1 PAIN EXPRESSION IN CHILDREN WITH AUTISM SPECTRUM DISORDER (ASD): A FOUNDATION FOR INSTRUMENT DEVELOPMENT By MELISSA DODD INGLESE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFULLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2008
2 2008 Melissa Dodd Inglese
3 To my parents, for their consta nt love and encouragement, and for instilling in me the importance of higher education; To my sister, who has the most generous heart, and teaches me how to live and love life; To my brother, whose own academic excellence challenges me daily, and inspired me to persevere these past years; And to my husband, for his unconditional love, and for steadfastly pointing me towards Christ when I need strength
4 ACKNOWLEDGMENTS I would lik e to express my sincere appreciation to my chair and mentor, Dr. Jennifer Elder, for her assiduous encouragement and support throughout this process. Her constant availability, guidance and insight allowed me to successfully navigate the complex world of research. I would also like to gratefully acknowledge Dr. Jo Snider, who saw potential in me before I realized it in myself, Dr. Ann Horgas, who provided me with an early opportunity to tangibly experience research, and Dr. Sheila Eyberg, w ho remained patient and always encouraging throughout this process. I wish to especially thank the many caregivers that participated in these studies. Their willingness to share valuable thoughts and descri be personal experiences enabled me to shed light on such an important issue for these childr en and families. Add itionally, I would like to specifically thank the autism researchers and c linicians who took time to offer their expert opinions in the second stage of this research. Th eir collective wisdom help ed me gain a better understanding of how to better se rve this population of children. Furthermore, this list would not be comp lete without acknowledging my own family. I imagine they wondered several times whenor ifthis day would come. Without their enthusiasm and belief in my ab ility to complete this research, I might have given up long ago. Thank you. I would like to thank my precious husband, who met, dated, and married me in the midst of this academic pursuit. He has never known me as anything but a doctoral student. You are my rock, friend, encourager, confidant, a nd the source of so mu ch joy in my life. Finally, I acknowledge my Lord and Savior, w ho daily enables me to see the world with compassion. His love for me drives the work I do and is the reason I desire to dedicate my life to serving others, as a nurs e, educator, and researcher.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................10 CHAP TER 1 INTRODUCTION..................................................................................................................12 Background and Significance of the Problem........................................................................ 13 Rationale and Need for This Work......................................................................................... 15 Theoretical Framework and Nursing Metaparadigm.............................................................. 17 The Development of Pain Theory................................................................................... 17 The Communications Model of Pain............................................................................... 18 Kings Interacting Systems Fram ework: Nursing Metaparadigm ...................................21 Dissertation Purpose........................................................................................................... ....22 Research Questions............................................................................................................. ....22 2 REVIEW OF THE LITERATURE........................................................................................ 25 Autism Spectrum Disorder (ASD).......................................................................................... 25 Prevalence........................................................................................................................26 Review of Possible Etiologies for ASD.......................................................................... 27 Genetic predisposition..............................................................................................28 Environmental influences......................................................................................... 29 Vaccines...................................................................................................................30 Psychological perspectives.......................................................................................30 Core Features Exhibited by Children with ASD............................................................. 37 Autism......................................................................................................................38 Aspergers disorder (AD)......................................................................................... 39 Pervasive developmental disorder-not otherwise specified (PDD-NOS)................ 41 Communication Impairment in ASD...................................................................................... 41 Overview of Communication Theory..............................................................................43 Language and Communication........................................................................................44 Language Development................................................................................................... 46 Innateness.................................................................................................................46 Developmental changes............................................................................................ 47 Processes..................................................................................................................47 Critical periods and continuity................................................................................. 48 Communication Impairment and Its Influence on Pain Expression in AS D...................50 Pain Assessment in Children.................................................................................................. 52
6 Pain Assessment in Vulnerable Pediatric Populations........................................................... 54 Cerebral Palsy and Neurological Impairment................................................................. 55 Neonates at Risk for Ne urological Impairm ent............................................................... 56 Children with Cognitive Impairment............................................................................... 57 Non-Communicating Children w ith Cognitive Impairm ent............................................ 58 Ethical Considerations............................................................................................................63 Vulnerable Populations................................................................................................... 63 The Study of Pain............................................................................................................65 Rationale and Need for Instrument Development.................................................................. 66 Summary.................................................................................................................................66 3 METHODS AND PROCEDURES........................................................................................ 68 Instrument Development........................................................................................................ 68 Study 1: Caregiver Perspectives of the Pain Experience in ASD ........................................... 68 Purpose............................................................................................................................69 Study Design................................................................................................................... 69 Participants: Recruitment, Consent, and Anonymity...................................................... 69 Measures/Instruments...................................................................................................... 