folkman & Lazarus, 19 80). When people perceive event controllability to be high, they tend to adopt active coping strategies; when people perceive event controllability to be low, they tend to adopt strategies dire cted towards managing their emotions (Aldwin, 1991; Carver, Scheier, & Weintraub, 1989). Other studie s find that the effectiveness of various coping strategies depends in large part on the controlla bility of the stressful situation, with active strategies proving most beneficial when the situ ation is controllable (A ldwin & Park, 2004; Park, 2001; Park, Armeli, & Tennen, 2004). 19

PAGE 20

Likelihood The second factor in determining the appropria te response to bad ne ws is the likelihood of possible negative outcomes. Bad news does not always indicate a guaranteed negative outcome. For example, a boss may have to info rm employees that the company must downsize without knowing who will lose their jobs. Physicians frequently give bad news that indicates the possibility of illness or injury based on initial evidence without the ability to diagnose a problem with complete certainty. For the purposes of the model, likelihood refers to how likely negative outcomes are to occur if the news recipient does not act to preven t them. For example, the likelihood that a suspicious lump indicates cancer should be eval uated irrespective of treatment options or the patients intentions to seek treatment. As suc h, likelihood is distinct fr om controllability. People may reduce the likelihood of negative outcomes by their response to bad news, but the initial evaluation of likelihood is sepa rate from controllability. Likelihood influences responding in two ways. First, and most intuitive, people consider the likelihood of a negative outcome in weighing th e costs and benefits of an effortful and costly response. (Ajzen & Fishbein, 1980; Becker, 1974; Edwards, 1954; Rogers, 1983). Second, and less intuitive, the initial percep tion of likelihood of a negative outco me influences later affective reactions should the worst actual ly occur. Negative outcomes ar e unpleasant in their own right, but they are particularly unpleasant when they are unexpected (van Dijk & van der Pligt, 1997; Shepperd & McNulty, 2002). People who respond as if a negative outcome is unlikely to occur may have a particularly unpleasant experience if the outcome does o ccur, more so than if they had expected the worst. This finding suggests th at people may benefit not only in terms of literal preparation, but also in terms of affective preparation by engaging in more active responses if the negative outcome is likely to occur. 20

PAGE 21

Severity The third factor in determining the appropriate response to bad news is the severity of the possible negative event. Bad news varies in term s of how important or co nsequential the possible negative outcome is. Clearly, a woman who learns that she is at risk for heart burn is hearing very different news than a woman who learns that she is at risk for a heart attack, and both the news-giver and the recipient of the news should proceed differently in these two situations. Of course, even news that has re latively non-severe consequen ces can be bad. The woman who learns she has a high risk for heart burn may have to make significant diet ary and other lifestyle changes. However, her response will differ in many ways from the woman learning of her risk for heart attack. People naturally account for the severity of potential negative outcomes when they anticipate and respond to bad news. The severity of potential hea lth outcomes predict whether people will engage in preventative health be havior (Becker, 1974; Rogers, 1983), and research on coping finds that people choos e active coping strategies when they judge the event to be highly stressful or important (Anderson, 1977; Parkes, 1986; Terry, 1991). In addition, research on bracing for bad news finds that people onl y embrace a negative outlook for outcomes or consequences that are importa nt (Shepperd, Findley-Klein, Kw avnick, Walker, & Perez, 2000). If possible negative outcomes are inconsequential or non-severe, news-recipients gain more from choosing relatively passive responses (watchful wa iting) than from engaging in physically or emotionally active responses (a ctive change or acceptance). Communicating Desired Responses Although the goal of the Bad News Response Mode l is not to elucidate specific details of the communication of bad news, the model suggests that the bad news-giver direct the recipient towards desired responses and offers insights in to which responses may be most effective in 21

PAGE 22

different situations. The direction on behalf of the bad news-giver can encourage people to respond in the most effective way even in the face of problems with comprehension, arousal, education, etc. When preparing to give bad news, the communi cator can evaluate th e news situation in terms of the likelihood, severit y, and controllability of negative outcomes and direct the communication towards encouraging the recipient to e ngage in the response that is most likely to be effective. The details of su ch direction will differ greatly de pending on the specific topic and nature of the bad news, but these broad generaliz ations should be effec tive across a variety of domains and situations. The Bad News Res ponse Model does not recommend that bad newsgivers manipulate the recipien t into responding in a particular way using whatever means necessary. Rather, news-givers should present all possible responses and the costs and benefits of each, and then give their opinion regarding the best possible res ponse (Epstein, Alper, & Quill, 2004). Furthermore, recipients of bad news can us e the model to evaluate their news and choose the best response when the news-giver is unable to guide them appropriately. Summary and Implications The medical literature suggests a number of goa ls to help people gi ve bad news well, but we propose that giving bad news well should be defined as guiding news-recipients towards desired responses. The Bad News Response Model suggests that news-givers can look to situational factors (con trollability, likelihood, and severity ) to determine which of three responses (watchful waiting, active change, and accep tance) is most likely to be effective. The Bad News Response Model can serve seve ral important purposes. First, the model can assist bad news-givers who otherwise must rely on their own limited experience or personal motivations when giving bad news. The Bad News Response Model provides a goal for bad 22

PAGE 23

news transmission that can reduce the impact of news-givers concerns on their news-giving strategies by guiding them toward s recipient-focused strategies. Second, bad news-givers can use the model to evaluate their transmission of news after the fact. If news-givers observe recipients maki ng an undesired response, they can examine their transmission strategy in light of the model. The news-giver may have incorrectly assessed one or more of the situational factors, or the suggestion of the best response may have been ineffective. For example, physicians may be unaware of their patients financial circumstances, and this lack of information could result in misjudgment of th e severity and/or controllability of patients medical conditions. Physicians might assume that expensive treatments are feasible when in fact the patient does not have insurance or the mean s to pay for the treatments, making the prognosis relatively uncontrollable. Third, bad news recipients can use the model to evaluate their responses to bad news, apart from the giver. After receiving bad news, recipients can use the model to determine the most appropriate response by evaluating the li kelihood, severity, and c ontrollability of the possible outcomes. For example, a woman who lear ns of upcoming layoffs at work can consider the likelihood that she will lose her job, how bad the consequences of a job loss would be, and if she has control over whether she is laid off. Having evaluated the situation, she may have a better sense of the most effective response. This process ma y help people to override responses based solely on anxiety or fear. In addition, recipients who find th at their response to some news is ineffective can reexamine the situational factors involved and possibly adjust their responses accordingly. If the woman facing a possible job loss responds with active change and then finds that she is making no progress towards keeping her job, she may deci de to shift towards acceptance by checking the want ads and telling her family about the layoffs. 23

PAGE 24

Controllability Likelihood Severity Guide recipient of the news towards a desired response Acceptance Maximize quality of life Non-responding Watchful waiting Active change Transmission Goal Desired Response Situational Factors Desired Outcome Figure 1-1. Bad news response model Reevaluate response effectiveness over time 24

PAGE 25

25 Table 1-1. Impact of situationa l factors on appropriate responding Low likelihood High likelihood Low severity Low control Watchful wa iting Watchful waiting High control Watchful waiting Active change High severity Low control Watchful waiting Acceptance High control Active change Active change

PAGE 26

CHAPTER 2 REVIEW OF THE LITERATURE God grant me the serenity to accept the thi ngs I cannot change, courage to change the things I can, and wisdom to know the differe nce (Serenity Prayer, source uncertain). In the book The Anatomy of Hope (Groopman, 2004), an oncologi st recounts the stories of two patients whose prognoses permitted little hope. The first patients physician repeatedly misled her into thinking that a cure was likely rather than providing more accurate information about her inevitable fate. When the cancer was a bout to take her life, the patient expressed dismay at the false hope promoted by her trusted physician. In contrast, the second patients physician conveyed the gravity of her situation at each step of the cancers progression. This patient lived her last moments to the fullest a nd died with little regret or dismay. Although the two diagnoses were equally dire, the patients experiences were de cidedly different. Such stories emphasize the role of those charged with givi ng bad news in providing the recipients with wisdom to know when their situation can be changed and when the situation simply calls for serene acceptance. These anecdotes suggest the need for a systematic model of giving and responding to bad news. Giving bad news is an unpleasant task (R osen & Tesser, 1970). Unfortunately, most people must transmit bad news at some point in their lives. They may have to break up with a lover or tell a student about a failing grade. Moreover, many professions entail bad news transmission as part of the job description. Health care empl oyees must convey diagnoses, military personnel must deliver news of wartim e casualties, and manage rs must occasionally hand out pink slips. Although giving bad news is uncomfortable for the giver, the opening medical examples point to the importance of givi ng bad news well. A number of researchers in the medical field have provided guidelines to he lp those who must give bad news, but what people mean by giving bad news well is less clear. Evaluating the success of a bad news 26

PAGE 27

transmission requires that news-givers have a goal in mind when giving bad news and compare the outcomes of their transmission to that goal. Some researchers focus on the goal of providing hope to the recipient of the news, others focu s on making the transmission easier and less painful for the news-giver, and others focus on increa sing recipients satisfac tion with the bad news transmission. What should be the primar y goal when communicating bad news? The purpose of this review is to investigate how to give bad news well. Central to the notion of giving bad news well is having a clear understanding of the goal of bad news transmission. We critically evaluate six goals su ggested by prior research and propose a broader, more comprehensive goal for giving bad news. We then offer a theoretical framework, the Bad News Response Model, which draws from resear ch in health and social psychology and is designed to maximize positive long-term outcomes for news-recipients. The model specifies four possible responses to bad news and three situational factors that influence the response choice. Finally, we discuss future directions for research. The goals of this review are, by necessity, limited in scope. We do not address specific aspects of bad news transmission, such as tone of voice, setting, eye contact and amount of information. Although many studies have addresse d these issues (e.g., Ptacek & Eberhardt, 1996; Loge, Kaasa, & Hytten, 1997; Holland, 1989; Ptacek & Ptacek, 2001), and these aspects of the communication can affect how people respond to bad news, they fall outside of the scope of this article. Furthermore, we do not address the process by which people respond to bad news. The literature on coping focuses on how people apprai se and respond to bad news (Snyder, 1999 for a review). We focus on the goals people have when giving bad news. Finally, we do not attempt to predict with certainty the best responses to bad news. The Bad News Response Model predicts 27

PAGE 28

how situational factor s might affect responding, but the m odel does not stipulate how people should respond. Giving Bad News Well A review of the medical literat ure reveals a rich yet disorgan ized picture of how to give bad news well. Some consensus exists as to the stages of the bad news transmission process and the important aspects of the situation and the message itself (Fallowfield & Jenkins, 2004; Faulkner, 1998; Ptacek & Eberhardt, 1996), but lit tle consensus exists about the goals these suggestions are designed to achieve and often no goal is mentioned at all. The medical literature suggests six possible goals of a bad news communication: 1) decr easing news-givers discomfort, 2) providing sufficien t information to recipients, 3) promoting recipients satisfaction with the transmission, 4) improving news-recipien ts memory for and understanding of the news, 5) reducing recipients distress in response to the news, and 6) promoting hope. We briefly review the evidence supporting each of these goals and then offer a new, alternative goal that incorporates the positive aspects of the other six goals. Although we treat these goals as distinct for the purpose of this review, it is notewort hy that a news-giver ma y simultaneously pursue multiple goals during a bad news communication. Decreasing the News-givers Discomfort One goal of bad news-givers is to decrease the discomfort they feel about giving bad news. Giving bad news is often extremely unpleasant for the news-giver. Physicians and nurses report discomfort with giving bad news due to lack of training, fear of patients emotional reactions, fear of their own emo tional reactions, fear of being bl amed for the bad news, fear of the patients suffering and dying, personal fear of illness and death, and uncertainty associated with not knowing all the answ ers (Ambuel & Mazzone, 2001; Buckman, 1984). Reducing newsgivers discomfort is an important goal because discomfort with giving bad news can produce 28

PAGE 29

negative consequences for both news-givers a nd recipients. One study found that burnout and poor mental health are common am ong physicians who must frequen tly give bad news, and that physicians who felt insufficiently trained in givi ng bad news experienced the greatest distress (Ramirez et al., 1995). Furthermore, research sh ows that physicians who are more comfortable and confident with giving bad news are perceive d as more trustworthy by patients, and patients who trust their physicians are more likely to co mply with treatment recommendations (Holland, 1989). With the importance of this goal in mind, several training programs for health care professionals aim to decrease discomfort and increase confidence and skills with giving bad news (Baile et al., 1999; Parath ian & Taylor, 1993; Unger, Alpe rin, Amiel, Beharier, & Reis, 2001). In addition to formal training programs, a nu mber of professionals suggest strategies to make the job of giving bad news easier (e.g., Clark & LaBeff, 1982; Eggly et al., 1997; McClenahen & Lofland, 1976; Radziewicz & Bail e, 2001). For example, one paper describes five strategies to smooth the proc ess of giving news of death and suggests that the best method of communication depends on several situational factor s (the type of death, the age at death, the place of death, and the occupation and experience of the news-giver) (Clark & LaBeff, 1982). Helping people to feel comf ortable giving bad news is cl early important. However, the goal of decreasing news-givers discomfort is problematic for several reasons. First, making news-givers as comfortable as possible may mean that they portray the ne ws in a more positive light than is warranted, omitting negative informa tion in an attempt to avoid eliciting negative responses from the receiver. Second, this goal rele gates to secondary importance the needs of the news-recipient. An appropriate goal for bad news transmission should account for the needs of both the news-giver and news recipient. 29

PAGE 30

Providing Sufficient Information A second goal for giving bad news described in the medical literature is to provide newsrecipients with sufficient information about th e news. Researchers who address the ethics of giving medical bad news primarily focus on how much information people should receive about diagnoses and prognoses. In medical settings, ethical treatment requires that patients receive clear, honest information because it allows the patients to accept the situation and make plans for the future (Fallowfield, Jenkins, & Beveridge, 2002; Girgis, Sanson-Fisher, & Schofield, 1999; Goldie, 1982; Ward, 1992). The information provi ded should also be consistent among patients and their family members to avoid distru st and suspicion (Doyle & OConnell, 1996). Furthermore, cultural, family, and personal prefer ences affect the amount of information patients wish to receive, and it is the physicians responsi bility to consider these preferences (Sabbioni, 1997). However, personal comfort with disclosing bad news often determines what and how much information many physicians disclose. Gi ving and receiving bad ne ws are both unpleasant experiences, and physicians and patients may be eager to avoid the experience at all costs. In one study, 40% of physicians admitted to giving patients inaccurate life expectancy estimates, mostly in an optimistic direction (Lamont & Chisakis, 2001). Although some people may not be prepared to hear the full truth about an undesirable diagnosis (Bor et al., 1993; Geer, Morris, & Perringale, 1979; Goldie, 1982; Lubinsky, 1994; Michaels, 1983; Radziewicz & Baile, 2001), evidence suggests that patients with serious conditions often suspect that they will hear bad news (Fallowfield, Jenkins, & Beveri dge, 2002). Knowing how much in formation to disclose is difficult, leading some to recommend that physic ians repeatedly ask patients how much they want to know, thus allowing the patients to determine the level of information conveyed (Freedman, 1993). 30

PAGE 31

Providing sufficient information is clearly a necessary goal in medical interactions, including bad news communication. Patients w ho do not receive sufficient or accurate information are unable to make informed decisions as to how they want to respond to their diagnosis. However, this goal is insufficient fo r guiding bad news-givers. The ethical guideline of providing clear, complete inform ation to patients is a means to an end, not an end in itself; it is only a starting point for guidi ng bad news transmission. Promoting Recipient Satisfaction A third goal for bad news-givers is to give th e news in a way that satisfies recipients. A predominant belief in the medical literature is that patients should be satisfied with the way they receive bad news and that they should have their needs met in the communication experience. One review of the literature concluded that, wh en giving bad news, the patients desires and needs are far more important than the physicia ns (Ptacek & Eberhardt, 1996). Many studies reinforce this idea by asking patients how they want to hear bad news and then using their responses to design bad news communication st rategies (e.g., Ambuel & Mazzone, 2001; Back & Curtis, 2002; Butow et al., 1996; Girgis et al., 1999; Hagerty et al., 2005; Randall & Wearn, 2005; Salander, 2002). Other studies measure patients satisfaction w ith their experience receiving bad news to assess the competence of bad news-givers (e.g., Damian & Tattersall, 1991; Derdiarian, 1989; Dunn et al., 1993; Ellis & Tattersall, 1999; Gillotti, Thompson, & McNeilis, 2002; Hurwitz, Duncan, & Wolfe, 2004; Mast, Kindlimann, & Langewitz, 2005; Ptacek & Ptacek, 2001; Reynolds, Sanson-Fisher, P oole, Harker, & Byrne, 1981). Such studies typically find that the needs patients express are in line with established guidelines for giving bad news (Girgis et al., 1999; Ptacek & Ptacek, 2001; Randall & Wearn, 2005). However, some differences arise when comparing patients needs and the ways physicians typically give bad news. For example, patients rate information about treatment and 31

PAGE 32

future outcomes as more important to them th an diagnostic information (Back & Curtis, 2002; Butow et al., 1996; Salander, 2002). Patients also view the experience of receiving bad news as an ongoing process throughout their treatment, not as a one-time conversation with their physician (Randall & Wearn, 2005). Finally, a number of studies find that small talk and general expressions of support, not in-depth conversation about diagnoses or prognoses, are most helpful to patients receiving bad news (Dean, 2002; Gillotti et al., 2002). Increasing patient satisfaction with bad ne ws communication is a worthy goal for bad news-givers. However, asking patients how they wa nt to hear bad news assumes that patients can objectively and accurately evaluate their own em otions and the reasons for them. People are generally poor at accurate introspection and t hus poor at understanding the reasons behind their emotional and cognitive responses (Nisbett & Wilson, 1977). For example, patients reporting that their physicians are unskilled at presenting bad news may be unaware of the effects that their location, mood, and relationship with the physicia n have on their feelings about the news transmission. In this light, it seems that pati ents opinions, although important and valid, may be a somewhat inaccurate source of suggestions for how to give bad news. Finally, satisfaction with a bad news communication is undoubtedly influenced to some extent by the content of the communication. Other things being equal, the wo rse the news, the less sa tisfied people will be with the communication. Although the medical literature does not sp eak to this point, it seems likely that a focus on promoting recipient sa tisfaction with the co mmunication creates the possibility that news-givers will alter or water down the bad news. Improving Memory and Understanding A fourth goal for giving bad news is to en sure that recipients understand and remember information about the bad news and its implications. People receiving bad news often find it difficult to understand and remember the informa tion they receive. For example, a patient may 32

PAGE 33

hear the word cancer during a diagnostic conve rsation and fail to process any information thereafter. Presenting information in a way that patients can understand is crucial in bad news communication because it can improve patients outcomes, avoid confusion and distress, and increase patients satisfaction with the communi cation (Baile et al., 2000; Ellis & Tattersall, 1999; Fallowfield & Jenkins, 2004; Loge et al., 1997; Quill, 1991). A number of researchers sugge st that when bad news is complicated or difficult to understand, bad news-givers should be careful to clarify, check for understanding, and summarize the information presented (Dias, Ch abner, Lynch, & Penson, 2003). Furthermore, a number of communication skills of the news-giver, such as self-c onfidence, warmth, and honesty, can help the patient to process bad ne ws (Myers, 1983). Finall y, providing recall aids (e.g., informational handouts, audiotapes of the bad news communication) often improves patients memory and understanding, and may reduce patient anxiety (Back & Curtis, 2002; Hogbin, Jenkins, & Parkin, 1992; McHugh et al., 1995; Reynolds et al., 1981). Improving news-recipients memory and understa nding of the information they receive is clearly an important goal for ba d news-givers. People who fail to comprehend the bad news they receive may be unable to cope with the news and may make unwise decisions in response. On the other hand, news-recipients need more than me mory and understanding of the facts to cope and respond effectively. For example, even if a breast cancer patient hears, understands, and remembers the details of her diagnosis, she is unl ikely to know much abou t the implications of the diagnosis or what course of action she should take. As such, improving memory and understanding represents an important intermed iate goal for bad news communication, but is insufficient as an overarching pr inciple for guiding news-givers. 33

PAGE 34

Reducing Recipients Distress A fifth goal for bad news-givers is to minimize recipients distress in response to the news. People receiving unexpected or traumatic news may be emotionally paralyzed by the experience, and poor communication by news-giver s can exacerbate recipients distress (Lerman et al., 1993). Many physicians a nd researchers recogni ze the importance of reducing emotional trauma following a bad news communication (Baile & Aaron, 2005; Boyd, 2001; Brewin, 1991; Fallowfield & Jenkins, 2004; Ptacek & Eberha rdt, 1996; Quill, 1991; Rabow & McPhee, 1999; Shields, 1998). Emotional distress may particular ly involve fear of death when bad news is health related (Penson et al., 2005), but all types of bad news are capable of producing distress. Researchers have noted that news -recipients distress is most severe during and just after a bad news communication, whereas the news-givers distress is most severe just before and during (Ptacek & Eberhardt, 1996). A consequence of this incongruity is that news-givers may be insensitive to recipients distress following the communication of bad news. In response to this problem, the medical literature provides a nu mber of suggestions for increasing sensitivity to patients distress. For example, physicians s hould prepare in advan ce for the communication (Holland, 1989; Michaels, 1983; Shields, 1998), demonstrate empathy, sensitivity, and compassion (Boyd, 2001; Brewin, 1991; Fogarty et al., 1999; Mast et al., 2005; Penson et al., 2005; Rabow & McPhee, 1999), allow patients to express their emotions (Boyd, 2001; Penson et al., 2005; Rabow & McPhee, 1999), take sufficient time in the bad news communication (Boyd, 2001; Penson et al., 2005), and help patients put the situation in perspective (Lalos, 1999). In addition, physicians shouldnt simply reassure the patient and move on; instead, they should acknowledge patients distress, determine the sour ces of distress, and check the patients needs before moving on to reassurance (Maguire, 1998). 34