70 Caregiver Survey on pain expres sion in children with ASD ................................... 70 Semi-structured interview guide to discuss pain in ASD ......................................... 70 Data Synthesis and Analysis........................................................................................... 71 Study 2: Expert Appraisal of Potential Pain Expressive Indicators in ASD .......................... 71 Purpose............................................................................................................................71 Participants: Expert Reviewers........................................................................................72 Measures/Instruments: Item List for Expert Review ....................................................... 72 Procedures and Data Analysis......................................................................................... 73 Establishing content validity.................................................................................... 73 Item-level content valid ity indexes (I-CVI) .............................................................74 Scale-level content validity index............................................................................ 76 Expert comments and suggestions........................................................................... 77 Summary.................................................................................................................................77 4 RESULTS...............................................................................................................................79 Study 1: Caregiver Survey Results......................................................................................... 79 Demographic Information............................................................................................... 79 Quantitative Results from Caregiver Surveys.................................................................80 Descriptive Information Provided by Caregivers ............................................................ 80 Study 2: Expert-Rated Relevance of Potential P ain Indicators..............................................82 Examination of Item-Level Content Validity Indexes.................................................... 82 Calculation of a Scale-Leve l Content Validity Index ......................................................84 Expert Comments and Suggestions................................................................................. 84 Summary.................................................................................................................................84
7 5 DISCUSSION.........................................................................................................................94 Pain Expression in ASD Compared to Typically Developing Children................................. 94 Determining When Children with ASD are in Pain............................................................... 96 Communication and Social Impairment s in ASD and Pain Expression .................................99 Items Determined to be Relevant to Pain Assessm ent in ASD............................................ 100 Primary Pain Indicators: Retained Items....................................................................... 101 Secondary Pain Indicators: Item Retention and Revision............................................. 103 Items Removed from the Item List Following Expert Review..................................... 106 Expert Comments and Suggestions............................................................................... 106 A Need for Individualized Pain Assessment in this Population....................................107 The Role of Direct Behavioral Obse rvation in Instrum ent Development..................... 108 Next Steps in this Instruments Development...............................................................109 Limitations.................................................................................................................... ........112 Survey Response Rate................................................................................................... 112 Quasi-Qualitative Design..............................................................................................113 Nursing Implications and Clinical Significance ................................................................... 114 Future Directions..................................................................................................................115 APPENDIX A STUDY 1: CAREGIVER SURVEY COVER LETTER EXPLAINING PARTICIPATION ................................................................................................................121 B STUDY 2: ITEM LIST FOR EXPERT REVIEW ............................................................... 122 C STUDY 2: ITEM LIST COVE R LE TTER EXPLAINING STUDY AND PARTICIPATION................................................................................................................126 LIST OF REFERENCES.............................................................................................................127 BIOGRAPHICAL SKETCH.......................................................................................................139
8 LIST OF TABLES Table page 3-1 Study 1: Caregiver survey and se m i-structured interview questions.................................78 4-1 Study 1: Quantitative results from caregiver surveys........................................................ 86 4-2 Caregiver survey (questi on 1) resu lts: Identification a nd description of frequently reported themes..................................................................................................................87 4-3 Caregiver survey (question 1) results : Iden tified themes and sample caregiver quotations...........................................................................................................................88 4-4 Pain indicators identified by caregivers of child ren with ASD ......................................... 90 4-5 Study 2: Calculated item-level conten t validity in dexes (I -CVI)* for each item reviewed by experts........................................................................................................... 91 5-1 Comparison of expert-identified pain indica tors f or use in ASD with pain indicators from three existing pain assessment tools........................................................................ 117 5-2 Study 2: Variability in expert-rated rele vance am ongst all items with an I-CVI = .67... 120
9 LIST OF FIGURES Figure page 1-1 The Communications Model of Pain................................................................................. 24
10 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 PAIN EXPRESSION IN CHIDLR EN WITH AUTISM SPECTRUM DISORDER (ASD): A FOUNDATION FOR INSTRUMENT DEVELOPMENT By Melissa Dodd Inglese August 2008 Chair: Jennifer Elder Major: Nursing Sciences Little is known about how children with autism spectrum disorder (ASD) experience and express pain. Communication im pairments frequently render children with ASD unable to reliably self-report pain, social impairments limit their understandi ng of interpersonal interaction (and possibly alter the social context surrounding the pa in experience), and pe rseverative and/or repetitive displays of behavior may influence th e pain indicators traditio nally associated with children who are physically hurting. Additionally reports of children with autism having a high tolerance for pain create confusion and generate misunderstandings regarding whether they are capable of feeling pain. The overall purpose of this research was to explore how children with ASD express pain, and determine whether it is feasible to devel op a pain assessment tool specifically for this population. In order to approach this global research question, a series of two studies were conducted to 1) explore caregiver beliefs about the pain experience in their children with ASD, 2) identify a number of potential pain indicators in this popula tion, and 3) determine which pain indicators are believed by expe rts to best indicate pain in children with ASD.