PAGE 35

Although news-givers are in part responsible for managing reci pients distress, this goal is not an end in itself and thus is an insufficient goal for bad news communication. Much like the goal of improving memory and understanding, the goa l of reducing recipients distress is an intermediate goal that makes it possible for reci pients to move on to the greater goal of coping and responding to the bad news. Even if the brea st cancer patient descri bed earlier understands and remembers her diagnosis and also maintains a manageable level of dist ress after hearing the news, she may nevertheless lack the resources to move forward and respond. Finally, as with promoting recipient satisfaction, focusing on redu cing distress may inadvertently influence the content of the communication. The worse the news the more likely the recipients will be distressed by the communication. Thus, similar to the goal of promoting recipient satisfaction, attending to recipient distress may lead news-givers to alter or water down the bad news. Promoting Hope A sixth goal for giving bad news is to prom ote hope or optimism in recipients, an idea that has received considerable attention in the medical literature (Bor et al., 1993; Bruhn, 1984; Charlton, 1992; Clayton, Butow, Arnold, & Tattersall, 2005; Groop man, 2004; Yates, 1993). Hope can be defined as a combination of desire s for the future, values and goals about future outcomes, and action to bring about hoped for outcomes (Simpson, 2004). The goal of promoting hope is supported by the finding th at hope may be a powerful force in predicting positive health outcomes, such as better adjustment to breast cancer (Taylor, Lichtman, & Wood, 1984), lower incidence of hypertension (Richman et al., 2005), better immune f unctioning (Segerstrom, Taylor, Kemeny, & Fahey, 1998), and faster recove ry from a number of illnesses (Groopman, 2004). A number of factors in crease the likelihood of promoting hope in a bad news communication. For example, fostering a good relationship between patient and physician 35

PAGE 36

(Bruhn, 1984; Salander, 2002), focusing on the pote ntial for successful treatment (Bruhn, 1984; Clayton et al., 2005; Peteet, Abrams, Ross, & St earns, 1991; Sardell & Tr ierweiller, 1993), and discussing the effects of the news on day-to-d ay living (Clayton et al., 2005) promote hope in patients receiving bad news. Promoting hope as a goal for news-givers is somewhat problematic. Although hope may lead to positive outcomes in many cases, it must be balanced with hone sty and realistic goals (Clayton et al., 2005; Groopman, 2004; Links & Kramer, 1994). This balance is particularly important when there is a possibility that hope ma y be shattered at some point down the road, as is often the case during the course of an illness. Shattered hopes can lead to disappointment and distrust of those who initia lly communicated hopefulness (Peteet et al., 1991; Van Dijk, Zeelenberg, & van der Pligt, 1999). On the other hand, hope can always be directed toward the possibility of improving outcomes down the road or having a productive life in spite of the bad news (Links & Kramer, 1994; Peteet et al., 1991 ; Yates, 1993), making hope a worthwhile goal for bad news-givers. However, providing hope is not the same as providing news-recipients with the information they need to cope and respond to negative life events. The breast cancer patient gains strength and other positive outcomes as a result of her physician promoting hopefulness, but she needs more than hope to know which c ourse of treatment to undergo. Hope may be a necessary component of coping with bad news, but news-givers must also help recipients engage in responses that will lead to the best long-term outcomes. Thus, promoting hope, along with improving memory and understanding and reducing distress, is an intermediate step in the greater goal of promoting eff ective responses to bad news. 36

PAGE 37

Guiding Recipients Toward Desired Responses The six goals for bad news-givers just described (reducing news -givers discomfort, providing sufficient information, promoting reci pients satisfaction, improving memory and understanding, reducing distress, and promoting hope) suggest different interpretations of what it means to give bad news well. However, these goals are means to an end, not ends in themselves. The six goals described are intermediate goals that may ultimately lead to positive outcomes for the recipient of the news but do not specify how these positive outcomes can be achieved. Furthermore, these goals were developed for us e in medical settings and may be difficult to apply to other types of bad news. We suggest an alternative, broader goal for ne ws-givers that incorporates aspects of each of the other goals. We propose that giving bad ne ws well is defined as guiding news-recipients towards desired responses responses that news-g ivers believe will result in the best long-term outcomes for recipients. Although we later provide suggestions as to which responses may be most effective, a desired response refers to the response deemed best by the news-giver. To illustrate, imagine a physician giving news of cancer. The physician must convey the diagnosis honestly and clearly, but ultimately the physician must encourag e the patient to seek the most effective course of treatment or perhaps choose no treatm ent, depending on the situation. With this goa l in mind, news-givers can provide sufficient information and feel confident in their ability to give bad news well. Furthermore, numerous studies suggest that a focus on options for the future increases satisf action with the communication, reduces distress, and promotes hope (Back & Curtis, 2002; Clayton et al., 2005; Peteet et al., 1991; Salander, 2002; Schofield et al., 2003). Finally, guiding news-re cipients towards the most effective course of action (or inaction) maximizes their chances of experiencing positive long-term outcomes and quality of life, although the definition of the best outcomes varies greatly across situations. In 37

PAGE 38

general, successful bad news tr ansmission should prompt the recipient to respond in a way that maximizes quality of life and minimizes nega tive life outcomes. These outcomes include financial stability, physical, mental, and emotional health, and general well-being. A handful of studies have examined positive long-term outcomes associated with communicating bad news and reveal that a variet y of factors can direct ly influence outcomes such as psychological adjustment to an illness and psychological and emotional health. Findings suggest that strategies such as expressing empathy, allowing sufficient time for the bad news communication, and engaging the patient in treatm ent decisions, among others, predict better adjustment to breast cancer (Butow et al., 1996 ; Roberts, Cox, Reintge n, Baile, & Gibertini, 1994). Another study of breast cancer patients fo und that perceptions of caring and emotional supportiveness during the bad ne ws communication predicted fewer cancer-related PTSD symptoms, less depression, and less general dist ress (Mager & Andrykowski, 2002). In addition, physicians personal manner, communication skills, t echnical skills, and ove rall care predicted emotional health in breast cancer pa tients (Silliman et al., 1998). Yet the studies just described do not indi cate how various aspects of the bad news communication lead to positive or negative hea lth outcomes. For example, how does emotional supportiveness by physicians lead to better emotional outcomes in patients? It may be the case that supportiveness leads to better treatment d ecisions, or any number of positive behaviors, which then lead to better long-term outcomes. Th e researchers typically offer no explanation of how factors such as perceptions of caring, em otional supportiveness, the physicians personal manner, communication skills or technical skills produce bene ficial outcomes. Moreover, because these studies are largely correlational and rely almost entirely on patients retrospective reports about how they received their diagnoses, the specific mechanisms are difficult to pin 38

PAGE 39

down. Thus, we propose that guid ing recipients towards desire d responses represents the mechanism by which news-givers can promote positive long-term outcomes. The medical literature supports the goal of guiding patients toward the best course of action during bad news communications. A number of physicians note that patients want to focus on the future, towards treatment and long-term outco mes, rather than just on the diagnosis (Back & Curtis, 2002; Baile and Aar on, 2005; Baile et al., 2000; Bor et al., 1993). Other physicians describe methods for giving bad news with th e stated purpose of impr oving coping and decisionmaking (Boyd, 2001; Clayton et al., 2005; Epstein, Alper, & Quill, 2004; Fogarty et al., 1999; Lalos, 1999). Finally, several physic ians discuss various possible re sponses to bad news and the outcomes of engaging in different responses (De Haes & Koedoot, 2003; Greer et al., 1979). The goal of guiding recipients toward th e most effective responses prompts two questions. First, what are the different ways people can respond to ba d news? Second, which responses should bad news-giver s suggest? We developed the Bad News Response Model to answer these two questions. We propose that all responses to bad news fall into one of four categories: watchful waiting, act ive change, acceptance, and non -responding. We further suggest that three factors of the outcomes of bad news (controllability, likelihood, and severity) indicate which response is likely to be effective (Figure 1-1). The Bad News Response Model The Bad News Response Model suggests that giving bad news well involves guiding news-recipients towards a desired response. Thus the model is aimed both at the person who must give bad news and at the recipient of the news. Ultimately, the goal of the Bad News Response Model is to elicit a desired response from the recipient of bad news, but bad newsgivers must evaluate the characteristics of th e possible bad outcome that determine what the desired response should be. Bad news-givers can th en tailor their communica tion of bad news to 39

PAGE 40

encourage the desired response from the recipien t. In addition, recipients of bad news can individually evaluate their situation and determine the most effective response to the news. It is important to note that the Bad News Response M odel does not attempt to precisely predict the response that will lead to the best quality of life. The model provides guidance for bad newsgivers as to which responses may be best under different circumstances, but the model addresses a wide spectrum of bad news and must make ge neralizations based on situational factors. Bad news-givers and recipients should choose the response that is most likely to result in the best outcomes, given their assessment of the situational factors. In addition, the Bad News Response Model focu ses on what lies in the future as a result of the bad news and not on the event that has pa ssed. The model addresses responses to bad news and the outcomes of those responses on the future Thus, the situational factors in the model do not pertain to the event that is being disclosed, but rather to the possible results of that event. These results include both direct outcomes of the bad news and i ndirect effects of the news on other parts of life. For example, a professor w ho must tell a student about a failing exam grade (the past event) should consider the impact of that exam on the students final grade in the course (a direct future outcome) and on the students overall academic status (an indirect future outcome) when determining the best strategy for bad news transmission. How Can People Respond to Bad News? The Bad News Response Model suggests that people can respond to ba d news in one of four ways: a) watchful waiting, b) active cha nge, c) acceptance, and d) non-responding. These response categories broadly appl y to many kinds of bad news, although the specific nature of each response may differ across domains. For example, a patient who responds to a diagnosis of cancer with active change will engage in differe nt specific behaviors than will a student who 40

PAGE 41

responds to a failed exam with active change. Howe ver, we suggest that these two responses will be similar in fundamental ways. As evident in Table 2-1, we anticipate that the four response categories will each elicit a unique pattern of characteristics in terms of anxiety, affect and activ ity level. Anxiety level refers to feelings of worry, concern, or fear. Affect refers to general positive or negative moods and emotions, such as happiness or sadness. Activity level refers to the extent to which energy is directed towards changing the outcomes of the ba d news. Of note, the characteristics described may be present with all four responses to vary ing degrees, but we suggest that they are more likely to occur with thei r respective response. The characteristics in Table 2-1 may be cau ses, consequences, or concomitants of each response choice. For example, people who experience high levels of anxiety in response to bad news may be more likely to see the need to take action, in which case anxi ety serves as a cause. However, people may also deliberately choose to respond to bad news in a certain way, which can then lead to a variety of consequential thoug hts and feelings. Finally, certain kinds of bad news may prompt both a particular response and particular thoughts and feelings independently. Within the present model, we simply discuss th e dimensions in Table 2-1 as characterizing a given response category. Watchful waiting The first category represents a relativel y passive form of responding. The medical literature has employed the term wat chful waiting as a specific c ontrast to aggressive treatment options (e.g., de Haes & Koedoot, 2003). Here, watc hful waiting indicates a more general wait and see mentality regarding the bad news. The term watchful emphasizes that people engaged in this response are aware that they are facing a possible threat and are vigilant to changes in their situation. However, they maintain the st atus quo rather than take action. To illustrate, 41

PAGE 42

consider a man diagnosed with pr ostate cancer. The man is in his late 80s, a widower, and has few financial responsibilities. Although this man registers and accepts his diagnosis of cancer, he may choose not to get a second opinion or underg o treatment but instead go on with his life largely as if nothing had change d. He may make annual appointme nts to reassess his response, but otherwise his life remains as it was prior to his diagnosis. Watchful waiting bears similarity to the ge neral conceptualizati on of emotion-focused coping (Folkman & Lazarus, 1980), although the sp ecific characterization of emotion-focused coping differs widely between studies (Carver, Scheier, & Weintraub, 1989). The similarity resides in the fact that both watchful waiting and emotion-focused coping focus on distraction and emotional regulation. Emotion-focused copi ng entails directing energy toward managing anxiety and other negative emoti ons arising from a stressful situ ation rather than engaging in active intervention. People in the watchful waiting category may engage in act ivities that distract them from the bad news. Behaviors that are design ed to take ones mind off of a threat may be beneficial if no actions will make a difference, or if action would be too costly or if dwelling on the threat is counterpr oductive (Lazarus, 1985). However, watchful waiting differs from em otion-focused coping in a fundamental way. Emotion-focused coping is not mutually exclus ive with other more ac tive forms of responding, and in fact people may engage in emotion-focuse d coping in all four re sponse categories of the Bad News Response Model. Emotion-focused coping complements all forms of responding by reducing the intensity of stressful emotions a nd allowing people to ga in perspective on their situation (Folkman & Lazarus, 1980). In contrast, watchful waiting involves a specific set of behaviors and emotions that represen t one way of responding to bad news. 42

PAGE 43

It seems likely that watchful waiting involves low anxiety, high general positive affect, and low activity level. Each of these characteris tics results from distraction from the bad news and attention towards other, presumably positive aspects of life. Excessive focus on the bad news would increase anxious thoughts and feelings, induce sadness and distress, and lead to high activity levels in an effort to mobilize acti on towards change. People engaged in watchful waiting avoid this process by distracti ng themselves from the bad news. Active change Active change represents the most vigorous, engaged form of responding. Unlike the distraction or irrelevant activity characterizing watchful waiting, active change involves specific responses directed toward addressing the bad news. Active change aligns most clearly with traditional views of productive coping strategies, such as problem-focused coping, that directly address the negative situation. Pr oblem-focused coping in part i nvolves taking action to solve a problem or change a negative situation (Carve r et al., 1989; Folkman & Lazarus, 1980; Folkman & Lazarus, 1985). Active change includes three types of be havior: information-seeking, prevention, and treatment. Information-seeki ng serves two purposes. First, information-seeking provides recipients of bad news with the information th ey need to make decisions about how to respond. Second, information-seeking serves to connect recipi ents with others who have dealt with similar experiences and provides a networ k of support. Of note, other re searchers have discussed these roles of information-seeking as part of pr oblem-focused or active coping (Aldwin & Revenson, 1987; Lazarus, 1981; Lazarus & Launier, 1978). The terms prevention and treatment have medical connotations, but in this context they broadly refer to behaviors directed towa rd preventing the situation from deteriorating (maintenance) and treating an undesirable situ ation that has emerged (improvement). To 43

PAGE 44

illustrate, consider a different man diagnosed with prostate cancer. This man is in his late 40s, has a wife and several children, and is the primar y breadwinner for the family. Unlike the man in his 80s who chooses watchful waiting, the s econd man may be very willing to undergo chemotherapy and radiation in hopes that it will eradicate the cancer and allow him to live a full and long life with his family. He should activ ely investigate his condi tion, perhaps seeking a second opinion or researching pros tate cancer online or at the lib rary, and undergo preventative and/or aggressive measures to prolong his life. Active change also involves high anxiety and high activity levels. The high levels of anxiet y result from acknowledgement that a negative event is likely to occur and/or that the consequences are severe The high activity levels results from mobilization of energy towards active responses. Acceptance Acceptance is the third and most complex form of responding. This response is similar to previous conceptualizations of acceptance in th e literatures on aging, disability, and death. Previous theories discuss acceptance as a last st age in coping with loss or impending death that comes after a process of denial (Gamliel, 2000; Kbler-Ross, 1969) Many theorists assert that acceptance is a positive coping strategy in uncon trollable circumstances. People who come to accept their circumstances are able to seek meani ng in their loss, reduce their dread over what lies ahead, and seek social suppor t to cope (Gamliel, 2000). On the other hand, other researchers have found little support for the assertion that acceptance is an adaptive coping strategy, and some studies even suggest that realistic acceptance might be predictiv e of negative outcomes (Greer et al., 1979; Reed, Kemeny, Taylor, Wang, & Visscher, 1994; Wortman & Silver, 1989). We view acceptance as action towards acceptanc e rather than passive resignation. People who respond to bad news with acceptance do not necessarily collapse in a heap, although this response may be unavoidable at first. Instead, they eventually direct their energy towards moving 44

PAGE 45

forward and addressing any consequences of th e bad news. Acceptance involves looking beyond the negative outcomes to the possibility for hope th at lies in the future Even in the case of imminent death, people can find hope in living life to the fullest during their remaining time and dying with dignity (Dean, 2002). A lthough this response is similar in many ways to previous conceptualizations of acceptance, it avoids the sense of passivity and hopelessness that may lead to negative outcomes. In addition, acceptance is not a final, static state of resignation; instead, it involves an ongoing positive process of ma king the best of a bad situation. Acceptance combines aspects of watchful wa iting and active change to most effectively address situations in which a l ack of engagement is inappropriate yet the person cannot change the outcome. People can direct their energy toward s changing their lives ra ther than changing the negative event. This response bear s similarity to the concept of secondary control, in which people change themselves to fit a situation rather than changing the situation to fit the self (i.e., primary control) (Rothbaum, Weisz, & Snyder, 1 982). Secondary control represents an important form of control over ones emotional responses, but it does not involve engaging effort towards changing the situation. Acceptance involves two types of behavior : information-sharing and accommodation. Information-sharing involves telling others a bout the negative event, although the extent of sharing with others may vary depending on the news. For example, certain types of bad news, such as testing HIV-positive, may stigmatize th e individual, and people may want to limit their information-sharing to close friends and family. Furthermore, the effectiveness of informationsharing depends in part on the receptiveness of the listener (Harber & Pennebaker, 1992; Kelly & McKillop, 1996). 45

PAGE 46

Information-sharing serves three purposes. Firs t, information-sharing helps people accept the negative event by making the event part of th eir social reality. People who keep negative events, such as a disease or a job loss, a secret fr om friends and family may be in denial that the event has occurred. Information-sharing is both a step towards acceptance and a sign that such acceptance has begun. Second, information-sharing elicits social support from friends and family. Researchers have distinguished between seeking social support for emotional reasons vs. seeking social support for practical reasons (advice, assistance, etc.) (Carver et al., 1989). acceptance focuses more on the emotional side of social support, rather than the more active, change-focused practical side. Third, informati on-sharing seems to serve an important function in an end unto itself (Pennebaker, 1988; Pennebaker, Zech, & Rim, 2001). Research finds that people who talk (or write) more about a trauma tic event ruminate less (Pennebaker & OHeeron, 1984), experience less anxiety (Pennebaker, Colder & Sharp, 1990), have fewer negative health outcomes (Pennebaker & OHeeron, 1984), and have better quality of lif e (Spera, Buhrfeind, & Pennebaker, 1994), even when the ex pressions are private. Accommodation involves making changes, not to affect the news-specific outcome, but rather to incorporate the negative event into ones life. When a negative outcome is uncontrollable, accommodation focuses peoples en ergy on productive activity rather than futile efforts to change the outcome. For example, in most cases a woman who receives a rejection letter from her first choice gr aduate school should consider alte rnative schools or career plans rather than continuing to pursu e admission at the school that re jected her. Accommodation often involves behavioral changes su ch as cutting back on strenuou s activities in the case of a debilitating disease or putting away a lost loved ones personal item s in the case of a death in the family. It often also involves cognitive changes that entail looking for reasons why the tragedy 46

PAGE 47

occurred (sense-making) and focusing on positiv e changes resulting from the tragedy (benefitfinding) (Davis, Nolen-Hoeksema, & Larson, 1 998; Rabow & McPhee, 1999). Of note, other researchers have used the term accommodation differently, referring to a passive means of coping with old age that involves weakened aspirations and lowere d standards of living (Brandtstadter, Dirk, & Werner, 1993). Here, we use accommodation to refer to an active process of reordering priorities a nd adjusting to the new situation. In addition, the response of acceptance involves general negative affect (including sadness, regret, guilt, etc.) and moderate activity level. People are likely to experience particularly negative feelings when a severe negative event occurs and they are helpless to change the outcomes. The specific types of negative affect people experience depend on the details of the bad news. For example, people may feel guilt and regret when they feel that they could have changed the outcome, as in the case of failing a cl ass, but people are more likely to experience sadness and grief when they believe they could no t have changed the outcome, as in the case of an unavoidable death. Regarding activity level, a cceptance does not involve the same level of energy mobilization as active change, but some effort is required to adapt to the negative event. People must direct their energy towards understa nding and accepting the s ituation created by the bad news and dealing with the consequences, rather than taking active steps towards making significant life changes in an effort to change the outcomes of the bad news. Non-responding The fourth category of responding captures a number of responses. Lubinsky (1994) distinguishes between four forms of non-responding: denial, disbel ief, deferral, and dismissal. Although the four may appear simila r, their sources differ. Denial is form of repression brought on as a defense mechanism. It involves vehement disagreement with any disliked information, even when evidence makes it clear that the info rmation is correct, and is a relatively rare 47

PAGE 48

response to bad news. Disbelief is marked by confusion rather than rejection of bad news and may result from a desire to maintain hope for longer than is warranted. Deferral is marked by avoidance of information about bad news as a re sult of inadequate resources to cope with the situation. People responding with deferral may accep t the basis for bad news (e.g., results of a medical test) but reject or ignore the implications of those findings (i.e., the necessity of lifestyle changes or treatment). Finally, dismissal is marked by anger at the bad news-giver and denial of the news-givers competence or legitimacy. These four reactions, though different in significant ways, all fall into the res ponse category of non-responding. Non-responding is distinct from watchful waiting. non-responding is not an attempt to reduce anxiety about bad news wh ile acknowledging it, but rather an attempt to pretend nothing has happened or wish away the bad news. Furt hermore, non-responding ma y be most likely to occur in situations when acceptance is called for. Both watchful waiting and active change are somewhat attractive responses: one allows people to monitor the news and defer action until it is appropriate, and the other involve s taking action to change thi ngs for the better (de Haes & Koedoot, 2003). Acceptance, in contrast, requires people to face the news head-on and does not offer the hope that things will turn out well. Although acceptance is necessary when a very bad outcome is unavoidable, people may prefer to embrace non-responding instead. Non-responding can feel good for a short time because it allows people to pretend that nothing has changed for the worse, but people eventually must face negati ve outcomes, such as the death of a loved one or a terminal illness, and cope with the consequences. On the other hand, non-responding may be an acceptable response in the short-term. A number of researchers and physic ians note that denial is a ne cessary response for some people under certain circumstances, and a number of theo rists have argued that news-givers should not 48

PAGE 49

force recipients to face bad news before they ar e ready (Bor et al., 1993; Faulkner, 1998; Greer et al., 1979; Radziewicz & Baile, 2001). As indicated in Figure 1-1, the Bad News Response Model indicates that non-responding is a legitimate but generally undesired response. News-givers may recognize that recipients are likely to engage in non-respondin g at first, but the Bad News Response Model suggests that the go al of the news-giver is to gui de people towards the response that will lead to the best long-term outc omes. Although non-responding may be functional at first, people must eventually face bad news and choose a different response. How Can People Respond? Summary and Conclusions The Bad News Response Model suggests that people can respond to ba d news in one of four ways: 1) watchful waiting, 2) active change, 3) acceptance, and 4) non-responding. Watchful waiting represents a relatively inactive res ponse characterized by di straction activities and managing anxiety. Active change is a highl y active response primarily characterized by direct attempts to change the situation. Accep tance involves activity directed toward changing ones life to incorporate bad ne ws rather than attempting to ch ange the outcomes of the news. Non-responding involves unproductive (at least in the long-term) avoidance or denial of bad news. Although thus far we have discussed the four response categories as though they were mutually exclusive and as though selecting one response means reje cting other responses, people may display (or appear to display) multiple responses. Multiple responding can manifest in several ways. First, people may engage in multiple responses simultaneously. By so doing, people hedge their bets by putting some effort towards one response (e.g., trusting that things will go well, as in watchful waiting) while also recognizing and prep aring for alternative possibilities (e.g., by taking measures to encourage a positive outcome as in active change). For example, someone who learns of upcoming layoffs can engage in watchful waiting by delaying 49

PAGE 50

the search for a new job while also engaging in active change by delaying large purchases. This form of multiple responding recognizes that th e future is uncertain and that what a person expects to occur may not occur. As Mohammed, the Muslim spiritual leader said, "trust in God, but tie your camel firs t" (Cleary, 2001). People who seem to be engaging in multip le, simultaneous responses may also be responding to multiple levels of abstraction of the bad news. A single news event may include more than one form of bad news. A young man who learns that he failed a major project in a class relevant to a desired career has essentially received two pieces of news. First, he must deal with the possibility that he will fail the course. Second, he must deal with the implications of his failure on his qualifications to en ter his desired career. He may respond with active change in regards to his course grade while simultaneou sly responding with acceptance in regards to his career path. Alternatively, he ma y respond with acceptance in regards to his course grade but take active measures to ensure that he pe rforms well on other car eer-relevant criteria. Finally, people may respond in different ways to one situation acro ss time. For example, imagine that a physician finds a lump in a patien ts breast. The physician may initially encourage watchful waiting, suggesting that th e patient proceed with life as usual until the biopsy results come in. If the biopsy reveals malignancy, the physician might then recommend active change. Finally, if subsequent tests reve al that the cancer is resistant to treatment, the physician may suggest acceptance. Thus, one broad situation may involve multiple news events, therefore allowing for the possibility of multiple respons es. The Bad News Response Model can account for longitudinal events if news-givers and recipien ts reevaluate the situation at each point when new information is available (Figure 1-1). 50

PAGE 51

Which Responses Should News-Givers Suggest? The Bad News Model suggests that bad news -givers should guide recipients towards desired responses; the model does no t attempt to predict with certain ty the best responses to bad news. The model ultimately relies on news-givers to determine the response that will lead to the best outcomes for recipients and then guide the recipients towards that response. However, bad news varies on a number of predictable dime nsions, and research suggests that certain dimensions may lead one response to be more effective than others, depending on the situation. Specifically, examination of the vast literature on risk perception, heal th behavior, and coping reveals three factors that repeatedly emerge as playing a particularly im portant role in peoples responses to the possibility of bad news and othe r stressful situations: the controlla bility of negative outcomes, the likelihood of negative outcomes, and the seve rity of negative outcomes. Table 1-1 presents a summary of the respons es that may be most effective for each combination of high and low controllability, likelihood, and severity. These suggestions represent the responses that seem most likely to be effective under different circumstances, in light of existing research on both responses to ba d news and situational factors of the news. In general, we suggest that people should engage in active change when two or three of the situational factors are high (high control, hi gh likelihood, and/or high severity) and watchful waiting when two or three of the situational fact ors are low (low control, low likelihood, and/or low severity). The only exception occurs when likelihood and severity are high but control is low. Under these circumstances, when severe ne gative outcomes are highly likely and little or nothing can be done to change the outcomes, we suggest that acceptance is the best response. It is noteworthy that perceptions of controll ability, likelihood, and severity are somewhat subjective. Numerous studies de monstrate that people often func tion under an illusion of control when, in fact, chance determines their fate (Crocker, 1982; Langer, 1975; Langer & Roth, 1975). 51

PAGE 52

In addition, people misperceive the likelihood of events due to misunderstandings of objective probabilities (Kahneman & Tversky, 1982; Tver sky & Kahneman, 1974), undue focus on salient examples (MacLeod & Campbell, 1992; Slovic, Fisc hoff, & Lichtenstein, 1982), and a desire to avoid disappointment or regret (Carroll, Sweeny, & Shepperd, 2006; Sweeny & Shepperd, 2006). Finally, people often base their perceptions of severity on misleading information, such as prevalence, personal relevance, or illness stereotypes (Jemmott, Ditto, & Croyle, 1986; Croyle & Williams, 1991). One prominent model of coping suggests that people engage in an appraisal process to determine whether a stressful s ituation demands coping resources (Folkman & Lazarus, 1980; Lazarus, 1966; Lazarus & Folkman, 1984) and this appraisal process is subject to the many biases that color judgments. Thus, ne ws-recipients undoubtedly choose responses that reflect misperceptions of bad news. Although peoples natural responses may be bi ased, the most effective response to bad news depends more on the actual controllabilit y, likelihood, and severity of potential negative outcomes than on subjective per ceptions of these factors. For example, a patient who misperceives the severity of his or her condi tion due to lack of know ledge or inaccurate understanding will not benefit from, and may even be hurt by, pursuing treatment based on this misperception. The purpose of this section is to di scuss situational factor s that may predict the effectiveness of responses to bad news. As su ch, objective levels of controllability, likelihood, and severity are more important for our purposes than subjective appraisals of these factors by news-recipients. Controllability The first factor that may influence effective responding to bad news is the controllability of the negative outcomes that may result from ba d news. The ability to control the outcomes of bad news varies greatly across di fferent situations. For example, a student who discovers s/he is 52

PAGE 53

failing a course several weeks befo re the semesters end may be able to improve his or her grade by completing extra credit assignments, getting he lp from the professo r, or studying long and hard for the final exam. However, as the semester draws to a close, control over the course grade diminishes, and once final course grades are tu rned in, there may be no remaining avenues to affect the outcome of the course. Controllability plays a significant role in predicting peoples responses to threat. The Health Belief Model (Becker, 1974; Janz & Becker, 1984; Kirscht, 1988) and Protection Motivation Theory (Floyd, Pr entice-Dunn, & Rogers, 2000; Ma ddux & Rogers, 1983; Rogers, 1983) include measures of controllability (respons e efficacy and/or self-effi cacy) as factors that predict whether people engage in preventative he alth behaviors, and the proactive coping model (Aspinwall & Taylor, 1997) indicates that perceive d control plays a role in peoples attempts to prevent negative events. Research on coping shows th at the controllability of a stressful situation affects the strategies people choose to adopt wh en coping with stressful situations (Folkman & Lazarus, 1980). When people perceive event contro llability to be high, they tend to adopt active coping strategies; when people perceive event co ntrollability to be low, they tend to adopt strategies directed towards managing their emotions (Aldwin, 1991; Carver, Scheier, & Weintraub, 1989). Other studies find that the effectiveness of va rious coping strategies depends in large part on the controllabili ty of the stressful situation, wi th active strategies proving most beneficial when the situation is controllable (Aldwin & Park, 2004; Par k, 2001; Park, Armeli, & Tennen, 2004). Likelihood The second factor in determining the appropria te response to bad ne ws is the likelihood of possible negative outcomes. Bad news does not always indicate a guaranteed negative outcome. For example, a boss may have to info rm employees that the company must downsize 53

PAGE 54

without knowing who will lose their jobs. Physicians frequently give bad news that indicates the possibility of illness or injury based on initial evidence without the ability to diagnose a problem with complete certainty. For the purposes of the model, likelihood refers to how likely negative outcomes are to occur if the news recipient does not act to preven t them. For example, the likelihood that a suspicious lump indicates cancer should be eval uated irrespective of treatment options or the patients intentions to seek treatment. As suc h, likelihood is distinct fr om controllability. People may reduce the likelihood of negative outcomes by their response to bad news, but the initial evaluation of likelihood is sepa rate from controllability. Likelihood influences responding in two ways. First, and most intuitive, people consider the likelihood of a negative outcome in weighing th e costs and benefits of an effortful and costly response. Several models include likelihood (or perceived vulnerabil ity) as a predictor of health behavior (Becker, 1974; Rogers, 1983) and preventative behavi or in general (Theory of Reasoned Action, Ajzen & Fishbein, 1980; and Subj ective Expected Utility Theory, Edwards, 1954). Second, and less intuitive, th e initial percepti on of likelihood of a negative outcome influences later affectiv e reactions should the worst actually occur. Expe ctations about future outcomes play a role in how bad a bad outcome feels. Negative outcomes are unpleasant in their own right, but they are particularly unpleasant when they are unexpected (van Dijk & van der Pligt, 1997; Shepperd & McNu lty, 2002). People who respond as if a negative outcome is unlikely to occur may have a particularly unpleas ant experience if the outcome does occur, more so than if they had expected the worst. This finding suggests that people may benefit not only in 54

PAGE 55

terms of literal preparation, but al so in terms of affective prepar ation by engaging in more active responses if the negative out come is likely to occur. Severity The third factor in determining the appropriate response to bad news is the severity of the possible negative event. Bad news varies in term s of how important or co nsequential the possible negative outcome is. Clearly, a woman who learns that she is at risk for heart burn is hearing very different news than a woman who learns that she is at risk for a heart attack, and both the news-giver and the recipient of the news should proceed differently in these two situations. Of course, even news that has re latively non-severe consequen ces can be bad. The woman who learns she has a high risk for heart burn may have to make significant diet ary and other lifestyle changes. However, her response will differ in many ways from the woman learning of her risk for heart attack, and the people giving the news to these women should also proceed differently. The consequences or severity of bad news may differ based on a characteristic of the outcome (e.g., financial impact, life expectan cy, effect on emotional well-being) or characteristics of the individual. The earlier examples of the tw o men diagnosed with prostate cancer illustrate how characteristics of the individual such as age, family circumstances, financial stability and responsibilities can influe nce the consequences of bad news. People naturally account for the severity of potential negative outcomes when they anticipate and respond to bad news. The severity of potential hea lth outcomes predict whether people will engage in preventative health be havior (Becker, 1974; Rogers, 1983), and research on coping finds that people choos e active coping strategies when they judge the event to be highly stressful or important (Anderson, 1977; Parkes, 1986; Terry, 1991). In addition, research on bracing for bad news finds that people onl y embrace a negative outlook for outcomes or consequences that are importa nt (Shepperd, Findley-Klein, Kw avnick, Walker, & Perez, 2000). 55

PAGE 56

People brace less for outcomes that are uni mportant because such outcomes are less consequential for them. For example, participan ts in one study who anticipated soon learning their test results for a medical condition shifted from optimism in their risk estimates only when the consequences of testing positive were severe If the consequences were not severe, their predictions remained unchanged (Taylor & Shep perd, 1998). If possible negative outcomes are inconsequential or non-severe, news-recipients gain more from choosing relatively passive responses (watchful waiting) than from engaging in physically or emotionally active responses (active change or acceptance). Communicating Desired Responses Although the goal of the Bad News Response Mode l is not to elucidate specific details of the communication of bad news, the model suggests that the bad news-giver direct the recipient towards desired responses and offers insights in to which responses may be most effective in different situations. The direction on behalf of the bad news-giver can encourage people to respond in the most effective way even in the face of problems with comprehension, arousal, education, etc. When preparing to give bad news, the communi cator can evaluate th e news situation in terms of the likelihood, severit y, and controllability of negative outcomes and direct the communication towards encouraging the recipient to e ngage in the response that is most likely to be effective. The details of su ch direction will differ greatly de pending on the specific topic and nature of the bad news, but these broad generaliz ations should be effec tive across a variety of domains and situations. It is important to note that the Bad News Response Model does not recommend that bad news-givers manipulate the r ecipient into responding in a particular way using whatever means necessary. Rather, news-giv ers should present all possible responses and 56

PAGE 57

the costs and benefits of each, and then give their opinion regarding th e best possible response (Epstein, Alper, & Quill, 2004). At first glance, the suggestion that news-giv ers should evaluate multiple aspects of the recipients situation to give the bad news in the best way may seem impractical. In many cases, news-givers may know little about the recipient or the circumstances surrounding the bad news they must disclose. However, the models suggestions represent an improvement over leaving bad news-givers to their own devices. News-giver s who attempt to evaluate the bad news and guide recipients towards responses that are most likely to be effective, as suggested by the Bad News Response Model, will likely do a greater serv ice for the recipients than would a news-giver with little or no guidance. Wit hout guidance, news-givers often fa ll victim to personal concerns, such as not wanting to upset the recipient or be blamed for the news, that often trump concern for the best interest of the recipi ent (Buckman, 1984). Furthermore, re cipients of bad news can use the model to evaluate their news and choose the best response when the news-giver is unable to guide them appropriately. Summary, Critique, and Future Directions The medical literature suggests a number of goals to help people give bad news well: new-givers should reduce their own discomfo rt, provide sufficient information, promote recipient satisfaction, improve memory and understanding, reduce recipients distress, and promote hope. However, none of these goals provides sufficient information or a broad enough objective to adequately guide people in giving bad news. We propose that giving bad news well should instead be defined as guiding news-recipients towards desi red responses responses that news-givers believe will result in the best l ong-term outcomes for recipients. The Bad News Response Model suggests that news-givers can look to situational f actors (controllability, 57

PAGE 58

likelihood, and severity) to determine which of three responses (watchful waiting, active change, and acceptance) is most likely to be effective. Strengths of the model The Bad News Response Model has a number of strengths that improve previous attempts in the medical literature to study the pr ocesses of giving bad news. First, the model is applicable to a broad set of situations and domains, including academic performance, professional news, interpersonal news, medical diagnoses, and news of death, among others. Second, the model addresses the ro les of both the bad news-giver and the recipient of the news by making suggestions for transmission based on the desired respons e. Third, the model systematically addresses different types of ba d news in terms of the likelihood, severity, and controllability of possible nega tive outcomes of the news. Although the model draws on the strengths of previous researc h, it represents the first comp rehensive model of giving and responding to bad news. The Bad News Response Model can serve seve ral important purposes. First, the model can assist bad news-givers who otherwise must rely on their own limited experience or personal motivations when giving bad news. The Bad News Response Model provides a goal for bad news transmission that can reduce the impact of the news-givers concerns on their news-giving strategies by guiding them toward s recipient-focused strategies. Second, bad news-givers can use the model to evaluate their transmission of news after the fact. If news-givers observe recipients maki ng an undesired response, they can examine their transmission strategy in light of the model. The news-giver may have incorrectly assessed one or more of the situational factors, or the suggestion of the best response may have been ineffective. For example, physicians may be unaware of their patients financial circumstances, and this lack of information could result in misjudgment of th e severity and/or controllability of patients 58

PAGE 59

medical conditions. Physicians might assume that expensive treatments are feasible when in fact the patient does not have insurance or the mean s to pay for the treatments, making the prognosis relatively uncontrollable. Even when physicians pe rfectly assess the situat ional factors, patients often mishear or forget information conveyed in a diagnostic communica tion (Croyle, Loftus, Klinger, & Smith, 1993). The best efforts of news-givers to prom pt desired responding can be lost if the recipient t unes out the transmission. Bad news-givers who notice seemingly ineffective responding by recipients can seek additional informa tion to better judge the situational factors or reevaluate the bad news transmission for signs of lack of attention or misunderstanding on the part of the recipient. Third, bad news recipients can use the model to evaluate their responses to bad news, apart from the giver. After receiving bad news, recipients can use the model to determine the most appropriate response by evaluating the li kelihood, severity, and c ontrollability of the possible outcomes. For example, a woman who lear ns of upcoming layoffs at work can consider the likelihood that she will lose her job, how bad the consequences of a job loss would be, and if she has control over whether she is laid off. Having evaluated the situation, she may have a better sense of the most effective response. This process ma y help people to override responses based solely on anxiety or fear. In addition, recipients who find th at their response to some news is ineffective can reexamine the situational factors involved and possibly adjust their responses accordingly. If the woman facing a possible job loss responds with active change and then finds that she is making no progress towards keeping her job, she may deci de to shift towards acceptance by checking the want ads and telling her family about the layoffs. Limitations of the model Although the Bad News Response Model is based on research from psychology, medicine and health, the model is largely sp eculative and remains unt ested. In addition, the 59

PAGE 60

model makes broad suggestions to allow the greate st breadth of application. This focus on the functionality of the model leads to an emphasis on generality over detail. As a result, the Bad News Response model may be imperfect in certain specific situations, while making suggestions that lead to the best outcomes overall. Peopl e often make miracle recoveries from medical conditions that were, by all accounts, beyond hope. Although the model would recommend acceptance in these cases, people can choose to ta ke risks and pursue unlikely cures in hopes of such a miracle. However, the model plays th e odds by suggesting the response that will most often result in the best outcomes. The model does not provide specific suggestions regarding how news-givers should communicate their suggestion of th e best response in a way that insures recipients will respond as desired. Other researchers have addressed techniques of news tr ansmission in both the medical literature and in the lit eratures on persuasion and communica tion, but future research may be required to determine the specific application of that research to the goal of guiding newsrecipients towards desired responses. One strategy that may prove successful is for news-givers to help recipients reach accura te conclusions about the controll ability, likelihood, and severity of potential negative outcomes of bad news. The research reviewed earlier suggests that people naturally respond to bad news in light of these si tuational factors, but their assessment of these factors may be inaccurate or biased. News-givers can provide recipients with more objective information about the bad news, thus making desired responding more likely. Finally, the model does not specify precise ly how people should evaluate the three situational factors, or how to determine whether the factors are high or low. The situational factors fall on a continuum, and the distinction between high vs. low is relative. For example, bad news that is low in severity may be significantly more severe than neutral news, but it is low 60

PAGE 61

in severity compared to other types of bad news. Research ex amining peoples perceptions of various events, as well as the most effective re sponses to these events, will address the question of how to evaluate the situational factors of bad news. Future directions The first step for future research is to te st the effectiveness of the Bad News Response Model as a model for giving bad news well. Four questions deserve attenti on. First, are the four responses in the model exhaustive, or are there other possible re sponses? Second, do the responses suggested in the model, which derive from the three situational factors, produce the best quality of life? Third, how do people naturally respond to bad news under various circumstances, and can bad news-givers improve th e likelihood that people will make the desired responses? Fourth, how can news-givers best guide recipients towards a desired response once the desired response is determined? The model makes predictions for each of these questions, and studies are curren tly underway to test these predictions. A second direction for future research is to examine the specific ch aracteristics of the four responses to bad news. Table 2-1 makes pred ictions regarding the cognitive, emotional, and behavioral characteristics of each response. For example, we suggest that watchful waiting is characterized by low anxiety, high general positiv e affect, and low arousal. Studies examining peoples emotional states and activity level wh ile engaging in watchful waiting, and likewise active change and acceptance, can examine these characteristics. A third area for future research is the influence of individual differences on peoples responses to bad news. The model attempts to ma ke predictions that generalize across people and circumstances. However, indivi dual differences may affect responding in two ways. First, individual differences likely aff ect peoples natural responses to bad news. For example, selfefficacy could increase the likelihood of choos ing active change over the other response 61

PAGE 62

categories. Second, individual differences like ly influence both the actual and perceived experiences of the likelihood, seve rity, and controllability of ne gative outcomes. The 80and 40year old men with prostate cancer described earl ier provide one example of how differences such as age, priorities, and resources affect the be st response to bad news. For example, the same disease with the same prognosis has more severe consequences for the man with responsibilities to his family than for the man with few responsibilities. Although the disease may be equally severe for the two men, the consequences of the di sease on other areas of their lives are likely to differ in severity. Finally, future studies can examine the appl ication of the Bad News Response Model to different cultures and developmental stages. Several studies find that people give medical bad news differently in different cu ltures (Searight & Gafford, 2005). For example, patients in China often receive less information a bout their diagnoses than patient s in the United States (Tse, Chong, & Fok, 2003), and cancer patients in England report that their doctors used the word cancer much less frequently than patients in the United States (Newall et al., 1987). These findings suggest that the Bad News Response Mode l may apply across cultures, but it is possible that cultural values and traditions may affect the way in which some aspects of the model are applied. As such, culture may act as an individual difference va riable that affects natural responses to bad news. For example, differen ces in personal agency between Eastern and Western cultures may lead people to respond with active change more in the West than in the East, and this difference would affect the ease wi th which news-givers are able to guide people towards the three responses in different cultures. Furthermore, although people of all ages re ceive bad news, the cognitive and emotional responses of children are likely not comparable to those of late-adole scents or adults. Young 62

PAGE 63

children and adolescents may have a difficult time expressing co mplex emotional reactions and making complex decisions (Inhelder & Piaget, 1958). The Bad News Response Model may be applicable to all ages, but the nature of its applicability likely differs across developmental stages. For example, the model may apply better to the primary caregiver than to the child diagnosed with a severe illness, or better to the ad ult child than to the senile parent given news of failing health. The Bad News Response Model assu mes that recipients of bad news are in a position to choose between different possible responses. In the cases just described, the family member, not the primary recipient of the news will make decisions about treatment options. Coda The medical literature provides many useful sugge stions for giving bad news but falls short of providing an overarching goal for bad ne ws-givers. The Bad News Response Model represents an improvement over the existing wo rk on giving bad news by providing a framework that includes all types of bad news, incorporates a number of valuable goals for bad news transmission, and addresses the ro le of both the news-giver and the recipient. A comprehensive and systematic model of bad news transmission benefits not only people who must give bad news, but also those receiving the news. People receiving bad news must not only address the subject of the news itself, but also their emotiona l reactions to the news. Poor coping can lead to depression, anxiety, and other ment al health concerns. The Bad Ne ws Response Model strives to provide bad news-givers the tools they need to improve the recipients ability to respond effectively to the situation at hand. 63