11 Findings revealed that pain is a serious concern among caregivers. Parents struggle to determine when their children are in pain. St udy 1 results indicated that children with ASD are capable of experiencing and expressing pain, however their pain expression is unique. Study 2 validated this uniqueness by identifying pain in dicators specific to ch ildren with ASD, and by demonstrating that indicators rele vant to other, similar populations of children are not considered to be relevant in this population. This research supports the initial belief that a pain assessment tool specifically for use in ASD is warranted. Future research will build upon these studies to continue development of a valid and reliable instru ment to assess pain in individuals with ASD. Accurate pain assessment is necessary in order to ensure that pain in this population does not go unrecognized, and thus untreated.
12 CHAPTER 1 INTRODUCTION In 1989, McCaffery and Beebe established what is now one of the m ost widely recognized definitions of pain. Their definition emphasizes the subjective nature of the pain experience, and proposes that pain is whatever the patient says it is and occurs whenever the patient says it does (p. 7). Given this well established definition, coupled with other nationally acknowledged definitions, the accepted gold standard for pain assessment and measurement is self-report. This measurement captures the subjective aspect s of the pain experience, and supports the prevailing belief that pain is a unique and individualized proce ss, heavily influenced by the perceptions of those experiencing it. This definition however, i nherently implies that the person experiencing pain can verbally communicate his or her subjective experi ence. The International Association for the Study of Pa in (IASP) defines pain as a s unpleasant sensory and emotional experience associated with actual or potential tissue damage, or described in terms of such damage (IASP, 2007, 17). This definition also emphasizes the subjective nature of pain, and supports the use of self-report, whic h solicits how an individual would describe his or her pain to others (Merskey, 1991). In 1996, Anand and Craig critiqued the most co mmonly cited pain definitions and stated that pain assessment must be designed to conform to the communicatio n capabilities of the suffering person (p. 3). They emph asized the responsibility of scientists and clinicians to adapt a working definition of pain that would incl ude certain vulnerable popul ations. Specifically, they referred to a) individuals unable to communicate verbally (i.e., infants, or perhaps children and adults with communication impairment), and b) individuals whose verbal self-report might be unreliable (i.e., younger children, and persons with cognitive impairment or dementia). Effective pain assessment and measurement in a variety of populati ons requires that one
13 understand how to identify and interpret specific nonverbal indicators of pain, especially when verbal pain expression is confounded by an underlying condition (Oberlander & Craig, 2002). This research focused on one particular vulnerable populationone that poses an extremely unique challenge to pain assessmentchildren diagnosed with an autism spectrum disorder (ASD). Communication impairments fre quently render children with ASD unable to reliably self-report pain, social impairments limit their understandi ng of interpersonal interaction (and possibly alter the social context surrounding the pa in experience), and unusual (perseverative and/or repetitiv e) displays of behavior may influence the pain indicators traditionally associated with children who are physically hurting. Additionally reports of children with autism having a high toleran ce for pain create confusion and generate misunderstandings regarding whether they are capable of feeling pain. If individuals, specifically healthcare providers, assume that chil dren with ASD are not capable of feeling pain as a result of their underlying diso rder, these children are at treme ndous risk for having their pain go unrecognized and untreated. At present, it is unclear how ch ildren on the autism spectrum cognitively perceive, apply m eaning to, and commonly express their physical discomfort. Background and Significance of the Problem Prior to 1989, the pain experi en ce in children was severely underestimated, if even acknowledged at all. The prevailing belief was th at young children did not feel pain in the same way as adults (Finley & McGrath, 1998, p. 1). It was also believed that if they do feel pain, their inability to remember the painful event negates the need for astute pain management. The seminal work of Anand (1989) not only established that children, infants, and neonates have fully functioning nervous systems capable of pro cessing pain, but pointed out that inadequately treated pain can have detrimental and permanent physiological effects. Anand emphasized the primitive biological function of pain, which is to alarm the body that something is wrong and
14 needs attention. Its presence from birth supports th e notion that pain is often vital to survival (Hadjistavropoulos, Craig, & Fuchs-Lacelle, 2004) These reports amended attitudes towards pediatric pain assessment and management. Rather than assume children do not feel pain unless overtly indicated, researchers and clinicians inve rted the paradigm to assume children always experience pain during procedures an d situations that would normally be painful to an adult. As previously pointed out however, assessing pain in children (especially infants, preverbal children, and noncommunicating an d/or cognitively impaired children) can be challenging, specifically when relying upon self-report. Duri ng the past several decades, emerging pediatric pain research has aimed to overcome several of these challenges. A number of tools are now available to assess pain in typically developing children, those capable and incapable of self-re port. Pediatric