PAGE 64

64 Table 2-1. Characteristics of the three response categories Watchful waiting Active change Acceptance Non-responding Anxiety Low High Moderate Low Positive affect Moderate Low Low High Negative affect Low High High Low Activity level Low High Moderate Low

PAGE 65

CHAPTER 3 STUDY 1 All bad news is not created equal. Some ne ws presents myriad options for taking action; other news provides only the option of accepta nce. Some news foreshadows additional bad outcomes in the future; other news is short-lived in its effects. Some news is serious; other news is relatively benign. The Bad News Response Mo del takes into account various facets of bad news and suggests that responses to bad news may be sensitiv e to the type of news people receive. In Study 1, we test several predictions of the Bad News Response Model in an academic domain. The purpose of Study 1 was to examine how th e controllabili ty, likelihood, and severity of the outcomes of bad news predict the responses people choose. The Bad News Response Model makes three predictions regarding the re lationship between the type of bad news and responses: Hypothesis 1 : People are more likely to engage in watchful waiting when perceived controllability, likelihoo d, and/or severity of outcomes are low than when they are high. Hypothesis 2 : People are most likely to engage in active change when perceived controllability, likelihoo d, and/or severity of outcomes are high than when they are low. Hypothesis 3 : People are most likely to engage in acceptance when they perceive the controllability of outcomes as low than when they perceive controllability as high. Methods Participants Participants were 230 undergraduates recru ited through the web-ba sed participant pool participating in sessions of one to five. Procedure All participants read eight scenarios descri bing a situation involv ing a poor grade on an exam (Appendix A). The scenarios varied based on a) the likelihood of a bad outcome (high vs. 65

PAGE 66

low likelihood of failing the course), b) the sever ity of the outcome (high vs. low importance of the course), and c) controllability of the outco me (high vs. low ability to improve the course grade). The order of all scen arios was counterbalanced. Participants then read descri ptions of the three response types (watchful waiting, active change, and acceptance) and indicated the extent to which they would engage in each response ( 1 = very unlikely, 9 = very likely ) and which response they would choose. Participants also indicated how likely they were to recei ve a poor grade based on the scenario ( 1 = very unlikely, 9 = very likely ), how severe that outcome would be if it occurred ( 1 = not at all bad, 9 = very bad), and how much control they had over improving their grade ( 1 = little or no control, 9 = full control ). Results Manipulation Checks To test the effectiveness of our manipulations, we conduc ted dependent t-tests on the three manipulation check items. The manipulations of controll ability, likelihood, and severity were quite successful. Analysis of the controllability manipulation check item revealed that participants rated the outcomes as less c ontrollable in the low control conditions ( M = 2.94, SD = 1.24) than in the high control conditions ( M = 4.78, SD = 1.49), t (229) = 20.10, p < .0001, d = 2.66. Analysis of the likelihood manipulation check item revealed that pa rticipants rated the negative outcome as less likely in the low likelihood condition ( M = 6.66, SD = 1.77) than in the high likelihood condition ( M = 7.18, SD = 1.41), t (229) = 5.86, p < .0001, d = .77. Analysis of the severity manipulation check item revealed that participants rated the outcomes as less severe in the low severity condition ( M = 6.50, SD = 1.34) than in the high severity condition ( M = 7.34, SD = 1.02), t (229) = 12.72, p < .0001, d = 1.68. 66

PAGE 67

Likert-Type Ratings We tested the first set of hypot heses using 2 (Controllability: high vs. low ) by 2 (Likelihood: high vs. low ) by 2 (Severity: high vs. low ) repeated measures ANOVAs on participants ratings of the likelihood that th ey would choose watchful waiting, active change, and acceptance. We had no predictions regarding wh ether or how the three factors might interact to influence participants. Nevertheless, we c onducted analyses to explore possible interactive effects. Watchful waiting We hypothesized that people w ould prefer watchful waiting when the controllability, likelihood, and severity of the negative outcome were low than when they were high ( Hypothesis 1). Our hypotheses were partly confirmed. First, participants were more likely to choose watchful waiting when likelihood was low ( M = 5.40, SD = 2.25) than when likelihood was high ( M = 3.18, SD = 1.62), F (1, 227) = 317.38, p < .0001, d = 2.36. Second, participants were more likely to choose watchful waiting when severity was low ( M = 4.58, SD = 1.95) than when severity was high ( M = 3.87, SD = 1.70), F (1, 227) = 79.07, p < .0001, d = 1.17. Contrary to predictions, participants were more likely to choose watchful waiting when controllability was high ( M = 4.43, SD = 1.92) than when controllability was low ( M = 4.02, SD = 1.76), F (1, 227) = 21.10, p < .0001, d = .61. We suspect that this unexpected result may be due to participants interpreting the high controllabil ity manipulation (i.e., the class has opportunities for extra credit that could improve the grade) as an indication that they could wa it until later in the semester to act. Analyses also yielded two unexpected interact ions for watchful waiting. The first was a Controllability by Likelihood interaction, F (1, 227) = 8.09, p < .01, d = .38 (Figure 3-1). Post hoc tests using the Bonferroni adjustment revealed that in high likelihood conditions participants 67

PAGE 68

were more likely to respond with watchful waiting when controllability was high ( M = 3.46, SD = 1.89) than when controllability was low (M = 2.89, SD = 1.71), t (229) = 5.42, p < .0001, d = .72. In low likelihood conditions, participants in high ( M = 5.40, SD = 2.40) and low ( M = 5.15, SD = 2.31) controllability conditions did not differ in their pr eference for watchful waiting, t (227) = 2.33, p = .02, d = .31. Analyses also revealed an unexpected three-way interaction for preferences for watchful waiting, F (1, 227) = 14.09, p < .001, d = .50. The marginal means for this interaction are shown in Table 3-1. Because th is interaction is not readily interpretable and was not predicted, we will not discuss it further. There were no other signif icant interactions for watchful waiting, all F s < 2.87, all p s > .09, all d s < .22. Active change We hypothesized that people would prefer active change when the controllability, likelihood, and severity of the negative outcome were high than when they were low ( Hypothesis 2). Once again, the analyses partly supported our hypot heses. First, participants were more likely to choose active change when likelihood was high (M = 7.58, SD = 1.13) than when likelihood was low ( M = 5.94, SD = 1.88), F (1 227) = 210.47, p < .0001, d = 1.92. Second, participants were more likely to choose active change when severity was high ( M = 7.18, SD = 1.23) than when severity was low ( M = 6.53, SD = 1.49), F (1, 227) = 90.61, p < .0001, d = 1.26. Unexpectedly, participants were equally likely to choose active change wh en controllability was high ( M = 6.79, SD = 1.36) and when controllability was low ( M = 6.91, SD = 1.36), F (1, 227) = 2.55, p = .11, d = .21. Analysis also yielded two unexpected inter actions for active change. The first was a Severity by Controll ability interaction, F (1, 227) = 19.16, p < .0001, d = .58 (Figure 3-2). Post hoc tests using the Bonferroni adjustment revealed that in high severity conditions participants were more likely to respond with active change when controllability was low ( M = 6.70, SD = 68

PAGE 69

1.59) than when controllability was high ( M = 6.36, SD = 1.69), t (229) = 3.79, p < .001, d = .50. In low severity conditions, participants in high (M = 7.23, SD = 1.34) and low ( M = 7.12, SD = 1.40) controllability conditions did not differ in their preference for active change, t (229) = -1.36, p = .18, d = .18. Analyses also revealed an unexpected thre e-way interaction for preferences for active change, F (1, 227) = 8.44, p < .01, d = .38. The marginal means for th is interaction are presented in Table 3-1. Because this interaction is not readily interpretable and was not predicted, we will not discuss it further. There were no other significant interactions for active change, all F s < 3.54, all ps > .06, all d s < .25. Acceptance We hypothesized that people would prefer acceptance more wh en the controllability of the negative outcome was low than when it was high ( Hypothesis 3 ). Analyses confirmed our hypothesis. Participants were more likely to choose acceptance when controllability was low (M = 4.74, SD = 1.66) than when controllability was high ( M = 4.58, SD = 1.66), F (1, 227) = 5.88, p = .02, d = .32. However, participants were also more likely to choose acceptance when likelihood was high ( M = 4.89, SD = 1.72) than when likelihood was low ( M = 4.51, SD = 1.81), F (1, 227) = 15.95, p < .0001, d = 53. The effect of severity was non-significant, F (1, 227) = .01. Analysis also yielded two unexpected inte ractions for acceptance. The first was a Severity by Controll ability interaction, F (1, 227) = 4.60, p = .03, d = .28 (Figure 3-3). Post hoc tests using the Bonferroni adjust ment revealed that in the high severity conditions participants were more likely to respond with acceptance when controllability was low ( M = 4.81, SD = 1.80) than when controllability was high ( M = 4.52, SD = 1.77), t (229) = 3.31, p < .01, d = .44. In the low severity condition, participants in the low ( M = 4.67, SD = 1.75) and high ( M = 4.63, SD = 1.82) controllability conditions did not differ in their preference for acceptance, t (229) = .41, p = 69

PAGE 70

.68, d = .05. The second was a Likelihood by Severity interaction, F (1, 227) = 6.16, p = .01, d = .33 (Figure 3-4). Post hoc tests re vealed that in high severity c onditions participants were more likely to respond with acceptance when likelihood was high ( M = 4.94, SD = 1.91) than when likelihood was low ( M = 4.38, SD = 1.94), t (227) = 4.30, p < .0001, d = .57. In contrast, in low severity conditions participants did not differ in their preference for a cceptance when likelihood was high ( M = 4.78, SD = 1.82) and when likelihood was low ( M = 4.53, SD = 1.87), t (229) = 2.33, p = .02, d = .31. There were no other significan t interactions for acceptance, all F s < .35, all ps > .56, all ds < .08. Forced-Choice Responses In addition to the Likert-type ratings for each response choice, participants also indicated which response they would choose if forced to pick one response to th e scenario. We conducted Chi-square analyses examining the effects of c ontrollability, likelihood, a nd severity with which participants chose each response. Our predictions for participants forced-choice responses were the same as our predictions fo r their Likert-type ratings. To examine our hypotheses, we conducted Chisquare analyses to compare the proportion of participants who chose each response (watch ful waiting, active change, or acceptance) in the high vs. low conditions of cont rollability, likelihood, and severi ty. The Chi-square analyses examined whether 50% of the response choices fell into the high and low conditions. In other words, we compared expected frequencies (50% of response choices falling into low conditions and 50% into high conditions, or chance responding) with actual estimate frequencies. If the situational factors had no effect on participan ts response choices, then, for example, 50% of participants who chose acceptance (or active change or watchful waiting) should have done so in the low controllability (or like lihood or severity) condition and 50% in the high controllability condition. In contrast, if our hypot hesis is correct, great er than 50% of participants who chose 70

PAGE 71

acceptance will have done so in the low controllability condition, and less than 50% of participants who chose acceptance will have don e so in the high controllability condition. Table 3-2 shows the number and percentage of participants who chose each response based on the manipulations of controllability, likelihood, severity, and role. The results were partly consistent with the hypotheses. Supporting Hypothesis 1, participants opted for watchful waiting more often when likelihood was low than when likelihood was high, 2(1, N = 484) = 203.72, p < .0001. Participants also opted for watchf ul waiting more often when severity was low than when severity was high, 2(1, N = 484) = 19.04, p < .0001. Contrary to predictions (yet consistent with the Likert-type responses), there was a marginally significant tendency for more participants to opt for watchful waiting when controllability was high than when controllability was low 2(1, N = 484) = 3.60, p = .06. As discussed earlier, we suspect that this unexpected result may be due to participants misinter pretation of our contro llability manipulation. Supporting Hypothesis 2, participants opted for active change more often when likelihood was high than wh en likelihood was low, 2(1, N = 1204) = 79.82, p < .0001. Participants also opted for active change more when severity was high than when severity was low, 2(1, N = 1204) = 6.73, p < .01. However, controllability had no effect on participants response choice, 2(1, N = 1204) = .65, p = .42. Supporting Hypothesis 3, there was a marginally significant tendency for participan ts to opt for acceptance more when controllability was low than when controllability was high, 2(1, N = 150) = 3.23, p = .07. However, likelihood and severity had no effect on participants response choice, 2s(1, N = 1204) = .65, ps = .42. Discussion Study 1 provided an initial test of th e Bad News Response Model a model for understanding and predicting responses to bad ne ws. We tested three hyp otheses in this study. First, as expected participants were more likely to choose watchful waiting when negative 71

PAGE 72

outcomes of the news were unlikely to occur and/or non-severe (Hypothe sis 1). Unexpectedly, however, participants were less likely to choose watchful wa iting when negative outcomes were uncontrollable. Second, participants were more likely to choose active change when negative outcomes were controllable and likely to occur (Hypothesis 2). However, we did not find that severity influenced choices of active change. Third, participants were more likely to choose acceptance when negative outcomes were uncontrol lable (Hypothesis 3) and relatively likely to occur. These findings are important in that they provide support for the Bad New Response Model and offer preliminary eviden ce that the model is useful in predicting responses to bad news. However, the numerous interaction effects ar e difficult to interpret and inconsistent with out predictions. We failed to find complete support for our three hypotheses, and we found a number of unexpected interac tions that make our supported h ypotheses difficult to interpret. Several limitations to this study may have contributed to these problems. First, using exam scenarios may have caused participants to add their own interpretations to the scenarios beyond our intended manipulations. That is students at the college level regularly receive exam grades, and as such they may have predetermined beliefs about what is realistic in their personal experience. For example, some participants may ha ve felt that even our controllable event would be uncontrollable in real life, or they may have believed that they could control even our supposedly uncontrollable event. Pa rticipants in this study may ha ve reinterpreted our scenarios in light of their pers onal experience, making our manipula tions of secondary concern. Second, participants may have interpreted the event as having drastically variable consequences, depending on their academic histor y. We used a D on a course exam as the negative event in this study. A D might be a devastating grad e for some students, whereas 72

PAGE 73

other students may regularly receive D grad es. Although we hoped that the within-subjects nature of our design would render these differences irrelevant, individual differences in academic history may have been a strong enough influenc e on responses to weaken our manipulations. 73

PAGE 74

2 3 4 5 6Low LikelihoodHigh Likelihood Watchful Waiting Low Control High Control Figure 3-1. Responses of watchful waiting as a f unction of the interaction between controllability and likelihood (Study 1). 4 5 6 7 8 Low ControlHigh Control Active Change Low Severity High Severity Figure 3-2. Responses of active change as a func tion of the interaction between controllability and severity (Study 1). 74

PAGE 75

2 3 4 5 6Low SeverityHigh Severity Acceptance Low Likelihood High Likelihood Figure 3-3. Responses of acceptance as a functio n of the interaction between likelihood and severity (Study 1). 2 3 4 5 6 Low SeverityHigh Severity Acceptance Low Control High Control Figure 3-4. Responses of acceptance as a function of the interaction between controllability and severity (Study 1). 75

PAGE 76

76 Table 3-1. Marginal means for 3-way interactions (Study 1) Watchful waiting Active change Condition M SD M SD Low controllability Low likelihood Low severity 5.31 2.56 5.96 2.23 High severity 5.00 2.47 6.34 1.92 Low controllability High likelihood Low severity 3.34 2.17 7.43 1.65 High severity 2.45 1.87 7.90 1.55 High controllability Low likelihood Low severity 5.90 2.61 5.53 2.26 High severity 4.90 2.57 6.66 1.85 Low controllability High likelihood Low severity 3.76 2.31 7.18 1.69 High severity 3.15 2.02 7.80 1.39 Table 3-2. Study 1 frequency an alyses of response choices Watchful waiting Active change Acceptance Condition Frequency % Frequency % Frequency % Controllability High 268 29% 588 64% 64 7% Low 216 24% 616 67% 86 9% Likelihood High 85 9% 757 82% 78 8% Low 399 43% 447 49% 72 8% Severity High 194 21% 647 70% 78 8% Low 290 32% 557 61% 72 8%

PAGE 77

CHAPTER 4 STUDY 2 Study 2 attempted to replicate and clarify th e findings from Study 1 in a health-related domain and using a between-subjects design. We designed Study 2 to improve some of the weaknesses of Study 1 in an effort to more cl early support the predictions of the Bad News Response Model. Whereas Study 1 used a familiar academic scenario, Study 2 used a health scenario that few participants are likely to have experienced. In addition, we used a betweensubjects design to reduce the likelihood that individual differences in experience would interfere with our manipulations. We also examined two additional hypotheses related to responses to bad news. Whereas the hypotheses tested in Study 1 examined the ef fects of the type of bad news on response decisions (Hypotheses 1-3), the additional hypothese s tested in Study 2 examine the possibility that responses to bad news may also depend on who chooses the response. In health-care situations, both patients and health-care professionals participate in trea tment decisions, but may disagree on the best course of treatment. Resear ch suggests that patients may prefer aggressive treatment options in an effort to do everything th ey can, even when such treatments are likely to do more harm than good. For example, one st udy of breast cancer patients found that 27% of women who reported that they made their own treatment decision opted for a mastectomy rather than a more conservative treatment. In contrast, only 16.8% of women who reported making the treatment decision jointl y with their surgeon and 5.3% of wo men who reported that their surgeon made the treatment decision underwent a mastectom y (Katz et al., 2005). This finding implies that people generally prefer act ive responses when they are the recipient of bad news but are more likely to suggest a more conservative respon se to someone else receiving bad news. 77

PAGE 78

The disconnect between the pref erences of the bad news-recipi ent and the bad news-giver could arise for a number of reasons. For example, patients often fail to un derstand the details of their diagnoses (Baile et al., 2000; Ellis & Tattersall, 1999; Fallowfield & Jenkins, 2004; Loge, Kaasa, & Hytten, 1997; Quill, 1991), which could lead them to make uninformed treatment decisions. Alternatively, patients may feel a need to control the outcomes of their disease (e.g., Verhoef & White, 2002), and this need for control may lead them to choose aggressive treatment options. One purpose of this study was to exam ine whether people are truly more likely to choose more active responses for themselves than for others. Hypothesis 4a: People are more likely to choose watchful waiting for someone else than for themselves. Hypothesis 4b : People are more likely to choose activ e change for themselves than for someone else. Finally, we examined the effect of experien ce with giving bad news on response choices. Although this hypothesis was somewhat speculative, we expected that experience with giving bad news increases understanding of the roles that controllability, likelihood, and severity play in choosing responses to bad news such that people with greater experience would recognize when each choice is most reasonable. Hypothesis 5 : People who report having experience givi ng bad news are more likely to make response suggestions that take controllability, likelihood, and severity into account (Hypotheses 1-3 in Study 1) than are people who report little expe rience giving bad news. That is, the effects predicted by Hypotheses 1-3 are stronger for pe ople with experience. Methods Participants Participants were 434 undergraduates recruited through the we b-based participant pool. 78

PAGE 79

Procedure Participants read one of 16 possible versi ons of a scenario about a person with a suspicious mole. The scenarios varied in terms of a) the controllability of the possible type of skin cancer, b) the likelihood that the mole is cancerous, c) the se verity of the possible type of skin cancer, and d) whether the participant took the role of the patient or the physician. Although we designed our scenarios to ma nipulate the variables of intere st rather than to precisely represent the situation patients might actually face in a dermatologists office, it is noteworthy that participants in a pilot test indica ted that the scenarios were believable. The patient condition scenario read as follows: Imagine that you make an appointment with your doctor about a suspicious mole on your back. After examining the mole, your doctor determines that, if the mole is cancerous, it most likely is a (non-severe / severe ) form of skin cancer that grows slowly and is unlikely to cause health pr oblems. Although the mole ( cannot / can ) be removed through surgery, your doctor determines that there is ( only a 1-2% / a 50-60% ) chance that the mole is cancerous. Your doctor decides to biops y the mole to determine if it is cancerous, and the biopsy results will be ready in 1-2 weeks. Similarly, the physician condition scenario read as follows: Imagine that you are a physician who must tell a patient about a suspicious mole on his back. After examining the mole, you determine that, if the mole is cancerous, it most likely is a ( non-severe / severe) form of skin cancer that grows slowly and is unlikely to cause health problems. Although the mole ( cannot / can ) be removed through surgery, you determine that there is ( only a 1-2% / a 50-60% ) chance that the mole is cancerous. You decide to biopsy the mole to determine if it is cancerous, and the biopsy results will be ready in 1-2 weeks. Participants next read descri ptions of the three response types (watchful waiting, active change, and acceptance) and indicated the likelihood that they would engage in ( patient condition ) or recommend ( physician condition ) each response ( 1 = very unlikely, 9 = very likely ). Participants also indicate d which response they would choose if forced to select one response. Participants then indicated how likely they were/the patient was to have cancer ( 1 = very unlikely, 9 = very likely ), how severe that outcome would be if it occurred ( 1 = not at all 79

PAGE 80

bad, 9 = very bad ), and how much control they/the patient had over the outcomes ( 1 = little or no control, 9 = full control ). Finally, participants in the phys ician condition indicated how much experience they had with giving bad news ( 1 = little to no experience giving bad news, 9 = frequent experience giving bad news ). Results Manipulations Checks To test the effectiveness of our manipulations, we conducted separate 2 (Controllability: high vs. low ) by 2 (Likelihood: high vs. low ) by 2 (Severity: high vs. low ) between-subjects ANOVA on the three manipulation ch eck items. The manipulations of controllability, likelihood, and severity were quite successful. Analysis of the controllability manipulation check item yielded a single main ef fect of controllability, F (1, 427) = 9.45, p < .01, d = .30. Participants rated the outcomes as less controllable in the low control condition ( M = 3.37, SD = 2.59) than in the high control condition ( M = 4.12, SD = 2.98).Analysis of the lik elihood manipulation check item yielded a single main effect of likelihood, F (1, 424) = 282.94, p < .0001, d = 1.63. Participants rated the negative outcome as less likely in the low likelihood condition ( M = 2.35, SD = 1.68) than in the high likelihood condition ( M = 5.11, SD = 1.71).Analysis of the severity manipulation check item yielded a main effect of severity, F (1, 426) = 41.55, p < .0001, d = .62. Participants rated the outcomes as less severe in the low severity condition ( M = 5.75, SD = 2.23) than in the high severity condition ( M = 7.08, SD = 2.11). Analysis also yielded an unexpected Severity by Controllability interaction for this item, F (1, 426) = 3.97, p = .05, d = .19. However, this interaction in no way qualified the main eff ect of severity. There were no other significant main effects or interactions, all F s < 1.41, all ps > .24, all d s < .12. 80

PAGE 81

Likert-Type Ratings We tested the first set of hypot heses using 2 (Controllability: high vs. low ) by 2 (Likelihood: high vs. low ) by 2 (Severity: high vs. low ) by 2 (Role: patient vs. physician ) between-subjects ANOVAs on partic ipants ratings of the likelihood that they would choose watchful waiting, active change, and acceptance. Watchful waiting We hypothesized that people w ould prefer watchful waiting when the controllability, likelihood, and severity of the negative outcome were low than when they were high ( Hypothesis 1) and when they were making a response suggestion for someone else vs. themselves ( Hypothesis 4a ). Our hypotheses were confirmed. First, participants were more likely to choose watchful waiting when controllability was low (M = 6.27, SD = 2.54) than when controllability was high ( M = 5.39, SD = 2.84), F (1, 416) = 10.80, p = .001, d = .32. Second, participants were more likely to choose watchful waiting when likelihood was low ( M = 6.23, SD = 2.63) than when likelihood was high (M = 5.45, SD = 2.76), F (1, 416) = 11.32, p = .001, d = .33. Third, participants were more likely to choose watchful waiting when severity was low ( M = 6.14, SD = 2.53) than when severity was high ( M = 5.54, SD = 2.87), F (1, 416) = 7.75, p < .01, d = .27. Finally, participants were more likely to choose watchful waiting for others ( M = 6.74, SD = 2.34) than for themselves ( M = 5.06, SD = 2.79), F (1, 416) = 43.48, p < .0001, d = .65. No other significant effects emerged for Watchful waiting, all F s < 3.28, all ps > .07, all ds < .18. Active change We hypothesized that people would prefer active change when the controllability, likelihood, and severity of the negative outcome were high than when they were low ( Hypothesis 2) and when they were making a response suggestion for themselves vs. someone else ( Hypothesis 4b ). Once again, the analyses supported our hypotheses. First, participants were 81

PAGE 82

more likely to choose active change when controllability was high ( M = 7.00, SD = 2.21) than when controllability was low ( M = 6.34, SD = 2.26), F (1, 418) = 7.55, p < .01, d = .27. Second, participants were more likely to choose active change when likelihood was high ( M = 7.03, SD = 2.07) than when likelihood was low ( M = 6.29, SD = 2.39), F (1 418) = 11.67, p < .001, d = .33. Third, participants were more likely to choos e active change when severity was high ( M = 6.92, SD = 2.09) than when severity was low ( M = 6.40, SD = 2.40), F (1, 418) = 8.97, p < .01, d = .29. Finally, analysis revealed a marginally signifi cant tendency for participants to prefer active change for themselves ( M = 6.82, SD = 2.26) than for someone else ( M = 6.48, SD = 2.24), F (1, 416) = 3.43, p < .06, d = .18. Analysis also yielded two unexpected inter actions for active change. The first was a Severity by Controll ability interaction, F (1, 418) = 3.95, p = .05, d = .19 (Figure 4-1). Post hoc tests using the Bonferroni adjust ment revealed that participants in the high severity condition were more likely to respond with active change when controllability was high ( M = 7.42, SD = 1.91) than when controllability was low ( M = 6.43, SD = 2.15), F (1, 418) = 11.68, p = .0007, d = .46. In the low severity conditi on, participants in the high (M = 6.55, SD = 2.43) and low ( M = 6.25, SD = 2.37) controllability condi tion did not differ in their preference for active change, F (1, 418) = .28, p = .60, d = .05. The second was a Severity by Role interaction, F (1, 418) = 3.81, p = .05, d = .19 (Figure 4-2). Post hoc tests revealed th at participants in the physicia n condition were more likely to respond with active change when severity was high ( M = 6.96, SD = 1.92) than when severity was low ( M = 5.94, SD = 2.46), F (1, 418) = 11.24, p = .0009, d = .47. In contrast, participants in the patient condition did not diffe r in their preference for active change when severity was high ( M = 6.89, SD = 2.24) vs. low ( M = 6.76, SD = 2.30), F (1, 418) = .60, p = .44, d = .08. 82

PAGE 83

Acceptance We hypothesized that people w ould prefer acceptance when the controllability of the negative outcome was low than when it was high ( Hypothesis 3 ). Analyses confirmed our hypothesis. Participants were more likely to choose acceptance when controllability was low (M = 4.12, SD = 2.61) than when controllability was high ( M = 3.50, SD = 2.12), F (1, 424) = 6.95, p < .01, d = .26. No other significant eff ects emerged for acceptance, all F s < 3.66, all p s > .06, all ds < .19. Forced-Choice Responses In addition to the Likert-type ratings for each response choice, participants also indicated which response they would choose if forced to pick one response to th e scenario. We conducted Chi-square analyses examining the effects of c ontrollability, likelihood, severity, and role on the frequency with which participants chose each res ponse. Our predictions for participants forcedchoice responses were the same as our pr edictions for their Li kert-type ratings. Table 4-1 displays the number and percentage of participants w ho chose each response based on the manipulations of controllability, likelihood, severity, and role. The results were generally consistent with th e hypotheses. Supporting Hypothesi s 1, participants opted for watchful waiting more often when controllability was high than when controllability was low, 2(1, N = 181) = 4.64, p = .03, and when likelihood was high than when it was low, 2(1, N = 181) = 4.02, p < .05. However, severity had no effect on participants response choices, 2(1, N = 181) = .04, p = .84. Supporting Hypothesis 2, participants opted for active change more often when controllability was low than when it was high, 2(1, N = 224) = 7.14, p < .01, and when likelihood was low than when it was high, 2(1, N = 224) = 5.79, p = .02. However, once again, severity had no effect on participants response chose, 2(1, N = 224) = .07, p = .79. 83

PAGE 84

The absence of an effect for severity on c hoices of watchful waiting and active change was surprising and inconsistent with the results fr om the Likert-type items. It is noteworthy that the mean response to the severity manipulation check item was above the scale midpoint of 5.0 in both the low severe and high severe condition. Perhaps the absen ce of an effect for severity was due to participants in both high and low seve rity conditions regarding the situation described in the scenario as above threshold in seriousness for a forced-choice response. Supporting Hypothesis 3, participants opt ed for acceptance more often when controllability was low than when controllability was high, 2(1, N = 30) = 8.53, p < .01. Supporting Hypothesis 4a and 4b, participants opted for watchful waiting more often when assuming the role of physician than when assuming the role of patient, 2(1, N = 181) = 7.56, p < .01, yet opted for active change more often when assuming the role of patient than when assuming the role of patient, 2(1, N = 224) = 18.28, p < .0001. Does News-Giving Experience Matter? Hypothesis 5 proposed that participants w ho reported having more experience with giving bad news would be more likely to suggest responses in line with our predictions (e.g., watchful waiting when controllability, likelihood, and severity were low, active change when controllability, likelihood, and severity were hi gh, and acceptance when controllability was low). To examine Hypothesis 5 we added experience (as a continuous variable after centering), to our 2 (Controllability) x 2 (Likelihood) x 2 (Severity) analysis of partic ipants Likert-type ratings of watchful waiting, active change, and acceptan ce (Cohen, Cohen, West, & Aiken, 2003). Recall that only participants in the physician condition completed the measure of news-giving experience. Support for Hypothesis 5 would appear as a significant tw o-way interaction of Experience and the other variables in the model. 84

PAGE 85

Regarding ratings of watchful waiting, only the two-way interacti on of Experience and Controllability approached significance, F (1, 182) = 3.57, p = .06, d = .28. For the purpose of illustration, we computed the mean watchful waiting response separately for participants in the low and high controllability conditions at three leve ls of experience: low (one standard deviation below the mean level of experience), medium (at the mean level of experience) and high (one standard deviation above the mean level of expe rience) (Cohen et al., 200 3). The results appear in Figure 4-3. As predicted, when controllabilit y was low, people with more experience were more likely to suggest watchful waiting than were people with less experience, t (104) = 2.09, p = .04. However, when controllability was high, leve l of experience was unrelated to preferences for watchful waiting, t (91) = -.52, p = .60. Analyses also revealed an unexpected four-w ay interaction for preferences for watchful waiting, F (1, 182) = 7.38, p = .007, d = .40. Because this interaction was not predicted, we did not examine it further. There were no other significant interactions for watchful waiting, all F s < .68, all ps > .41, all ds < .12. Regarding ratings of acceptance, the interact ion of Experience and Controllability was significant, F (1, 182) = 3.80, p = .05, d = .29. We once again plotted three levels of experience separately for participants in the high and low controllability conditions. The results appear in Figure 4-4. Contrary to predictions when controllability was low, people with more experience were less likely to suggest acceptance than were people with less experience, t (104) = -2.17, p = .03. When controllability was high, expe rience and acceptance were unrelated, t (91) = .50, p = .62. The interaction between Experience and Seve rity was also significant for ratings of acceptance, F (1, 182) = 10.15, p < .01, d = .47. The results appear in Figure 4-5. Contrary to our 85

PAGE 86

predictions, when severity was high, people with more experience were less likely to suggest acceptance than were people with less experience, t (105) = -3.92, p < .001. When severity was low, experience and acceptance were unrelated, t (89) = 1.11, p = .27. No other interactions for acceptance reached conventional levels of significance, all F s < 3.04, all p s > .08, all d s < .26. We also found no interactions involving experi ence for preferences for active change, all F s < 1.98, all ps > .16, all d s < .21. Discussion Study 2 had three goals. First, we attempted to replicate and clarify the findings of Study 1 using a different paradigm. Second, we examined whether people select different responses for themselves vs. someone else. Third, we examined the effect of experi ence of giving news on response choice. In most instances our hypotheses received strong support. First, as expected participants were mo re likely to choose watchful waiting when negative outcomes of the news were uncontrolla ble, unlikely to occur, and/or non-severe (Hypothesis 1 ) Second, participants were more likely to choose active change when negative outcomes were controllable, likely to occur, an d/or severe (Hypothesis 2). Third, participants were more likely to choose acceptance when ne gative outcomes were uncontrollable (Hypothesis 3). Study 2 was partly inconsiste nt with the Study 1, particular ly regarding the effects of controllability on respon ses of active change and watchful waiting and the effect of likelihood on responses of acceptance. Future studies can examine whether the model-supportive findings in Study 2 a replicable in different domains and using different methods Although it is possible that the model-inconsistent re sults in Study 1 are suggestive of limitations of the model, we strongly suspect that they are in stead reflective of limitations in the design of the scenarios used in Study 1. Finally, although we also found a few unexpected interact ions in Study 2, the results 86

PAGE 87

of this study provide stronger support for the Bad News Response Model in comparison to the results of Study 1. As predicted, participants were also more likely to choose watchful waiting for someone else vs. themselves and were more likely to ch oose active change for themselves vs. someone else ( Hypothesis 4a and 4b ). Finally, we speculated that participants who reported having experience giving bad news would make response s uggestions more in line with our predictions than would participants who repor ted having little experience ( Hypothesis 5 ). Only one finding supported this hypothesis. As predicted, when controllability was low participants with experience giving bad news were more likely to suggest watchful waiting than were participants with relatively little experien ce. However, no other effect of experience was in line with our predictions, and the effects of experience on sugg estions of acceptance we re inconsistent with our hypotheses. Why did news-giving experience do so littl e to guide people in suggesting responses? The most likely explanation is a restriction of range. Our participants were undergraduates, not health-care providers. It is likely that the participants classified as having high experience giving bad news actually had very little experience, a nd certainly not the kind of experience that would guide them in suggesting how a patient should respond to health-related news. Supporting this explanation are the findings that 80% of our participants report ed experience at or below the midpoint of our 9-point scale. Indeed, the mean experience rating was only 3.5. Clearly, research is needed examining participants with more experience giving bad news (physicians, nurses, etc.). 87

PAGE 88

4 5 6 7 8 9 Low SeverityHigh Severity Active Change Low Control High Control Figure 4-1. Responses of active change as a func tion of the interaction between controllability and severity (Study 2). 4 5 6 7 8 9 Patient Condition Physician Condition Active Change Low Severity High Severity Figure 4-2. Responses of active change as a functi on of the interaction betw een role and severity (Study 2). 88

PAGE 89

Table 4-1. Study 2 frequency an alyses of response choices Watchful waiting Active change Acceptance Condition Frequency % Frequency % Frequency % Controllability High 76 35% 588 62% 7 3% Low 105 48% 616 42% 23 10% Likelihood High 77 35% 757 58% 16 7% Low 104 49% 447 44% 14 7% Severity High 89 40% 647 51% 20 9% Low 92 43% 557 52% 10 5% Role Patient 72 31% 647 61% 19 8% Physician 109 55% 557 40% 11 5% 89

PAGE 90

6 6.5 7 7.5Low Control High ControlWatchful Waiting Low Exp Moderate Exp High Exp Figure 4-3. Suggestions of watc hful waiting by participants reporting low, moderate, and high news-giving experience and for low and high controllab ility (Study 2). 3 3.5 4 4.5Low Control High ControlAcceptance Low Exp Moderate Exp High Exp Figure 4-4. Suggestions of acceptanc e by participants reporting low, moderate, and high newsgiving experience and for low and high controllabili ty (Study 2). 90

PAGE 91

91 3 3.5 4 4.5Low Severity High SeverityAcceptance Low Exp Moderate Exp High Exp Figure 4-5. Suggestions of acceptanc e by participants reporting low, moderate, and high newsgiving experience and for low and high severity (Study 2).

PAGE 92

CHAPTER 5 STUDY 3 Study 3 examines how people perceive their response options. We propose that although different responses may be best suited to different types of news, people are more likely to choose some responses than others, regardless of the circumstances. We further suggest that two causes of differential response pr eferences may be 1) the emoti ons associated with a given response, and 2) the impressions people believe they will make by responding to bad news in a given way. To illustrate how impression management concerns could influence response choices in the face of bad news, imagine an 80-year-old man who receives a diagnosis of prostate cancer. After talking with several physicians, the man choos es not to receive treatment at that time and informs his family of his decision. The man, lik e many prostate cancer patients interviewed by researchers (Chapple et al., 2002), may face signi ficant opposition to his decision. His family members may see him as weak, preferring to see hi m as the type of man who would aggressively face any challenge that comes his way. In the fa ce of such a negative impression, the cancer patient may reverse his decision and undergo an unnecessary and invasive treatment. Our hypotheses regarding peoples perceptions of their response options are summarized in Table 5-1. Regarding emotional associati ons with response options, we hypothesized that people would associate the most positive emotio ns with active change, the least positive emotions with non-responding, and watchful waiting and acceptance somewhere in between. However, we also hypothesized that people might perceive active change which could require difficult or costly behaviors, as highly anxi ety-provoking in comparison with other responses. Based on the finding that people often prefer active responses to more conservative responses (Chapple et al., 2002; Denberg et al, 2006.; Mazur & Merz, 1996), we hypothesized 92

PAGE 93

that people would generally form the most positi ve impressions of a person who chooses active change (i.e., as capable and admirable), the most negative impressions of a person who chooses non-responding (i.e., as weak, vul nerable, and disengaged), and moderately positive impressions of a person who chooses accepta nce or watchful waiting. People may have nuanced beliefs regarding some responses, so Study 3 also examined two speculative hypotheses. First, despite its bene fits, active change entail s the risk that action might be premature, excessive, or ineffectiv e. Under some circumstances, taking immediate action could waste valuable time and energy, or take time and energy away from more important needs. For example, an elderly man who chooses expensive surgery for prostate cancer over a more conservative watchful waiting approach ma y compromise his familys financial security for little medical gain and risk dangerous si de-effects (Bangma et al., 2007). As such, we expected that people would anti cipate making an impression of impulsivity by responding with active change. In addition, non-re sponding corresponds closely with denial, which may be seen as an immediate and highly reactive (i.e., impul sive) response to unbearable bad news. In a sense, people who react to bad news with deni al are entirely focused on immediate gratification with little thought for long-term outcomes of their behavior. As such, we expected that nonresponding would also be seen as highly impulsive and that watchful waiting and acceptance would be seen as less impulsive. Second, previous research on acceptance as a re sponse to loss or tragedy suggests that some people require a great deal of time to r each the point of acceptance (Gamliel, 2000; KblerRoss, 1969). It may be that acceptance requires a hi gh degree of understanding or maturity that is achieved after engaging in other re sponses or with the passage of time. However, one pattern of responding to bad news cannot fit al l situations or people (Wortman & Silver, 1989; Silver et al., 93

PAGE 94

2004), and thus some people may engage in ac ceptance immediately upo n receiving bad news. We expected that people would perceive a person who responds with acceptance as having a high degree of maturity. We also anticipated that people woul d perceive a person who responded with active change as highly mature, in line wi th my other predictions regarding the general positivity of active change, and that people would perceive a person who responds with watchful waiting or non-responding as less mature. Methods Participants We recruited 58 undergraduates through the we b-based participant pool (45 males, 113 females). Procedure Participants responded to four questionnaires in random order. For each questionnaire participants were instructed to imagine that their best fr iend responded to some bad news (rather than specific news for th e purpose of generalizability) in one of four ways: watchful waiting, active change, acceptance, and non-resp onding. Each questionnaire provided a brief description of the response type. Be st friends were used in an effo rt to reduce biases that might result from participants j udging their own responses while maximizing the likelihood that participants could vividly im agine the target responding in each way. After reading each response description, par ticipants first responded to an it em that read, By responding with [watchful waiting, active change acceptance, or non-respondi ng], I think my friend would feel. Participants were then presented with 29 affect items, and they indicated the extent to which they thought their friend would feel each way (1 = strongly disagree 9 = strongly agree ). Participants then responded to an item that read, By responding with [watchful waiting, active change, acceptance, or non-responding], I thi nk my friend is. Part icipants were then 94

PAGE 95

presented with 38 trait items, and they indicated th e extent to which they would characterize their friend in each way (1 = strongly disagree 9 = strongly agree ). Results Affect Items Affect items were grouped into five reliable cat egories based on a priori assignments and pilot testing (N = 30). The resultant categories and the included items for each were as follows (Cronbachs alphas represent the results from the sample used in Study 1, N = 158): positive (excited, glad, happy, pl eased, enthusiastic; = .89), negative (depressed, sad, confused, angry, regretful, disappointed, distressed, weary; = .90), anxious (anxious, tense, scared, on edge, nervous, worried; = .90), calm (calm, relieved, serene; = .70), hopeful (hopeful, energetic, inspired, determined; = .76). Three items (alert, attentive, active) did not correlate well with any of the groupings or with one another; includ ing or excluding these items did not change the nature of our conclusions, so th ey will not be mentioned further. We hypothesized that participants would asso ciate different emotions with different responses to bad news. Did response type make a difference in how participants rated the affect items? Table 5-2 presents the means for each response type. To test our hypotheses we conducted a one-way repeated measures MANOVA for each of the trait categories. As shown in Table 5-2, analyses revealed signifi cant main effects of response type on all 5 affect categories. We further examined the specific pattern of m eans by testing specific contrasts between affect ratings for each response type. Recall our hypothesis that participants would rate active change most positively and non-respondi ng least positively across most affect categories. This hypothesis was confirmed for ratings of negative and hopeful and partially supported for ratings of positive (Table 5-2). We further hypothesized th at acceptance and watchful waiting would both be viewed as equally calm and more so than active change or non-responding. This 95

PAGE 96

hypothesis was partially supporte d. Although watchful waiti ng was seen as the least calm acceptance was seen as more calm than active change and non-responding. Finally, we anticipated that participan ts would rate active change and non-responding as more anxious than either watchful waiting or acceptance. This hypot hesis was also partially supported. Once again, watchful waiting was seen as the most anxious, but acceptance was seen as less anxious than either active change or non-responding (Table 5-2). Trait Items Trait items were grouped into seven re liable impression categories based on a priori assignments and pilot testing (N = 30). The resultant categories and items were: capable (capable, competent, res ponsible, practical, decisive, sensible, prepared; = .87), admirable (admirable, respectable; = .90), mature (mature, wise, careful, calm, relaxed; = .81), impulsive (impulsive, reckless, headstrong, impatient, overconfident; = .81), weak (weak, hopeless, pathetic, lazy, helpless; = .83), vulnerable (vulnerable, fragile, unstable; = .84), and disengaged (disengaged, indifferent, resigned, detached, apathetic; = .80). Six items (resilient, hesitant, powerless, cautious, controlled, and stoic) did not correlate well with any of the groupings or with one another; including or excl uding these items did not change the nature of the conclusions, so they ar e not discussed further. We hypothesized that participants form different impressions of their best friend based on how the friend responded to bad news. To test our hypotheses we conducted a one-way repeated measures MANOVA for each of the trait categories. As shown in Table 5-3, analyses revealed significant main effects of res ponse type on all 7 trait categorie s. We further examined the specific pattern of means by te sting specific contrasts between impression ratings for each response type. Recall our hypothesis that participan ts would rate active change most positively and non-responding least positively across most impression categories. This hypothesis was 96