pain assessme nt tools generally rely on behavioral, physiologic, and/or f acial indicators to make infere nces about whether a child is hurting (Stevens, 1998). More recently developed instruments assess pain from a multidimensional perspective by including indi cators from each of the aforementioned categories. When self-report is available, it is solicited. If self-report is consistent with behavioral and physiologic indicators, the most comprehensive picture of a childs pain experience is obtained. For groups of children where self-report is unav ailable or unreliable, assessment tools must focus solely on behavioral, physiolo gic, and facial indicators of pain. Some of these children have traditionally been excluded from pain re searchsecondary to the challenges associated with assessing their painand include, for ex ample, children with cognitive impairment, noncommunicating children, children with cerebra l palsy, and children with other neuromuscular disorders that significantly limit their physical capabilities. Recent pain study initiatives have
15 resulted in a slightly increased understanding of how some children with cognitive and motor impairments experience and express pain. Severa l researchers have begu n to describe pain expression in children with cognitive and neurological impairment, identifying through caregiver interviews and surveys those behaviors most indicative of pain (Br eau, McGrath, Camfield, Rosmus, & Finley, 2000; Gilbert-MacLeod, Craig, Rocha, & Mathias, 2000; McGrath, Rosmus, Canfield, Campbell, & Hennigar, 1998; St allard, Williams, Velleman, Lenton, & McGrath, 2002). To date however, there have been minima l attempts to comprehensively articulate how children on the autism spectrum experience and expre ss pain. This dissertation research adds to the growing body of knowledge re garding the pain experience in vulnerable subpopulations of children; specifically, it a dds to what is known about pain ex pression amongst the vastly diverse group of children who have been diagnosed with ASD. Rationale and Need for This Work The notable gaps in th e literature indicate th at while some work reflects an attempt to understand how nonverbal children and children w ith cognitive impairment communicate pain, the pain experience in children on the autism spectrum has hardly been explored. This is problematic because, as previously mentioned, th e reported high tolerance for pain in children with ASD leads many to assume they are incap able of feeling pain. Nader, Oberlander, Chambers, and Craig (2004) published what is pe rhaps the first and only empirical investigation examining pain expression in children with au tism. They reported that children with autism display a significant behavioral reaction in response to a painfu l stimulus, and these findings are in sharp contrast to the prevailing beliefs of pain insensitivity described in the literature to date (p. 88). Interestingly, th ey also highlighted the challenges parents encounter when assessing pain in their children with autism. Parents were aske d to report whether they believed their children were sensitive to everyday painful events at home. When parental reports were compared to the
16 childs objective facial responses to an acute pain event, those children reacting most to the painful stimulus had been previously identified by their parents as having a high tolerance for pain. This discord led the authors to highlight the need for additional exploration into how caregivers perceive their childrens ability to feel pain. Additionally, they stated that research needs to identify the factors that might influence pain expression and parental report of pain in this population. What do caregiver s believe about whether their ch ildren can feel pain, and how do they assess pain in this population? Is it possible to develop a pain assessment tool specifically for use in this uni que group of individuals? Thes e were the foundational questions for this research. Streiner and Norman (1995) poi nt out in their in troduction to health measurement scales that most researchers tend to magnify the deficiencies of existing measures and underestimate the effort required to develop an adequate new m easure (p. 4). They subsequently state that perhaps the most common error committed by clini cal researchers is to dismiss existing scales too lightly, and embark on the de velopment of a new instrument with an unjustifiably optimistic and naive expectation that they can do better (p. 4). Despite the daunting and long-term nature of instrument development, a thorough review of the literature revealed that there is no existing psychometrically sound pain assessment tool ava ilable for use with children who have ASD. Relatively little is known about how this populati on experiences and expresses physical pain. An assessment tool that caregivers and health care professionals can use to determine when children with ASD are in pain is needed. This research established a foundation for futu re research into how children on the autism spectrum express pain. The follo wing discussion addresses seve ral identified gaps in the literature, and clarifies the di scordant messages circulating when it comes to whether these