PAGE 97

confirmed for ratings of capable, admirable weak, vulnerable, and disengaged (Table 5-3). We further hypothesized that acceptance and activ e change would be viewed as equally mature and more so than watchful waiting or non-respond ing. This hypothesis was fully supported. Finally, we anticipated that participants would ra te active change and non-responding as more impulsive than either watchful waiting or acceptance. This hypothesis was also full y supported (Table 5-3). Discussion The results of Study 3 suggest that on a wi de variety of measures, people view active change to be most positive, acceptance to be somewhat more positive than watchful waiting, and non-responding to be least positiv e. These results suggest that, regardless of the type of bad news, people perceive some responses to be mo re positive than others across a variety of measures. Of course, participants in Study 3 may have brought to mind a wide array of bad news examples. We did not specify the type of bad ne ws participants should imagine in this study to maximize the generalizability of our findings to different types of bad news, but as a result participants were free to imagine different type s of bad news in response to each of the four prompts. In other words, people may have imagined a type of bad news that demands action when asked to think about active change, but a type of bad news that calls for serene acceptance when asked to think about acceptance. Thus, people may have rated each response as more positive than they would have if they had imagined the same event across responses or imagined the response apart from any particular bad news. Although this limitation of Study 3 makes it difficult to predict how people would perceive ea ch response within a sp ecific context, it does not undermine the importance of the finding that people consistently rate some responses as more positive than others. In addition, although this study did not attemp t to examine whether these perceptions influence response choices, the emotional and imp ression associations people have with their 97

PAGE 98

response options may predict the responses people will choose. People may be particularly likely to avoid responses that they believe would make a poor impression on others, such as being seen as weak or immature (Schlenker, 1980). Combined with the results of Studies 1 and 2, the findings of Study 3 suggest that responses to ba d news may be foreseeable based not only on the type of news people receive but also their perceptions of their response options. 98

PAGE 99

99 Table 5-1. Predicted results (Study 3) Nonresponding Watchful waiting Active change Acceptance Affect categories Positive Low Moderate Moderate High Negative High Moderate Moderate Low Anxious High Low Low High Calm Low High High Low Hopeful Low Moderate Moderate High Impression categories Capable Low Moderate Moderate High Admirable Low Moderate Moderate High Mature Low Low High High Weak High Moderate Moderate Low Disengaged High Moderate Moderate Low Vulnerable High Moderate Moderate Low Impulsive High Low Low High Table 5-2. Affect category ra tings across response types Response type Affect category Nonresponding Watchful waiting Acceptance Active change F (3,154) p Positive 2.68a 2.90a 3.44 b 3.99c 21.90 < .0001 Negative 6.09a 5.58 b 5.10c 4.64d 27.28 < .0001 Anxious 6.25a 7.20 b 4.78c 5.97a 72.30 < .0001 Calm 3.16a 2.63 b 4.88c 3.92d 59.10 < .0001 Hopeful 3.36a 4.42 b 5.20c 6.46d 95.50 < .0001 Table 5-3. Impression category ratings across response types Response type Impression category Nonresponding Watchful waiting Acceptance Active change F (3,154) p Capable 3.34a 4.98 b 6.11c 7.13d 149.90 < .0001 Admirable 2.89a 5.31 b 6.83c 7.71d 191.19 < .0001 Mature 3.68a 5.11 b 6.13c 5.94c 72.51 < .0001 Impulsive 3.86a 3.19 b 3.22 b 4.09a 20.45 < .0001 Weak 4.75a 3.30 b 2.66c 1.76d 108.94 < .0001 Vulnerable 6.16a 4.85 b 4.02c 3.17d 75.45 < .0001 Disengaged 5.15a 3.84 b 3.63 b 2.66c 65.33 < .0001

PAGE 100

CHAPTER 6 STUDY 4 Studies 1 and 2 showed that the type of bad news people receive influences their response choices, and Study 3 showed that people associate specific emoti ons and impressions with each response option. Study 4 further examined the question of why people might choose certain responses over others by asking if people perceive certain responses to be more or less effective than other responses. More specif ically, we asked people to recall a time when they had received bad news and then indicate whether their respon se had positive or negative consequences. Like Study 3, Study 4 did not specifically examine whether people choos e certain responses based on perceptions of effectiven ess. Instead, this study serves as a preliminary examination of another potential influence on response choices. Based on both the general response preferences we found in Studies 1 and 2 (52% of participants chose active change in both Studies 1 and 2) and the beli efs people expressed in Study 3 (generally positive beliefs about active change, less positive beliefs about watchful waiting and acceptance), we expected that participan ts would perceive active change to be the most effective response choice, followed by wa tchful waiting and acceptance. That is, we hypothesized that participants who reported responding with ac tive change would report that their response had generally positive consequenc es, compared with participants who reported responding with watchful waiting or acceptance. Methods Participants 99 undergraduates were recruited through the we b-based participant pool and participated in sessions of one to five. 100

PAGE 101

Procedure Participants completed a questionnaire in response to both written instructions and instructions read by the experimenter. First, the experimenter aske d participants to write about a time when they received bad news. The experime nter encouraged participants to provide as much detail as they could remember, but they we re asked to limit this description to the time when they received the news and not to discuss anything after that experience. The experimenter described several examples to clarify the desired response. Participants th en indicated how likely they were to experience negative outcomes as a result of the bad news ( 1 = very unlikely, 9 = very likely ), how severe or important they expected these outcomes to be at the time they received the bad news ( 1 = not severe, 9 = very severe ), and how much control they had over the consequences of the news at the time they received it ( 1 = no control, 9 = complete control ). These items were included to test the genera lizability of the findings in Studies 1 and 2. Participants then wrote about their res ponse to the bad news, including thoughts, emotions, and actions. Participants read descri ptions of the three response types (watchful waiting, active change, and acceptance) and indicated the extent to which they had engaged in each response ( 1 = not at all, 9 = very much ) and which response best described their response to the situation they described. Finally, participants wrote about shortand long-term consequences of their response to the news a nd indicated how their re sponse affected their consequences or outcomes ( 1 = made things much worse, 9 = made things much better ). Participants also indicated whether they experien ced negative consequences as a result of their response to the bad news ( yes or no ), and if so, how long these negative outcomes lasted ( 1 = not very long, 9 = a very long time) and how severe they were (1 = not very severe, 9 = extremely severe ). 101

PAGE 102

Results Situational Factors and Response Ratings We hypothesized that participants ratings of controllability, li kelihood, and severity would be related to their respons es to the bad news, as found in Studies 1 and 2. The relevant correlations are reported in Table 6-1. First, we hypothesized that participants would be more likely to report responding with watchful waiting to the extent that the controllability, likelihood, and severity of the news were low. This hypot hesis was not supported. Pa rticipants ratings of controllability, likelihood, and severity were not significantly co rrelated with the extent to which participants engaged in watchful waiting, all r s < 14, all p s > .18. Second, we hypothesized that participants woul d be more likely to report responding with active change to the extent that the controllability, likelihood, and severity of the news were high. This hypothesis was partially supported. Participants were more likely to report responding with active change to the extent that th e controllability of the news was higher, r (99) = .51, p < .0001. There was also a non-significant trend towa rds participants respond ing more with active change when the severity of the bad news was higher, r (99) = .17, p = .10. However, responses of active change were not significantly correlated with re port of likelihood, r (99) = .12, p = .23. Finally, we hypothesized that participants w ould be more likely to report responding with acceptance to the extent that controllabil ity was low. This hypothesis was supported, r (99) = .38, p = .0001. Thus, the results of Study 4 partially, but not completely, replic ated the results of Studies 1 and 2. That is, as in Studies 1 and 2 participants were more likely to respond with active change when controllability and severity were high and we re more likely to respond with acceptance when controllability was low. However, participants in this study were no more likely to respond with watchful waiting when controllability, likelihood, a nd severity were low, or with active change wh en likelihood was high. 102

PAGE 103

Consequences of Responding The primary purpose of Study 4 was to ex amine participants perceptions of the consequences of various types of bad news and various responses to bad news. Table 6-1 displays the correlations between the three res ponse measures (watchfu l waiting, active change, and acceptance) and the three consequence measur es (positivity/negativity of consequences, duration of any negative consequences, and severity of negative consequences). Table 6-1 also displays the correlations between each of the situational factors (controllabilit y, severity, and likelihood) and the three consequence measures. Likert-type ratings of responding We hypothesized that participants would perc eive more positive consequences and fewer negative consequences to the extent that they reported responding w ith active change. This hypothesis was partially supported. Participants indicated that th eir response had a more positive effect on their outcomes to the extent that they responded with active change, r (99) = .41, p < .0001. To examine the relationship between responses and the duration and severity of negative consequences, we included only participants who indicated that they experienced negative consequences as a result of their response ( n = 36). Regarding active change, we found a marginally significant relationship between report ed responses of active change and the duration of negative consequences, r (36) = -.30, p = .08, such that participants who reported responding with active change indicated that negative consequences were shorter in duration. However, responses of active change were not signifi cantly correlated with severity of negative consequences, r (36) = -.18, p = .30. We also hypothesized that participants would perceive mo re negative consequences and fewer positive consequences to the extent that they reported responding with watchful waiting. 103

PAGE 104

This hypothesis was also partially supported. Part icipants indicated that their response had greater negative consequences to the extent that they responded with watchful waiting, r (99) = .29, p < .01. We also found a non-significant trend between reported responses of watchful waiting and the severity of negative consequences, r (36) = .28, p = .10, such that participants who reported responding with watc hful waiting indicated that nega tive consequences were more severe. However, responses of watchful waiting were not significantly correlated with duration of negative consequences, r (36) = .03, p = .87. Finally, we also hypothesized that participants woul d perceive more negative consequences and fewer positive consequences to the extent that they responded with acceptance. This hypothesis was fully supported. Participants indicated that their response had a more negative effect on their outcomes to the extent that they responded with acceptance, r (99) = -.38, p = .0001. We also found a significant relati onship between reported responses of acceptance and the duration of negative consequences, r (36) = .46, p < .01, such that participants who reported responding with acceptance indicated that negative consequences lasted longer. Furthermore, we found a significant relationship between reported responses of acceptance and the severity of negative consequences, r (36) = .43, p < .01, such that participants who reported responding with acceptance indicated that nega tive consequences were more severe. Forced-choice measures of responding In addition to the Likert-type ratings for each response choice, participants also indicated which response they engaged in by choosing fr om a list of the three response types. We examined the effects of the forced-choice res ponse selections on percep tions of consequences. Our predictions for participants forced-choice responses were the same as our predictions for their Likert-type ratings. That is, we predicted that participants who indicated that they responded with active change would perceive mo re positive consequen ces and fewer negative 104

PAGE 105

consequences, and participants who indicated th at they responded with watchful waiting or acceptance would perceive more negative cons equences and fewer positive consequences. We conducted a one-way ANOVA with three levels (choice of watchful waiting ( n = 13), active change ( n = 54), or acceptance ( n = 32) for each of the consequence measures. Analyses revealed a main effect of res ponse choice on overall consequences, F (2, 96) = 7.52, p < .001, d = .56. We then conducted planned contrasts compari ng the consequences of each response. As predicted, participants who res ponded with active change percei ved more positive consequences ( M = 6.98, SD = 2.23) than did participants w ho responded with watchful waiting ( M = 5.62, SD = 1.45), F (1, 96) = 4.45, p = .04, d = 43, and more positive consequences than did participants who responded with acceptance (M = 5.25, SD = 2.08), F (1, 96) = 13.70, p < .001, d = .75. There was no difference in perceived consequences betw een participants who responded with watchful waiting and participants who responded with acceptance, F (1, 96) = .28, p = .60, d = .11. For the measures of duration and severity of negative consequences we again included in our analyses only participants w ho indicated that they experien ced negative consequences of their response ( n = 36). Analyses revealed a signifi cant main effect of response choice on duration of negative consequences, F (2, 33) = 3.84, p = .03, d = .66. We once again conducted planned contrasts comparing the duration of negati ve consequences for each response choice. As predicted, participants who re sponded with active change ( n = 54) indicated that negative consequences were shorter in duration ( M = 3.59, SD = 2.06) than did participants who responded with acceptance ( n = 32) ( M = 5.67, SD = 2.64), F (1, 33) = 6.52, p = .02, d = .86. There was also a marginally significant differenc e in duration of negative consequences between participants who responded with watchful wa iting and participants who responded with acceptance, F (1, 33) = 3.49, p = .07, d = .63. However, participants who responded with active 105

PAGE 106

change did not perceive the duration of negative consequences differently than did participants who responded with watchful waiting (n = 13) ( M = 3.25, SD = 1.71), F (1, 33) = .07, p = .79, d = .09. Finally, there were no significant effects of response choice on severity of negative consequences, F (2, 33) = 2.11, p = .14, d = .49. Situational factors and consequences of responding We also explored the relationship between the situational factors of the bad news (controllability, likelihood, and severity) and the consequences of responding (Table 6-1). Regarding general consequences of responding, part icipants indicated that their response had a more positive effect on their outcomes to the extent that the event was controllable, r (99) = .24, p = .02. Participants also indicated that their response had more negative consequences to the extent that the event was likely to result in negative outcomes, r (99) = -.21, p = .03. Regarding duration of consequences, we found a positive relations hip between likelihood of negative outcomes and the dura tion of negative consequences, r (36) = .35, p = .04. For events that were likely to produce negative outcomes, part icipants indicated that negative consequences of their responses were longer in duration. We also found a positive relationship between likelihood of negative outcomes and the severity of negative consequences, r (36) = .47, p < .01. For events that were likely to produce negative outcomes, particip ants indicated that negative consequences of their responses were more severe. Finally, we found a marginally positive relationship between severity of negative event outcomes and th e severity of negative response consequences, r (36) = .29, p = .08. For events that had severe negative consequences, participants indicated th at negative consequences of their re sponses were more severe. No other correlations between situational factors and response consequences were significant, r s < .27, p > .10. 106

PAGE 107

Mediation Analyses We explored the possibility that the rela tionships between situational factors and response consequences might be mediated by the way participants responded to the bad news. However, as described by Baron and Kenny (1986) we could not examine mediations involving likelihood or severity because thes e factors were not correlated with any of the three response measures (watchful waiting, active change, and ac ceptance). As such, we focus our attention on potential mediators of the effect of controll ability on response consequences. We hypothesized that uncontrollable events might predict negative consequences because people are more likely to respond to these events with acceptance. In contra st, perhaps controllable events predict positive consequences because people are more likely to respond to these events with active change. To examine the mediating ro le of acceptance, three condi tions must hold true. First, controllability (the predictor) of the bad news should correlat ed with consequences (the outcomes). As previously mentioned, it did, r (99) = .24, p < .02. Second, the extent to which people responded with acceptance (the mediator) should be significantly correlated with both the controllability of the bad news and the reported general consequences. It did, both r s(99) = -.38, p < .0001. Finally, to establish mediation, the relations hip between controllability of the bad news and consequences should be signifi cantly reduced when the effects of the mediator (acceptance) are partialled out (Baron & Kenny, 1986). We conducted multiple regression to test this condition. Before acceptance was entered into the regression, controllability of the bad news significantly predicted outcomes of the event, = .24, t (97) = 2.40, p = .02. However, when acceptance was added to the regression, controllability no longer predicted outcomes, = .11, t (96) = 1.07, p = .29. Thus, the extent to which participants responded with acceptance completely mediated the relationship between cont rollability and outcomes as a result. We also 107

PAGE 108

conducted a Sobel test as a more conservative measure of the mediating effect of acceptance. The Sobel test for mediation was significant, z = 2.53, p = .01, further supporting the mediating role of acceptance in the relationship betw een controllability and outcomes. To examine the mediating role of active cha nge responses, we first tested the appropriate bivariate correlations. As a reminder, controlla bility (the predictor) of the bad news was positively correlated with general consequences (the outcomes). Second, the extent to which people responded with active change (the mediator) was significantly correlated with both the controllability of the bad ne ws and general consequences, r s(99) = .51 and .41, ps < .0001. Finally, we conducted multiple regression to test whether active change mediated the relationship between controllab ility and outcomes. When activ e change was added to the regression, controllability no longer predicted outcomes, = .04, t (96) = .35, p = .73. Thus, the extent to which participants responded with active change also completely mediated the relationship between controllab ility and outcomes. We also conducted a Sobel test of the mediating effect of active change. The S obel test for mediation was significant, z = 3.05, p < .01, further supporting the mediating ro le of active change in the rela tionship between controllability and outcomes. Discussion Combined with the findings of Study 3, the results of Study 4 sugge st that people may prefer active change because they perceive it to be both the most desirable and most effective response, far more so than watchful waiting a nd acceptance. Although participants in Study 4 did not indicate that the situ ational factors of bad ne ws predicted their responses in quite the same ways as were shown in Studies 1 and 2, Study 4 suggests that the perceived effectiveness of response options may be powerful predictor of responding. 108

PAGE 109

However, the impact of these findings is lim ited by the retrospective nature of this study. That is, participants reflected on an experience with bad news that occurred in the past, and their memories of the event and of the consequences of their response may be less than accurate. As such, we can only conclude with confidence that people perceive in hindsi ght that active change led to positive consequences, whereas watchf ul waiting and acceptance led to negative consequences. This finding may not be surprising in light of rese arch suggesting that people tend to regret actions in the short-term but inactions over longer periods of time (Gilovich & Medvec, 1995). That is, if participants described bad news events that occu rred far in the past, they may have been more likely to report that choos ing watchful waiting or acceptance was a mistake simply because they represent relatively inactiv e choices. Researchers suggest that people may regret inactions over time for several reasons (Gilovich & Medvec, 1995). First, people may forget the reasons that they were reluctant to take action at the time (e.g., fear, lack of confidence, etc.). Second, nega tive consequences of choices pe ople make are far more salient than potential negative cons equences of choices they dont make. Thus, future research can pinpoint whether people trul y perceive active change to be th e most effective response or if people simply regret inactions more than acti ons when reflecting on their own experiences. However, in light of the fact that people may choose responses based in part on their memories of how the responses turned out in the past, the results of Stud y 4 provide a sense of the role perceived effectiveness may play when people respond to bad news. 109

PAGE 110

110 Table 6-1. Correlations between situational f actors, responses, and response consequences 1 2 3 4 5 6 7 8 9 1. Controllability 1.0 2. Likelihood -.03 1.0 3. Severity .02 .68* 1.0 4. Watchful waiting -.14 -.04 -.11 1.0 5. Active change .51* .12 .17+ -.47* 1.0 6. Acceptance -.38* .16 .00 .31* -.54* 1.0 7. Positive/negative consequences .24* -.21* .12 -.29* .41* -.38* 1.0 8. Duration of consequencesa -.27 .35* .23 .03 -.30+ .46* -.55* 1.0 9. Severity of consequencesa -.16 .47* .29+ .28+ -.18 .43* -.53* .72* 1.0 a Includes only participants who indicated that the response had negative consequences ( n = 36). p < .05 + p < .10

PAGE 111

CHAPTER 7 GENERAL DISCUSSION The goal of the present set of studies was to examine several questions related to the Bad News Response Model. First, Studies 1 and 2 addr essed how the type of bad news people give or receive affects response choices. In these studies we had severa l hypotheses. Regarding watchful waiting, we anticipated that people would be more likely to select watchful waiting when the controllability, likelihoo d, and severity of negative outcomes were low. This hypothesis was largely supported. In both Studies 1 and 2, part icipants were more likely to choose watchful waiting when likelihood and severity were low vs high, and in Study 2 participants were more likely to choose watchful waiting when controllability was low vs. high. Regarding active change, we anticipated that people would be more likely to choose active change when the controllability, likelihoo d, and severity of negative outcome s were high. Again, this hypothesis was largely supported. In both Studi es 1 and 2, participants were more likely to choose active change when likelihood and severity were high vs low, and in Study 2 participants were more likely to choose active change when controllabili ty was high vs. low. Regarding acceptance, we anticipated that people would be more likely to choose acceptance when controllability was low vs. high. This hypothesis was suppor ted in both Studies 1 and 2. Finally, Study 2 also examined the possibility that the role people play (i.e., news-giver vs. news-recipient) would affect their response ch oices. Indeed, we found th at people playing the role of the news-giver tended to suggest watchf ul waiting, whereas people pl aying the role of the news-recipient tended to choose active change. This last finding is particularly interesting in light of the findings of Study 3, which showed that people view watchful waiting in a generally negative light. Although people were relatively unlikely to choose watchful waiting for themselves, they apparently viewed watchful waiting as somewhat more appropriate for others. 111

PAGE 112

Perhaps participants playing th e role of the physician in Study 2 were less focused on the negative beliefs they may hold about watchful wa iting and more focused on the best suggestion they could make as a responsible physician. Of course, we found several inconsistencie s in Studies 1 and 2. Most notably, we predicted that controllability would predict resp onses of watchful waiting or active change, but these hypotheses were not supported in Study 1. As discussed earlier, these inconsistencies may indicate limitations of the ability of the Bad News Response model to predict responses. However, we believe it to be more likely that th e exam scenarios used in Study 1 led participants to reinterpret our scenarios to make them c onsistent with their personal experience with academic bad news. That is, participants in Study 1 may not have seen our scenarios as believable given their typical exam performance, or they may have interpreted the bad news of a D grade differently depending on their typical exam performance. Future studies will examine whether the more model-consistent findings in Study 2 generalize to additional domains and methodologies. Second, Studies 3 and 4 addressed the possibi lity that people may prefer certain responses over others, regardless of the type of bad news they receive. In Study 3, we hypothesized that people would associate primar ily positive emotions and impressions with active change, primarily negativ e emotions and impressions w ith non-responding, and moderate emotions and impressions with acceptance and watchful waiting. This hypothesis was supported. Participants in Study 3 indicated that they felt most positive and hopeful and least negative when thinking about active change. Partic ipants also indicated that so meone who responds with active change is capable and admirable and not weak, vulnerable, or disengaged. Participants felt exactly the opposite abou t non-responding: they felt most ne gative and least positive and hopeful 112

PAGE 113

when thinking about non-responding, and they felt that someone who responded this way was weak, vulnerable, and disengaged and not capab le or admirable. As predicted, people had moderate associations with watchful waiting and acceptance, although people generally saw acceptance as somewhat more positive than watchful waiting. Finally, Study 4 examined how people perceive the effectiveness of response options. We hypothesized that people may generally prefer active change in part because they believe it to be the most effective response. As expected, participants reported that active change had positive consequences in a personal experience receiving bad news. In contrast, pa rticipants reported that watchful waiting and acceptance had negative consequences in a personal experience receiving bad news. Furthermore, the respon se participants reported mediated the effects of controllability on negative outcomes, such that choosing active ch ange when outcomes were controllable led to positive consequences, and choosing acceptance when outcomes were uncontrollable led to negative consequences. Implications As a whole, the results of these four studies suggest that the type of bad news people receive, their feelings about each potential response and their perceptions of those responses effectiveness may combine to predict response choices. Furthermore, the findings of Studies 1 and 2 provide support for the predictions of the Bad News Response Model, and the findings of Studies 3 and 4 suggest additiona l avenues for predicting and improving responses to bad news. These findings can assist bad news-givers w ho otherwise must rely on their own limited experience or personal motivations when giving bad news. For example, bad news-givers can use these findings to evaluate their transmission of news after the fact. If news-givers observe recipients making an undesired response, they can examine their transmission strategy in light of these findings. The news-giver may have incorre ctly assessed one or mo re of the situational 113

PAGE 114

factors, or s/he may have failed to account for the news-recipients personal beliefs about the response options. For example, physicians may be unaware of their patients financial circumstances, and this lack of information could result in misjudgment of the severity and/or controllability of patients medical conditions. Physicians might assume that expensive treatments are feasible when in fact the patient does not have insurance or the means to pay for the treatments, making the prognosis relatively uncontrollable. Alterna tively, physicians might assume that patients find a watch and wait approach appealing when in fact patients believe this approach to be undesirable and ineffective. Furthermore, bad news-recipients can use thes e findings to evaluate their responses to bad news, apart from the giver. After receiving bad news, recipients can evaluate the likelihood, severity, and controllability of the possible outcomes to understand how they should respond. For example, a woman who learns of upcoming la yoffs at work can consider the likelihood that she will lose her job, how bad the consequences of a job loss would be, and if she has control over whether she is laid off. Ha ving evaluated the situation, she ma y have a better sense of the most effective response. This process may help pe ople to override response s based solely on fear or anxiety, misguided impression management concerns, or narrow beliefs about potential response effectiveness. In addition, recipients who find that their response to some news is ineffective can reexamine the s ituational factors involved and possibly adjust their responses accordingly. If the woman facing a possible job loss responds with active change and then finds that she is making no progress towards keeping her job, she may deci de to shift towards acceptance by checking the want ads and telling her family about the layoffs. Limitations and Future Directions Although the studies presented here provide a clear sense of how people are likely to respond to bad news, they stop short of demonstra ting how responses affect shortor long-term 114

PAGE 115

outcomes. As such, the next step for future resear ch is to test the effectiveness of particular responses to bad news under different circum stances. The Bad News Response Model makes specific suggestions as to which responses will fa re best given the controllability, likelihood, and severity of negative outcomes (Table 1-1), a nd future studies can examine whether these predictions hold true. Of course Study 4 provides some sense of the most and least effective responses across situations, but the retrospective nature of th is study limits our ability to generalize from these findings. A second limitation of our studies is the la rgely hypothetical nature of the study designs. Studies 1-3 asked participants to imagine how they might respond to bad news or how they would feel about various responses to bad news, and as such their responses may or may not reflect how they would respond in a real-world situation. Although Study 4 attempted to use participants personal ex periences with bad news to exam ine responding and consequences of responding, the retrospective nature of this design may also limit the conclusions we can draw. Thus, another direction for future research is to examine responding and perceptions of responding in real bad news experiences. Althou gh we have little reason to believe that responses will be systematically different in re al vs. hypothetical situations, it is possible that people measure more carefully the cost of their response options when faced with a potentially consequential decision. A third area for future research is the infl uence of individual differences on peoples responses to bad news. The model attempts to ma ke predictions that generalize across people and circumstances. However, indivi dual differences may affect responding in two ways. First, individual differences likely aff ect peoples natural responses to bad news. For example, selfefficacy could increase the likelihood of choos ing active change over the other response 115

PAGE 116

categories. Second, individual differences like ly influence both the actual and perceived experiences of the likelihood, seve rity, and controllability of ne gative outcomes. The 80and 40year old men with prostate cancer described earl ier provide one example of how differences such as age, priorities, and resources affect the be st response to bad news. For example, the same disease with the same prognosis has more severe consequences for the man with responsibilities to his family than for the man with few responsibilities. Although the disease may be equally severe for the two men, the consequences of the di sease on other areas of their lives are likely to differ in severity. Finally, future studies can examine the appl ication of the Bad News Response Model to different cultures and developmental stages. Several studies find that people give medical bad news differently in different cu ltures (Searight & Gafford, 2005). For example, patients in China often receive less information a bout their diagnoses than patient s in the United States (Tse, Chong, & Fok, 2003), and cancer patients in England report that their doctors used the word cancer much less frequently than patients in the United States (Newall et al., 1987). These findings suggest that the Bad News Response Mode l may apply across cultures, but it is possible that cultural values and traditions may affect the way in which some aspects of the model are applied. As such, culture may act as an individual difference va riable that affects natural responses to bad news. For example, differen ces in personal agency between Eastern and Western cultures may lead people to respond with active change more in the West than in the East, and this difference would affect the ease wi th which news-givers are able to guide people towards the three responses in different cultures. Furthermore, although people of all ages re ceive bad news, the cognitive and emotional responses of children are likely not comparable to those of late-adole scents or adults. Young 116

PAGE 117

117 children and adolescents may have a difficult time expressing co mplex emotional reactions and making complex decisions (Inhelder & Piaget, 1958). The Bad News Response Model may be applicable to all ages, but the nature of its applicability likely differs across developmental stages. For example, the model may apply better to the primary caregiver than to the child diagnosed with a severe illness, or better to the ad ult child than to the senile parent given news of failing health. The Bad News Response Model assu mes that recipients of bad news are in a position to choose between different possible responses. In the cases just described, the family member, not the primary recipient of the news will make decisions about treatment options. Conclusions The Bad News Response Model is a systema tic and theoretical model of responding to bad news, and several of the present findings provide empirical support for several of the models predictions. Our findings suggest that pe ople do not take a one-si ze-fits-all approach when responding bad news but rath er tailor their responses to the type of news they receive. Most importantly, responses to bad news are not ar bitrary but instead follow predictable patterns according to the controllability, likelihood, and se verity of negative consequences that may follow from the news. Our findings further suggest that although both news-recipients and newsgivers are sensitive to these variations in bad ne ws, news-recipients lean toward active responses and news-givers toward more conservative resp onses. Finally, news-re cipients might choose particular responses based not only on the type of news they face, but also on the perceived emotional and social consequences and the perceived effectiveness of their response options. These considerations may explain why news-recipients prefer activ e responses over more conservative responses. Future studies can use our findings as a starting point to both predict and improve peoples responses to bad ne ws under different circumstances.

PAGE 118

APPENDIX A STUDY 1 SAMPLE QUESTIONNAIRE (Participants read one of the following 8 scen arios and answered the following questions): 1) Imagine that you receive a D on a course exam. The exam counts for 50% of your grade in this course, so it is very likely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. However, the course includes several opportunities for extra credit that could significantly improve your grade. In addition, you are taking this course with the pass/fail option. 2) Imagine that you receive a D on a course exam. The exam counts for 50% of your grade in this course, so it is very likely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. The course does include seve ral opportunities for extra credit that could significantly improve your grade. However, you cannot take this course pass/fail. 3) Imagine that you receive a D on a course exam. The exam counts for 50% of your grade in this course, so it is very likely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. In addition, the course has no opportunities for extra credit to improve your grade, and dropping the course is not an op tion. However, you are taking this course pass/fail. 4) Imagine that you receive a D on a course exam. The exam counts for 50% of your grade in this course, so it is very likely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. In addition, the course has no opportunities for extra credit to improve your grade, and dropping the course is not an opti on. Furthermore, you cannot take this course pass/fail. 5) Imagine that you receive a D on a course exam. The exam counts for only 10% of your grade in this course, so it is very unlikely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. Furthermore, the course in cludes several opportunitie s for extra credit that could significantly improve your grade. In addition, you are ta king this course pass/fail. 6) Imagine that you receive a D on a course exam. The exam counts for only 10% of your grade in this course, so it is very unlikely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. Furthermore, the course in cludes several opportunitie s for extra credit that could significantly improve your grade. Howe ver, you cannot take this course pass/fail. 7) Imagine that you receive a D on a course exam. The exam counts for only 10% of your grade in this course, so it is very unlikely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. In additi on, the course has no opportunities for extra credit to improve your grade, and dropping the c ourse is not an option. Howeve r, you are taking this course pass/fail. 8) Imagine that you receive a D on a course exam. The exam counts for only 10% of your grade in this course, so it is very unlikely that the poor exam grade could lead to a poor grade (i.e., a C) in the course. In additi on, the course has no opportunities for extra credit to improve your grade, and dropping the c ourse is not an option. Furthe rmore, you cannot take this course pass/fail. Read carefully the following three descriptions of possible responses to the exam grade: Watchful Waiting: This response involves a wait and see mentality regarding the course. People engaged in this response are aware that th ey are facing a possible threat. However, they 118

PAGE 119

go about life as usual rather than take action. For example, if you re sponded in this way you would not get a tutor or spend much extra time studying. Instead, you would go about the final half of the course just as you did the first half. Active Change: This response involves specific responses directed toward addressing the bad news. It includes three types of behavior: information-seeking, prevention, and addressing the problem. For example, if you responded in this wa y you might talk to others who have taken the course before you, get a tutor, or study harder for the final exam. Accommodation and Acceptance: This response involves two t ypes of behavior: informationsharing and accommodation. Information-shari ng involves telling othe r people about the negative event. Accommodation involves making cha nges, not to affect the outcome, but rather to incorporate the negative event into ones li fe. For example, if you responded in this way you might tell your parents about the grade and l ook into ways to improve your GPA the following semester. 1. How likely would you be to respond with Watchful Waiting? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 2. How likely would you be to respond with Active Change? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 3. How likely would you be to respond with Accommodation and Acceptance? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 4. Of the three responses, how would you most li kely respond to the failing exam grade? (Check one) _____ Watchful Waiting _____ Active Change _____ Accommodation and Acceptance 5. Imagine that the event described in the scenar io was really happening to you. How likely is it that, if you remained in the course you would receive a poor grade (i.e., a C)? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 119

PAGE 120

120 6. How bad would it be if you received a poor grade (i.e., a C) in the course? 1 2 3 4 5 6 7 8 9 Not at all Bad Very Bad 7. How much control would you ha ve in improving your grade? 1 2 3 4 5 6 7 8 9 Little or No Control Full Control

PAGE 121

APPENDIX B STUDY 2 SAMPLE QUESTIONNAIRE (PATIENT CONDITION) Imagine that you make an appointment with your doctor about a suspicious mole on your back. After examining the mole, your doctor determines that, if the mole is cancerous, it most likely is a (non-severe / severe) form of skin cancer that grows (slowly / quickly) and is (unlikely / likely) to cause health problems. Although the mo le (cannot / can) be removed through surgery, your doctor determines that there is (only a 1-2% / a 50 -60%) chance that the mole is cancerous. Your doctor decides to biopsy th e mole to determine if it is cancerous, and the biopsy results will be ready in 1-2 weeks. The following descriptions represent three possible types of responses to bad news. Please read each description carefully. Watchful Waiting: This res ponse involves a wait and see mentality. People engaged in this response are aware that they are faci ng a possible threat. However, they go about life as usual rather than take ac tion. For example, if the patient in the scenario engaged in Watchful Waitin g, s/he would not undergo surgery or make any significant life changes. Instead, s/he would check the mole periodically for changes. Active Change: This response involves specific actions directed toward addressing the bad news. It includes three types of beha vior: information-seeking, prevention, and treatment. For example, if the patient in the scenario engaged in Active Change, s/he may read up on skin cancer, get a second opinion, stay out of the sun, and/or have the mole removed. Acceptance: This response involves coming to terms with bad news rather than taking action to change the s ituation. Acceptance involves making changes, not to affect the outcome, but rather to incorporate the negative event into ones life. For example, if the patient in the scenario engaged in Acceptance, s/he may update his/her will and tell others about the mole to get social support. 1. How likely would you be to recommend Watchf ul Waiting to the pati ent in the scenario? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 2. How likely would you be to recommend Active Change to the patient in the scenario? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 121

PAGE 122

122 3. How likely would you be to recommend Acce ptance to the patien t in the scenario? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 4. Of the three responses, which woul d you be most likely to recommend? (Check one) _____ Watchful Waiting _____ Active Change _____ Acceptance 5. How likely is it that the patien t in the scenario has cancer? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 6. How bad would it be if the patient in the scenario does have cancer? 1 2 3 4 5 6 7 8 9 Not at all Bad Very Bad 7. How much control does the patie nt in the scenario have over the situation at this point? 1 2 3 4 5 6 7 8 9 Little or No Control Full Control

PAGE 123

APPENDIX C STUDY 2 SAMPLE QUESTIONNAIRE (PHYSICIAN CONDITION) Imagine that you are a physician who must tell a patient about a suspicious mole on his back. After examining the mole, you determine that, if the mole is cancerous, it most likely is a (non-severe / severe) form of skin cancer that grows (slowly / qui ckly) and is (unl ikely / likely) to cause health problems. Although the mole (cannot / can) be removed through surgery, you determine that there is (only a 1-2% / a 50-60 %) chance that the mole is cancerous. You decide to biopsy the mole to determine if it is cancerous, and the biopsy results will be ready in 1-2 weeks. The following descriptions represent three possible types of responses to bad news. Please read each description carefully. Watchful Waiting: This res ponse involves a wait and see mentality. People engaged in this response are aware that they are faci ng a possible threat. However, they go about life as usual rather than take ac tion. For example, if the patient in the scenario engaged in Watchful Waitin g, s/he would not undergo surgery or make any significant life changes. Instead, s/he would check the mole periodically for changes. Active Change: This response involves specific actions directed toward addressing the bad news. It includes three types of beha vior: information-seeking, prevention, and treatment. For example, if the patient in the scenario engaged in Active Change, s/he may read up on skin cancer, get a second opinion, stay out of the sun, and/or have the mole removed. Acceptance: This response involves coming to terms with bad news rather than taking action to change the s ituation. Acceptance involves making changes, not to affect the outcome, but rather to incorporate the negative event into ones life. For example, if the patient in the scenario engaged in Acceptance, s/he may update his/her will and tell others about the mole to get social support. 1. How likely would you be to recommend Watchf ul Waiting to the pati ent in the scenario? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 2. How likely would you be to recommend Active Change to the patient in the scenario? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 123

PAGE 124

124 3. How likely would you be to recommend Acce ptance to the patien t in the scenario? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 4. Of the three responses, which woul d you be most likely to recommend? (Check one) _____ Watchful Waiting _____ Active Change _____ Acceptance 5. How likely is it that the patien t in the scenario has cancer? 1 2 3 4 5 6 7 8 9 Very Unlikely Very Likely 6. How bad would it be if the patient in the scenario does have cancer? 1 2 3 4 5 6 7 8 9 Not at all Bad Very Bad 7. How much control does the patie nt in the scenario have over the situation at this point? 1 2 3 4 5 6 7 8 9 Little or No Control Full Control

PAGE 125

125 APPENDIX D STUDY 3 SAMPLE QUESTIONNAIRE Imagine that your best friend received some bad news. The following questionnaire asks you to imagine four possible ways your best friend could respond to bad news. For each possible response, read the description car efully and then answer the subsequent questions about how you think your friend would be fee ling after responding in the described manner. Watchful Waiting This response involves a wait and see mentality. People engaged in this response are aware that they are facing a possible threat. However, they go about life as usual rather than take action. They may wait to see if the situation resolves itself, or they may wait until they have more information before making an active response. For the items below, please indicate the extent to which you think your friend would feel each emotion as a result of responding with Watchful Waiting. Use the scale below to respond to the following questions: 1 2 3 4 5 6 7 8 9 Strongly Strongly Disagree Agree I think my friend would feel _____ Anxious _____ Relieved _____ Depressed _____ Active _____ Tense _____ Calm _____ Inspired _____ Scared _____ Alert _____ Excited _____ Sad _____ Glad _____ Serene _____ Confused _____ Angry _____ Attentive _____ Happy _____ On edge _____ Nervous _____ Regretful _____ Disappointed _____ Determined _____ Energetic _____ Distressed _____ Pleased _____ Enthusiastic _____ Worried _____ Weary _____ Hopeful

PAGE 126

APPENDIX E STUDY 4 QUESTIONNAIRE The experimenter will read inst ructions for each question on this form. Please do not respond to any questions until the experimenter has comp leted the instructions for that question. 1. Please describe a time you received bad news: ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ 2. When you received this bad news, how likely was it that negative consequences or outcomes (including emotional, financial, hea lth-related, academic, etc.) would follow? 1 2 3 4 5 6 7 8 9 Not Likely Very Likely 3. When you received this bad news, how importan t or severe did you expect the consequences or outcomes (including emotional, financial, health-related, academic, etc. ) of the news to be? 1 2 3 4 5 6 7 8 9 Not Severe Very Severe 4. When you received this bad news, to what ex tent did you have control over the consequences or outcomes (including emotional, financial, health-related, academic, etc.) of the news? 1 2 3 4 5 6 7 8 9 No Control Complete Control 5. Please describe how you responded to the bad news. How did you feel? What were your thoughts? What actions did you take? If you responde d in more than one way, please describe all of the ways you responded (continue on the back if you need more space). ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ 126

PAGE 127

Below are a variety of things a person might do in response to bad news. Please indicate the extent to which you engaged in each in res ponse to the bad news on the following scale: 1 2 3 4 5 6 7 8 9 Not at All Very Much (If the statement is Not Applicable indicate that with an x.) _____ 6. I tried to distract myself by thinking about other things. _____ 7. I took immediate action to address th e situation. _____ 8. I resigned myself to the fact that my life was in some way different now. _____ 9. I thought of reasons why this event was not a cause for alarm. _____ 10. I learned more about what options we re available to improve the situation. _____ 11. I focused my energy towards understanding and accepting the situation. _____ 12. I went about my life as though nothing had changed. _____ 13. I made changes in my life to keep the situation from getting worse. _____ 14. I told others about the situa tion so they could help me cope. _____ 15. I took a wait and see a pproach to the situation. _____ 16. I set up a plan to take action. _____ 17. I spent time alone to try to adjust to the situation. The following descriptions represent three possible types of responses to bad news. Please read each description carefully. Watchful Waiting: This response involves a wait and s ee mentality. People engaged in this response are aware that they are facing a possible threat. However, they go about life as usual rather than take action. For example, people en gaging in Watchful Waiting might do things to take their mind off the bad news and simp ly check in on the situation periodically. Active Change: This response involves specific actions di rected toward addressing the situation. It includes behaviors like seeking information, trying to change th e situation for the better, and trying to prevent the situation from getting worse. For example, people engaging in Active Change might talk to other people in similar situ ations and look for ways to directly deal with their situation. Acceptance: This response involves coming to terms with bad news rather than taking action to change the situation. Accep tance involves making changes, not to affect the outcome, but rather to incorporate the negative event into ones li fe. For example, people engaging in Acceptance might tell friends and family about the bad news to get social support and adjust their future plans as a result of their situation. 127

PAGE 128

18. To what extent would you say you responded to the bad news with Watchful Waiting ? 1 2 3 4 5 6 7 8 9 Not at All Very Much 19. To what extent would you say you responded to the bad news with Active Change ? 1 2 3 4 5 6 7 8 9 Not at All Very Much 20. To what extent would you say you responded to the bad news with Acceptance ? 1 2 3 4 5 6 7 8 9 Not at All Very Much 21. Of the three responses, which one describes best how you responded to the bad news? If you responded in more than one way, select the one that best describes your primary response. (Check one) _____ Watchful Waiting _____ Active Change _____ Acceptance 22. Write about the short-term and long-term consequences or outcomes of the bad news (including emotional, financial, health-related, academic, etc.). Specifically address how your response to the situation affected the outcomes. In other words, did your response to the news make the situation better or worse, or have no impact? ______________________________________________________________________________ ______________________________________________________________________________ ______________________________________________________________________________ 23. How did your response to the situation affect the consequences or outcomes? -4 -3 -2 -1 0 1 2 3 4 Made Things Much Worse No Effect Made Things Much Better 128

PAGE 129

129 If the consequences or outcomes of your responses were not nega tive, check this box instead of answering #24 and #25 (if the outcomes we re negative, proceed to #24): 24. If the consequences or outcomes of your resp onse were negative, how long did the negative outcomes last? 1 2 3 4 5 6 7 8 9 Not Very Long A Very Long Time 25. If the consequences or outcomes of your resp onse were negative, how important or severe were they? 1 2 3 4 5 6 7 8 9 Not Very Severe Extremely Severe

PAGE 130

LIST OF REFERENCES Ajzen, I., & Fishbein, M. (1980). Understanding attitudes an d predicting behavior. Englewood Cliffs, NJ: Prentice Hall. Aldwin, C. M. (1991). Does age affect the st ress and coping process? Implication of age differences in perceived control. Journal of Gerontology, 46, 174-180. Aldwin, C. M., & Park, C. L. (2004). Coping and physical health outcomes: An overview. Psychology and Health, 19, 277-281. Aldwin, C. M., & Revenson, T. A. (1997). Does coping help? A reexamination of the relation between coping and mental health. Journal of Personality and Social Psychology, 53, 337-348. Ambuel, B., & Mazzone, M. F. (2001). Breaking bad news and discussing death. Primary Care, 28, 249-267. Anderson, C. R. (1977). Locus of control, coping behaviors, and performan ce in a stress setting: A longitudinal study. Journal of Applied Psychology, 62, 446-451. Aspinwall, L., & Taylor, S. (1997). A stitch in time: Self-regulation and proactive coping. Psychological Bulletin 121, 417-436. Back, A. L., & Curtis, A. R. (2002). Communicating bad news. Western Journal of Medicine, 176, 177-180. Baile, W. F., & Aaron, J. (2005). Patient-physician communication in oncology: Past, present, and future. Current Opinion in Oncology, 17, 331-335. Baile, W. F., Buckman, R., Lenzi, R., Glober, G., Baile, E. A., & Kudelka, A. P. (2000). SPIKES: A six-step protocol for delivering bad news. Oncologist, 5, 302-311. Baile, W. F., Kudelka, A. P., Beale, E. A., Glober, G. A., Myers, E. G., Greisinger, A. J., Bast Jr., R. C., Goldstein, M.G., Novak, D., & Lenzi, R. (1999). Communication skills training in oncology. Cancer, 86, 887-897. Baron, R. & Kenney, D. (1986). The moderator-m ediator variable distinction in social psychological research: Conceptu al, strategic and statistical considerations. Journal of Personality and Social Psychology, 51 (6), 1173-1182. Becker, M. H. (Ed.) (1974). The health belief model and persona l health behavior. Health Education Monographs, 2 Bor, R., Miller, R., Goldman, E., & Scher, I. (1 993). The meaning of bad news in HIV disease: Counseling about dreade d issues revisited. Counseling Psychology Quarterly, 6, 69-80. 130