17 children feel pain. Caregiver pe rceptions regarding th e pain experience in children with ASD were explored in a first study, while insight from autism experts was obtained and examined in a second study. Understanding pain expression in children on the autism spectrum is important for healthcare providersespecially nurseswho often in teract directly with children and families, and function frequently as advocates and educators. Ultimately, this research is important because in gaining a better understanding of how children w ith ASD express pain, families of newly diagnosed children can be well-informed. This info rmation has the potential to alleviate suffering in an often silent and isolated population. Theoretical Framework and Nursing Metaparadigm Theoretical and concept ual models inform th e research process by providing an organized foundation from which relational questions and hypotheses may be derived. These questions and hypotheses may be descriptive and e xploratory in nature, or they may aim to test a prescribed intervention. While some theoretical models ar e more abstract (metaparadigms), and speak to the philosophical assumptions and/or worldview of a particular disc ipline, those theories utilized to guide research (middle-range theories) must pr ovide clear definitions for the central concepts, and articulate the multiple relationships be tween concepts (Alligood & Tomey, 2002; Peterson, 2004). Concepts need to be linked to empirical indicators in order to provide a means for directly measuring a particular concept and its po stulated relationship to other concepts (Fawcett, 2005). The research process aims to test these th eoretical relationships an d either support, or refute them. The Development of Pain Theory A num ber of theories can house a research program focused on the study of pain. Pain theory has evolved rapidly duri ng the past 40 years, ignited by the introducti on of a groundbreaking model by Melzack and Wall (1965), which delineated a multidimensional pain
18 processing system for the first time. Their mode l also articulated the hypo thesized role of the brain (central processing) in pai n, and it is now realized that th e brain is a significant modulator in the pain experience. Most current pain theories emphasi ze the interplay between physiology, neurobiology, and psychology (cognition and emotion) in the perception and expression of pain. This research examined how children with ASD mi ght process, perceive and ultimately express pain to others. It focused primarily on determ ining how to identify whether a child with ASD is in pain. This inquiry needed to be theo retically guided by a mode l that addresses the relationships between concepts important to processing, perc eiving, expressing, observing, and measuring pain. Specifically, because of the impairments in social and communicative abilities common to children with ASD, a theory that accounts for how these capacities influence the perception and expression of pain is most appropriate for pain research in this population. The Communications Model of Pain Craig (1998) exam ined available pain models and concluded that cu rrent models ignored interpersonal processes whereby the pain experience is encoded in expressive behavior, which observers must decipher [italics added] in order to appreciate the s ubjective nature of pain (p. 104). He and others set forth to develop a m odel that would integrat e various aspects of communicating pain to others, and focused spec ifically on how observational strategies might allow a third party to decode pain behaviors in an individual who could not verbally self-report pain. The resulting Communications Model of Pain (CMOP; Hadjistavropoulos & Craig, 2002) was used to guide this research. Hadjistavropoulos and Craig (2002) depict the pain experien ce in an A -> B -> C model (Figure 1-1). The internal state (A, in figure 11) during a physically pain ful event is affected by intrapersonal and contextual influences. These influences include the affective and motivational components of pain, the biological nature of the brain and nervous system, and the cognitive
19 capacity to apply meaning and perceive pain. The experience is encoded (B, in Figure 1-1) and expressed through various voluntar y and involuntary behaviors reflecting the interpersonal component of the model. The higher mental pr ocessing pathway encompa sses ones ability to voluntarily communicate his or her pain, or to perhaps even hide th e fact that one is in pain. This represents the shared experience of pain with others, particularly with an observer. Coupled with the voluntary pain expressive signals shared w ith others are various i nvoluntary (or reflexive) indicators of pain. Finally, everything an obs erver knows about the pain experience of another individual must come from the individual experi encing pain. This process of decoding (C, in Figure 1-1) someone elses e xperience depends heavily on the clarity of the messages coming from the verbal and nonverbal programs. Add itionally, individuals bri ng their own experiences and perceptions with them when they attempt to decode, or interpret, another persons pain expressive behaviors and/or reports. The CMOP is a comprehensive model, appropriate for this inquiry into how children with ASD might experience and express pain. Its emphasis on inferring pain from verbal and nonverbal indicators, and inclusi on of both voluntary and involuntary responses to pain, made this framework useful for the conducted research. Theoretical models allo w researchers to test specific relationships between concepts. This requires a formal definition of each concept, and a way to empirically measure concepts. In this particular research, it was not yet determined how children with ASD might encode pa in in expressive behaviors. And thus, because this research involved conducting exploratory desc riptive studies, empirical indi cators were not derived from the current model. Rather, this research aimed to determine what specific empirical indicators would be appropriate for each concept described by the model, given the population of interest. For example, in many children with ASD, communi cation is impaired, and thus children cannot