PAGE 131

Boyd, J. R. (2001). A process for delivering ba d news: Supporting families when a child is diagnosed. Journal of Neuroscience Nursing, 33, 14-20. Brandtstadter, J., Dirk, W., & Werner, G. (1993). Adaptive resources of the aging self: Outlines of emergent perspectives. International Journal of Behavioral Development 16, 323349. Brewin, T. B. (1991). Three ways of giving bad news. The Lancet, 337, 1207-1209. Bruhn, J. G. (1984). Therapeutic value of hope. Southern Medical Journal, 77, 215-219. Buckman, R. (1984). Breaking bad news : Why is it still so difficult? British Medical Journal, 288, 1597-1599. Butow, P. N., Kazemi, J. N., Beeney, L. J., Gr iffin, A., Dunn, S. M., & Tattersall, M. H. N. (1996). When the diagnosis is cancer: Patient communication experiences and preferences. Cancer, 77, 2630-2637. Carroll, P. J., Sweeny, K., & Shepperd, J. A. (2006). Forsaking optimism. Review of General Psychology, 10, 56-73. Carver, C. S., Scheier, M. F., & Weintra ub, J. K. (1989). Assessing coping strategies: A theoretically ba sed approach. Journal of Personality and Social Psychology, 56, 267-283. Charlton, R. C. (1992). Breaking bad news. Medical Journal of Australia, 157, 615-621. Clark, R. E., & LaBeff, E. E. (1982). Death te lling: Managing the delivery of bad news. Journal of Health and Social Behavior, 23, 366-380. Clayton, J. M., Butow, P. N., Arnold, R. M., & Ta ttersall, M. H. N. (2005). Fostering coping and nurturing hope when discussing the future with terminally ill cancer patients and their caregivers. Cancer, 103, 1965-1975. Cleary, T. (2001). The wisdom of the prophet: Sayings of Muhammad, selections from the Hadith. Boston: Shambhala. Crocker, J. (1982). Biased questions in judgment of covariation studies. Personality and Social Psychology Bulletin, 8, 214-220. Croyle, R. T., Loftus, E. F., Klinger, M. R., & Smith, K. D. (1993). In J. R. Schement & B. D. Ruben (Eds.), Between communication and information. Information and behavior, Vol. 4 (255-268). New Brunswick, NJ: Transaction Publishers. Croyle, R. T., & Williams, K. D. (1991). Reactio n to medical diagnosis: The role of illness stereotypes. Basic and Applied Social Psychology, 12, 227-241. Damian, D., & Tattersall, M. H. N. (1991). Le tters to patients: Improving communication in cancer care. Lancet, 338, 923-926. 131

PAGE 132

Davis, C. G., Nolen-Hoeksema, S., & Larson, J. (1998). Making sense of loss and benefiting from the experience: Two construals of meaning. Journal of Personality and Social Psychology, 75, 561-574. De Haes, H., & Koedoot, N. (2003). Patient centered decision making in palliative cancer treatment: A world of paradoxes. Patient Education and Counseling, 50, 43-49. Dean, A. (2002). Talking to dying pa tients of their hopes and needs. Nursing Times, 98, 34-35. Derdiarian, A. K. (1989). Effects of informa tion on recently diagnosed cancer patients and spouses satisfaction with care. Cancer Nursing, 12, 285-292. Dias, L., Chabner, B. A., Lynch, T. J., & Penson, R. T. (2003). The Oncologist, 8, 587-596. Doyle, D., & OConnell, S. (1996). Break ing bad news: Starti ng palliative care. Journal of the Royal Society of Medicine, 89, 590-591. Dunn, S.M., Butow, P.N., Tattersall, M.H., Jone s, Q.J., Sheldon, J.S., Taylor, J.J., & Sumich, M.D. (1993). General information tapes i nhibit recall of the cancer consultation. Journal of Clinical Oncology, 11, 2279-2285. Edwards, W. (1954). The theory of decision making. Psychological Bulletin, 51, 380-417. Eggly, S., Afonso, N., Rojas, G., Baker, M., Cardoza, L., & Robertson, R. S. (1997). An assessment of residents competence in the delivery of bad news to patients. Academic Medicine, 72, 397-399. Ellis, P. M., & Tattersall, M. H. N. (1999). How should doctors communicate the diagnosis of cancer to patients? Annals of Medicine, 31, 336-341. Epstein, R. M., Alper, B. S., & Quill, T. E. (2006). Communicating eviden ce for participatory decision making. Journal of the American Medical Association, 291, 2359-2366. Fallowfield, L., & Jenkins, V. (2004). Communicating sad, bad, and difficult news in medicine. Lancet, 363, 312-319. Fallowfield, L., Jenkins, V., & Beveridge, H. A. (2002). Truth may hurt but deceit hurts more: Communication in palliative care. Palliative Medicine, 16, 297-303. Faulkner, A. (1998). ABC of palliative care: Comm unication with patients, families, and other professionals. British Medical Journal, 316, 130-132. Floyd, D., Prentice-Dunn, S., & Rogers, R. (2000) A meta-analysis of research on protection motivation theory. Journal of Applied Social Psychology 30, 407-429. Fogarty, L. A., Curbow, B. A., Wingard, J. R., McDonnell, K., & Somerfield, M. R. (1999). Can 40 seconds of compassion reduce patient anxiety? Journal of Clinical Oncology, 17, 371379. 132

PAGE 133

Folkman, S., & Lazarus, R. S. (1980). An anal ysis of coping in a middle-aged community sample. Journal of Health and Social Behavior, 21, 219-239. Folkman, S., & Lazarus, R. S. (1985). If it change s it must be a process: A study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology, 48, 150-170. Freedman, B. (1993). Offering trut h: One ethical approach to the uniformed cancer patient. Archive of Internal Medicine, 153, 572-576. Gamliel, T. (2000). The lobby as an arena in the confrontation between a cceptance and denial of old age. Journal of Aging Studies, 14, 251-271. Gillotti, C., Thompson, T., & McNeilis, K. (200 2). Communicative competence in the delivery of bad news. Social Science and Medicine, 52, 1011-1023. Gilovich, T., & Medvec, V. H. (1995). The experience of regret: What, when, and why. Psychological Review, 102, 379-395. Girgis, A., Sanson-Fisher, R. W., & Schofield, M. J. (1999). Is there consensus between breast cancer patients and providers on gu idelines for breaking bad news? Journal of Behavioral Medicine, 25, 69-77. Goldie, L. (1982). The ethics of telling the patient. Journal of Medical Ethics, 8, 128-133. Greer, S., Morris, T., & Pettingale, K. W. (1979). Psychological res ponse to breast cancer: Effect on outcome. Lancet, 785-787. Groopman, J. (2004). The anatomy of hope: How people prevail in the face of illness. New York: Random House. Hagerty, R. G., Butow, P. N., Ellis, P. M., L obb, E. A., Pendlebury, S. C., Leighl, N., Mac Leod, C., & Tattersall, M. H. N. (2005). Comm unicating with realism and hope: Cancer patients views on the disclosure of prognosis. Journal of Clinical Oncology, 23, 12781288. Harber, K. D., & Pennebaker, J. W. (1992). Ov ercoming traumatic memories. In S. A. Christianson (Ed.), The handbook of emotion and memory (pp. 359). Hillsdale, NJ: Erlbaum. Hogbin, B., Jenkins, V. A., & Parkin, A. J. (1992). Remembering bad news consultations: An evaluation of tape-recorded consultations. Psychooncology, 1, 147-154. Holland, J. C. (1989). Now we tell, but how well? Journal of Clinical Oncology, 7, 557-559. Hurwitz, C. A., Duncan, J., & Wolfe, J. (2004). Ca ring for the child with cancer at the close of life: There are people who make it, and Im hoping Im one of them. Journal of the American Medical Association, 292, 2141-2149. 133

PAGE 134

Inhelder, B., & Piaget, J. (1958). The growth of logical thinking: From childhood to adolescence. New York: Basic Books, Inc. Janz, N., & Becker, M. (1984). The h ealth belief model: A decade later. Health Education Quarterly 11 1-47. Jemmot, J. B., Ditto, P. H., & Croyle, R. T. ( 1986). Judging health status: Effects of perceived prevalence and personal relevance. Journal of Personality & Social Psychology, 50, 899905. Kahneman, D., & Tversky, A. (1982). On th e study of statis tical intuitions. Cognition, 11, 123141. Kelly, A. E., & McKillop, K. J. (1996). Consequences of revealing personal secrets. Psychological Bulletin, 120, 450-465. Kirscht, J. (1988). The health belief model and pr edictions of health actions. In D. S. Gochman (Ed.), Health behavior: Emerging research perspectives (pp. 27-41). New York: Plenum Press. Kbler-Ross, E. (1969). On death and dying New York: Macmillan. Lalos, A. (1999). Breaking bad news concerning fertility. Human Reproduction, 14, 581-585. Lamont, E. B., & Christakis, N. A. (2001). Prognostic disclosure to patients with cancer near the end of life. Annals of Internal Medicine, 134, 1096-1105. Langer, E. J. (1975). The illusion of control. Journal of Personality and Social Psychology, 32, 311-328. Langer, E. J., & Roth, J. (1975). Heads I win, ta ils it's chance: The illusion of control as a function of the sequence of outcomes in a purely chance task. Journal of Personality and Social Psychology, 32, 951-955. Lazarus, R. S. (1966). Psychological stress and the coping process New York: McGraw-Hill. Lazarus, R. S. (1981). The stress and coping paradigm. In C. Eisdorfer, D. Cohen, A. Kleinman, & P. Maxim (Eds.), Models for clinical psychopathology (pp. 177-214). New York: Spectrum. Lazarus, R. S. (1985). The costs and benefits of de nial. In A. Monat & R. S. Lazarus (Eds.), Stress and coping (2nd ed., pp. 154). New York: Columbia University Press. Lazarus, R. S. & Launier, R. (1978). Stress -related transactions between person and environment. In L. A. Pervin & M. Lewis (Eds.), Perspectives in interactional psychology (pp.287-327). New York: Plenum Press. 134

PAGE 135

Lerman, C., Daly, M., Walsh, W. P., Resch, N., Seay, J., Barsevick, A., Birenbaum, L., Heggan, T., & Martin, G. (1993). Communication between patients with breast cancer and health care providers: Determinants and implications. Cancer, 72, 2612-2620. Links, M., & Kramer, J. (1994). Breaking bad news: Realistic versus unrealistic hopes. Supportive Care in Cancer, 2, 91-93. Loge, J. H., Kaasa, S., & Hytten, K. (1997). Di sclosing the cancer di agnosis: The patients experiences. European Journal of Cancer, 33, 878-882. Lubinsky, M. S. (1994). Bearing bad news: Dealing with the mimics of denial. Journal of Genetic Counseling, 3, 5-12. Maddux, J., & Rogers, R. (1983). Protection motivation and self-efficacy: A revised theory of fear appeals and attitude change. Journal of Experiment al Social Psychology 19, 469479. Mager, W. M., & Andrykowski, M. A. (2002) Communication in the cancer bad news consultation: Patient perceptions and psychological adjustment. Psycho-Oncology, 11, 35-46. Maguire, P. (1998). Breaking bad news. European Journal of Surgical Oncology, 24, 188-199. Mast, M. S., Kindlimann, A., & Langewitz, W. (2005). Recipients perspectives on breaking bad news: How you put it really makes a difference. Patient Education and Counseling, 58, 244-251. McHugh, P., Lewis, S., Ford, S., Newlands, E., Ru stin, G., Coombes, C., Smith, D., OReilly, S., & Fallowfield, L. (1995). General informa tion tapes inhibit recall of the cancer consultation. British Journal of Cancer, 71, 388-392. MacLeod, C., & Campbell, L. (1992). Memory accessibility and probability judgments: An experimental evaluation of th e availability heuristic. Journal of Personality and Social Psychology, 63, 890-902. McClenahen, L., & Lofland, J. (1976). Bearing bad news: Tactics of the Deputy U. S. Marshal. Sociology of Work and Occupations, 3, 251-272. Michaels, E. (1983). Deliver bad news tactfully. Canadian Medical Association, 129, 1307-1308. Myers, B. A. (1983). The informing interview: Enabling patients to h ear and cope with bad news. American Journal of Disabled Children, 137, 572-577. Newall, D. J., Gadd, E. M., & Priestman, T. J. (1987). Presentation of information to cancer patients: a comparison of two centres in the UK and USA. British Journal of Medical Psychology, 60, 127-131. 135

PAGE 136

Nisbett, R. E., & Wilson, T. D. (1977). Telli ng more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231-259. Parathian, A. R., & Taylor, F. (1993). Journal of Advanced Nursing, 18, 801-807. Park, C. L., Folkman, S., & Bostrom, A. (2001). Appraisals of controllability and coping in caregivers and HIV+ men: Test ing the goodness-of-fit hypothesis. Journal of Consulting and Clinical Psychology, 69, 481-488. Park, C. L., Armeli, S., & Tennen, H. (2004). A ppraisal-coping goodness of fit: A daily internet study. Personality and Social Psychology Bulletin, 30, 558-569. Parkes, K. R. (1986). Coping in stressful ep isodes: The role of individual differences, environmental factors, and situational characteristics. Journal of Personality and Social Psychology, 51, 1277-1292. Pennebaker, J. W. (1988). Confidi ng traumatic experiences and health. In S. Fisher & J. Reason (Eds.), Handbook of life stress, cognition and health (pp. 669-682). Oxford, England: John Wiley & Sons. Pennebaker J. W., Colder, M., & Sharp, L. K. (1990). Accelerating the coping process. Journal of Personality and Social Psychology, 58, 528-537. Pennebaker, J. W., & O'Heeron, R. C. (1984). Conf iding in others and illness rate among spouses of suicide and accidental-death victims. Journal of Abnormal Psychology, 93, 473-476. Pennebaker, J. W., Zech, E., & Rim, B. Disclo sing and sharing emotion: Psychological, social, and health consequences. In M. S. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut, Handbook of bereavement re search: Consequences, coping and care (pp. 517-543). Washington, DC: American Psychological Association. Penson, R. T., Partridge, R. A., Shah, Muhammad A., Giansiracusa, D., Chabner, B. A., & Lynch Jr., T. A., (2005). Fear of death. The Oncologist, 10, 160-169. Peteet, J. R., Abrams, H. E., Ross, D. M., & Sterns, N. M. (1991). Presenting a diagnosis of cancer: Patients views. Journal of Family Practice, 32, 577-581. Ptacek, J. T., & Ptacek, J. J. (2001). Patients pe rceptions of receiving bad news about cancer. Journal of Clinical Oncology, 19, 4160-4164. Ptacek, J. T., & Eberhardt, T. L. (1996). Br eaking bad news: A review of the literature. Journal of the American Medical Association, 276, 496-502. Quill, T. E. (1991). Bad news: Delivery, dialogue, and dilemmas. Archive of Internal Medicine, 151, 463-468. Rabow, M. W., & McPhee, S. J. (1999). Beyond breaking bad news: How to help patients who suffer. Western Journal of Medicine, 171, 260-263. 136

PAGE 137

Radziewicz, R., & Baile, W. F. (2001). Communication skills: Break ing bad news in the clinical setting. Oncology Nursing Forum, 28, 951-953. Ramirez, A. J., Graham, J., Richards, M. A., Cu ll, A., Gregory, W. M., Leaning, M. S., Snashall, D. C., & Timothy, A. R. (1995). Burnout and psychiatric disorder among cancer clinicians. British Journal of Cancer, 71, 1263-1269. Randall, T. C., & Wearn, A. M. (2005). Receiving bad news: Patients with haematological cancer reflect upon their experience. Palliative Medicine, 19, 594-601. Reed, J. M., Kemeny, M. E., Taylor, S. E., Wang, H. J., & Visscher, B. R. (1994). Realistic acceptance as a predictor of decreased survival time in gay men with AIDS. Health Psychology, 13, 299-307. Reynolds, P. M., Sanson-Fisher, R. W., Poole, A. D., Harker, J., & Byrne, M. J. (1981). Cancer and communication: Informationgiving in an oncology clinic. British Medical Journal, 282, 1449-1451. Richman, L. S., Kubzansky, L., Maselko, J., Kawachi, I., Choo, P., & Bauer, M. (2005). Positive emotion and health: Going beyond the negative. Heath Psychology, 24, 422-429. Roberts, C. S., Cox, C. E., Reintgen, C. S., Ba ile, W. F., & Gibertini, M. (1994). Influence of physician communication on newly diagnosed breast cancer patients psychologic adjustment and decision-making. Cancer, 74, 336-341. Rogers, R. W. (1983). Cognitive and psychological processes in fear appeals and attitude change: A revised theory of protection motivation. In J. T. Cacioppo & R. E. Petty (Eds.), Social psychophysiology (pp. 153-176). New York: Guilford Press. Rosen, S., & Tesser, A. (1970). On reluctance to communicate undesirable information: The MUM effect. Sociometry, 33, 253-263. Rothbaum, F., Weisz, J. R., Snyder, S. S. ( 1982). Changing the world and changing the self: A two-process model of perceived control. Journal of Personality and Social Psychology, 42, 5-37. Sabbioni, M. E. (1997). Informing can cer patients: Whose truth matters? Annals of the New York Academy of Sciences, 809, 508-513. Sardell, A. N. & Trierweiler, S. J. (1993). Disclosing the cancer diagnosis: Procedures that influence patient hopefulness. Cancer, 72, 3355-3365. Schofield, P. E., Butow, P. N., Thompson, J. F., Tattersall, M. H. N., Beeney, L. J., & Dunn, S. M. (2003). Psychological responses of pa tients receiving a dia gnosis of cancer. Annals of Oncology, 14, 48-56. Searight, H. R., Gafford, J. (2005). Cultural diversity at the end of life: Issues and guidelines for family physicians. American Family Physician, 71, 515-522. 137

PAGE 138

Segerstrom, S., Taylor, S., Keme ny, M., & Fahey, J. (1998). Optimism is associated with mood, coping and immune change in response to stress. Journal of Personality and Social Psychology, 74(6), 1646-1655. Shepperd, J. A., Findley-Klein, C., Kwavnick, K. D., Walker, D., & Perez, S. (2000). Bracing for loss. Journal of Personality and Social Psychology, 78, 620-634. Shepperd, J. A., & McNulty, J. K. (2002). The affective consequen ces of expected and unexpected outcomes. Psychological Science, 13, 85-88. Shields, C. E. (1998). Giving patients bad news. Primary Care, 25, 381-390. Silliman, R. A., Dukes, K. A., Sullivan, L. M ., & Kaplan, S. H. (1998). Breast cancer care in older women: Sources of information, social support, and emotional health outcomes. Cancer, 83, 706-711. Simpson, C. (2004). When hope makes us vulnera ble: A discussion of the patient-healthcare provider interactions in the context of hope. Bioethics, 18, 428-447. Slovic, P., Fischoff, B., & Lichtenstein, S. (1982). Facts versus fears: Understanding perceived risk. In D. Kahneman, P., Slovic, & A. Tversky (Eds.), Judgment under uncertainty: Heuristics and biases (pp.463-489). New York: Cambridge University Press. Snyder, C. R. (1999). Coping: The psychology of what works New York: Oxford University Press. Spera, S. P., Buhrfeind, E. D., & Pennebaker, J. W. (1994). Expressive writing and coping with job loss. Academy of Management Journal, 37, 722-733. Taylor, S., Lichtman, R., & Wood, J. (1984). Attri butions, beliefs about co ntrol, and adjustment to breast cancer. Journal of Personality and Social Psychology 46(3), 489-502. Taylor, K. M., & Shepperd, J. A. (1998). Bracing fo r the worst: Severity, testing and feedback as moderators of the optimistic bias. Personality and Social Psychology Bulletin, 24, 915926. Terry, D. J. (1991). Coping resource s and situational appraisals as predictors of coping behavior. Personality and Individual Differences, 12, 1031-1047. Tse, C. Y., Chong, A., & Fok, S. Y. (2003). Breaking bad news: A Chinese perspective. Palliative Medicine, 17, 339-343. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185, 1124-1131. Ungar, L., Alperin, M., Amiel, G. E., Behari er, Z., & Reis, S. (2002). Breaking bad news: Structured training for fam ily medicine residents. Patient Education and Counseling, 48, 63-68. 138

PAGE 139

139 Van Dijk, W. W., & van der Pligt, J. (1997). The impact of probabi lity and magnitude of outcome on disappoint ment and elation. Organizational Behavior and Human Decision Processes, 69, 277-284. van Dijk, W., Zeelenberg, M., & van der Pligt, J. (1999). Not having what you want versus having what you do not want: The impact of t ype of negative outcome on the experience of disappointment and related emotions. Cognition & Emotion 13, 129-148. Ward, C. G. (1992). The die is cast: Te lling patients they are going to die. Journal of Burn Care and Rehabilitation, 13, 272-274. Wortman, C. B., & Silver, R. C. (1989). The myths of coping with loss. Journal of Consulting and Clinical Psychology, 57, 349-357. Yates, P. (1993). Toward a reconceptualization of hope for patients with a diagnosis of cancer. Journal of Advanced Nursing, 18, 701-706.

PAGE 140

BIOGRAPHICAL SKETCH Kate Sweeny was born in 1980 in Burlington, Ve rmont. In 2002 she received a Bachelor of Science degree in psychology from Furman Univer sity (Greenville, SC), graduating summa cum laude. She then began her graduate education in social psychology at the Un iversity of Florida, where she received her Master of Science degr ee in December 2003 and w ill receive her Doctor of Philosophy in 2008. Kate now heads to the University of California at Riverside as an assistant professor in health and social psychology. 140