20 verbally self-report pain. The CMOP emphasizes the importance of identifying and utilizing automatic processing and nonverbal indicators of pa in when self-report is not available. This research began to make it possible to attach specific empirical indicators (i.e., those items comprising a pain assessment tool for children with ASD) to the concept labeled nonverbal programs in the model. Other concepts explored in this research in clude observer bias, message clarity, and for those children with verbal ability, ver bal programs of pain expression. Caregiver perceptions regarding whether they believe their children can feel pain were solicited. These perceptions can now be examined in light of th is model, and the impact of how observer bias might influence a caregivers ability to decode pain expression in their children can be formally tested in future studies. Add itionally, message clarity was explor ed as caregivers described the challenges associated with asse ssing pain in their children. The CMOP is one of the first thorough attemp ts to emphasize that pain processing is not comprised solely of neural and sensory components. This model highlights not only the role of the individual experience, but al so the interpersonal experience. The model also allows one to consider how joint attention might influence the understanding and expression of pain in individuals. If children with ASD are capable of feeling physical pa in, their intrapersonal experience might be similar to a typical childs. It may never be possible to accurately depict how they perceive or apply meaning, because as a decoder, all information must be interpreted. Knowing, however, th at children with ASD have di fficulty jointly attending to a shared experience, one can postulate that atte nding to a painful even t might be impaired. Dawson et al. (2004) report that attention to the distress of ot hers is related to ones joint attentional skills. These findings indicate that th e social intersubjectivity of the pain experience
21 likely plays a more important role than previously considered in earlier models of pain (Deyo, Prkachin, & Mercer, 2004). If intersubjectivity is impaired in individuals with ASD, their sharing of a painful experience is also subj ect to impairment. The CMOP offers a creative framework which also accommodates deficits in social interaction and allows researchers to begin to theorize how social impa irment might influence pain e xpressionwhich is critically important when considering the pain experience in children with ASD. Given the challenges associated with assessing pain in children with ASD, the CMOP accommodates those modes of pain expression (i nvoluntary and nonverbal indicators) that will likely need to be relied upon for accurate pain assessment in this population. As the exploratory, descriptive studies were conducted, findings were examined in light of how this model depicts the pain experience in individuals with communication and/or social impairments. Kings Interacting Systems Framework: Nursing Metaparadigm Parse (2000), a well known nursing theorist, em phasizes that m iddle-range theories are hypothetical statements that can be tested by quantitative re search through the operationalization of terms (p. 91). She purports that these groups of hypothetical statements are without meaning until they are framed with in a conceptual framework. Some nursing theorists, when addressing the advancement of nursing science, believe that middle-range theories may be borrowed from other discipline s only if they are linked to nursing conceptual frameworks and philosophy (Fawcett, 2005; Pars e 2000; Villarruel et al., 2001). Thus, the particular pain model guiding this research was linked to a more abstract nursing metaparadigm; specifically, to the one developed by Imogene King (1981). The general concepts and underlying worldview of the CM OP align nicely with the philosophical and conceptual development of Kings (1981) systems framework, which emphasizes interaction, and the communicative nature of human beings as they relate to others
22 and their environment. Imogene Kings conceptu al model defines the major concepts pertinent to the discipline of nursing, such as those in the overall metaparadigm of nursing: human beings, environment, health, and nursing. She delineates her philosophy regarding how these abstract concepts relate to one another. Her conceptual framework derives from systems theory, and focuses on three specific interacting systems: pers onal systems, interpersonal systems, and social systems. The CMOP approaches pain from a framework that assumes individu als bring their entire being to a situation, and that they interact with others, conveyi ng communicative signals (i.e., in this study, expressing pain via verbal and nonverbal modalities) for interpretation by others. The CMOP emphasizes the importance of the environm ent, communication, and so cial factors in the experience, expression, and obser vation/interpretation of pain, both of which are described abstractly by King (1981). The metaparadigm developed by King reflects the overall nursing approach that was applied when conducting this research. The CMOP however, more specifically defines the concepts pertinent to this research, and will ultimately allow for specific testing of relationships between concepts. Dissertation Purpose The overall purpose of this research was to explore how children with ASD express pain, and determ ine whether it is feasible to devel op a pain assessment tool specifically for this population. In order to approach this global research question, a series of two studies were conducted to address several mo re specific questions. Research Questions Two exploratory descriptive studies addr essed the following research questions: 1. What do caregivers of children with autis m spectrum disorder (A SD) believe about the way their children experien ce and express pain?
23 2. Do caregivers find it difficult to determine when their children with ASD are in pain? Further, if they can determine when their child is in pain, what specific indicators are utilized in pain assessment? 3. Do caregivers of children with ASD believe healthcare providers adequately assess and treat their childrens pain? 4. How do autism experts describe the nature of the pain experience in children with ASD? 5. What specific indicators do autism and pain e xperts believe are relevant to pain assessment in children with ASD? 6. Is it feasible to develop a pain assessment tool for us e in the ASD population?
24 Figure 1-1. The Communications Model of Pain. (Hadjistavropoulos & Craig, 2002) A Internal Experience Higher Mental Processing Automatic Processing C Decoding Pain Stimulus B Encoding in Expressive Behavior Intrapersonal Influences Contextual Influences Message Clarit y Message Clarit y Automatic Processing Higher Mental Processing Verbal Pro gr ams N onverbal Pro g rams Observer Bias
25 CHAPTER 2 REVIEW OF THE LITERATURE In order to appreciate the challenges encountered when attempting to determ ine whether a child with autism spectrum disorder (ASD) is in pain, one must be familiar with the specific nature of the disorder. The purpose of the followi ng review is to a) prov ide an overview of ASD, by describing current positions regarding prevalence and proposed etiologies, and by examining the specific strengths and wea knesses associated with children on the autism spectrum, b) explore communication theory, par ticularly as it explains the impairments in ASD, and focus specifically on the implications of this deficit s influence on pain expression, c) present a brief discussion of pain expression in normal children, d) review research focused on the development of pain assessment tools for use in several re lated vulnerable populations of children, e) explore the applicability of using these existing pain a ssessment tools in ASD, and f) discuss ethical issues pertinent to this research. Autism Spectrum Disorder (ASD) In 1943, an original paper by Leo Kanner de scribed the presentation of children who would later be labeled as autistic In his words, There is fr om the start an extreme autistic aloneness that, whenever possible, disregards, igno res, shuts out anything that comes to the child from the outside (Frith, 2003, p. 6). This aloneness emerges from a fundamental impairment in social interaction. It is this social impairment, along with impairments in communication, and the presence of restricted pa tterns of behaviors an d interests that ch aracterize a group of disorders known as autism spectrum disorder s (ASDs; Charman & Baird, 2002; Dover & Le Couteur, 2007; National Institute of Mental H ealth [NIMH], 2007; Walker et al., 2004). ASDs are sometimes also referred to as pervasive developmental disorders (PDDs), and the PDDs include autistic disorder, Aspe rgers disorder, childhood disinteg rative disorder (CDD), Retts
26 disorder, and PDD-not otherwise specified (PDD-NOS; American Psychiatric Association [ DSM-IV-TR ], 2000; Volkmar & Klin, 2005). In clinical practice, Retts disorder and CDD are sometimes not included in the grouping of ASD, and therefore some sources may describe ASD as including autism, AD, and PDD-NOS (Kutscher, 2006). For the purpose of this review, ASD will refer to autism, AD, and PDD-NOS. These disorders are collectively described as a spectrum of disorders because the symptoms can occur in a variety of combinations and present with varyin g degrees of severity (Autism Society of America [ASA], 2006). The emergence of language describing a spectrum of disorders highlights the possible interrelatedn ess of these diagnoses and accommodates the nature of the variability in symptom severity of ten seen in children on the spectrum (Filipek et al., 1999). More recent atten tion to the category of PDDNOS has contributed to the conceptualization of a broader phenotype of ASD, as this category includes children who display various patterns of skills and difficultie s related to, but not adequately meeting the restrictive criteria for other pervasive developmen tal disorders such as Aspergers disorder, or autism (Dawson et al., 2002; Filipe k et al., 1999; Volkmar, Lord, Bailey, Schultz, & Klin, 2004). Despite the challenges encountered by clinicians regarding classification, all children with ASD have difficulties in the three previously described domains: social interact ion, patterns of communication, and restricted and/or repetitive behaviors or interests. Prevalence It was assum ed for many years that autism was rare, occurring at a rate of about four to five cases per 10,000 children. Prevalence report s in the late 80s and early 90s however, indicated that the rate of ASDs might be somewhere around 30 per 10,000 children, with about one quarter of those (10 per 10,000) meeti ng criteria for true autism (Filipek et al., 1999; Fombonne, 2003, 2005; Volkmar, Chawarska, & Klin, 2005). This alarming rise in the
27 reported prevalence compelled many to explore why the number of children with ASD seemed to be increasing, and sparked debate about whethe r there is an autism epidemic (Barbaresi, Katusic, Colligan, Weaver, & Jacobsen, 2005; Barclay, 2005; Fombonne, 2001; Gernsbacher, Dawson, & Goldsmith, 2005; Newschaffer, Fal b, & Gurney, 2005; Williams, Higgins, Brayne, 2006). In response to the ci rculating concerns regarding the true prevalence of autism and related disorders, the Centers for Disease Control (CDC) conducted a study examining 8-year-old children living in 14 sites in the United States (C enters for Disease Control [CDC], 2007). Their findings state that 1 in 150 childre n are living with an ASD. The rate of autism in boys is about four times higher than in girls, with a reported ratio of 4.3:1 (Fombonne, 2005). According to the CDC, using this current prevalence data, we can estimate that if 4 million children are born in the United States ev ery year, approximately 24,000 of these children will eventually be diagnosed with an ASD (CDC, 2007, 2). In addition, the CDC points out that using this prevalence rate, if it has been constant over the past two decades, we can estimate that up to 500,000 individuals between the ages of birth and 21 have an ASD (CDC, 2007). After mental retardation (intellectual impairme nt), ASDs are the most common developmental disability (CDC, 2007). These st atistics are widely adopted and currently reported throughout the United States by organizations and experts fo cused on autism. The prevalence of ASD is currently higher than that of spina bifida, can cer, or Downs syndrome (Filipek et al., 1999; Muhle, Trentacoste, & Rapin, 2004; NIMH, 2007) Inadequate pain assessment in this population has national implications and potentially neglects the needs of a large population of children. Review of Possible Etiologies for ASD Unlike other m edical conditions where a precise causal mechanism can usually be delineated, the etiology of ASD continues to perplex many experts, clinicians, and researchers in
28 the field. It is believed that autism is a neurobehavioral disord er and has no single cause (ASA, 2006; Happe, Ronald, & Plomin, 2006; Sunil, 2006) Growing evidence supports the idea that ASD may be caused by several factors, includi ng genetic susceptibili ty and environmental influences. The precise nature of the interpla y between genetic predisposition and environmental factors is the targeted question of many current research initiatives. The heterogeneity of the phenotype of ASD highlights the complexity scien tists encounter as they examine the proposed etiologies of ASD. There is emphasis on the impor tance of integrating findings from a variety of disciplines attempting to understa nd why children with ASD present with such varying degrees of difficulty in the three primary domains. The following sections will review current research initiatives exploring possible infectious, gastro intestinal, nutritional, environmental toxin, and inflammatory, metabolic, and neuroendocrine factors and ASD. Additionally, a few psychological perspectives are briefly presented. Genetic predisposition Fa mily studies have reported that there is a 60% concordance for classic autism in monozygotic twins versus 0% in dizygotic tw ins (Muhle et al., 2004; Sunil, 2006). When a broader phenotype for autism was used, the monoz ygotic concordance incr eased to 92%. These findings point to a strong genetic base for autism, but also highlight the fact that there are likely environmental or nongenetic factors influenci ng the expressionand t hus the severityof autism traits (Freitag, 2007; Muhle et al., 2004; Sunil, 2006; Szpi r, 2006; Zafeiriou, Ververi, & Vargiami, 2007). While most experts agree ther e is a strong genetic component to ASD, the more controversial discussions involve the nongenetic, e nvironmental factors that may influence the phenotypic expression of ASD. No particular environmental fact ors have been scientifically proven to trigger ASD, however current ongoi ng studies are examining the possible gene-
29 environment interactions that may lead to th e development of ASD, and will hopefully provide more concrete evidence for particular culprits (Szpir, 2006). Environmental influences Several s tudies examining parental beliefs about the causes of autism demonstrate that there are a number of environmental factors cons idered to influence the development of ASD. Most commonly cited factors in clude immunizations (child a nd maternal immunization during pregnancy), environmental exposure (to infectio n, medications, toxins), intolerance to food (primarily those containing casein and gluten), and specific perinatal events (fetal distress or anoxia, prematurity, low birth we ight, uterine bleeding or indu ced labor; Harrington, Patrick, Edwards, & Brand, 2006; Mercer, Creighton, Hold en, & Lewis, 2006). In addition to those factors commonly mentioned by parents, other prenat al, perinatal, and post natal factors are under investigation (Kolevzon, Gross, & Reichenberg, 2007). It is not clear how, or whether, heavy metal exposure (mercury, cadmium, lead) might c ontribute to ASD. It has been hypothesized that children with ASD have a diminished ability to detoxify, and theref ore exposure to metals results in increased circulating levels thr oughout the body, and brain (Zafeiriou et al., 2007). Studies now underway are also i nvestigating the relationship betw een factors such as atypical placental growth, abnormal gut tissue, inflamed ti ssue in the brain, maternal and paternal age, and food allergies (Kolevzon et al., 2007; Neimark, 2007). What was once considered a primary disorder of the brain is now being conceptualized by some experts as a disorder whereby the interaction between genetic predis position and environmen tal factors leads to a change in cellular function. This then manifests itself throughout the body, and presents as an array of biological, neurological, and psychological abnormalities which have come to represent the autism spectrum disorders (Herbert, 2005; Neimark, 2007).
30 Vaccines Issues involving vaccination con tinue to be a controversial to pic. There are two separate, but related, concerns regarding immunizations. The fist involves th e MMR vaccine and the belief that administration of this vaccine causes ASD in some genetically predisposed children. In 1998, a report by Wakefield et al. described 12 children who presented with ASD and comorbid bowel symptoms. He contended that children with regressive autism and bowel symptoms reflected a unique pr ocess linked to the MMR vaccine (Brown, Berkovic, & Scheffer, 2007; Francois et al., 2005; S ilverman & Brosco, 2007; Taylor, 2006). This proposition drew enormous attention from parents and clinicians. Despite criticism regarding the study design, the retraction of this interpretation by almost every co-author of the paper, rigorous follow-up studies, and the recent reports from the Institute of Me dicine (IOM) and the CDC that there is no causal relationship between the MMR vaccine and AS D, this issue remains at the forefront of many discussions and continues to be on the au tism research agenda (CDC, 1999, 2000; Institute of Medicine [IOM], 2001; Mu rch et al., 2004; Silverman & Brosco, 2007; Taylor, 2006). The second hotly debated con cern related to vaccination involves the sheer number of vaccines children receive, and th e possible relation of increased immunizations to the perceived increase in autism prevalence (Silverman & Brosco, 2007). Attenti on to the preservative thimerosal (which contains ethylmercury) in vaccines, has been postulated to cause neurotoxicity, and thus ASD. Removal of th imerosal, and subsequent research (IOM, 2004) does not reveal any link between immunizations and developmental disabilities, however it is acknowledged that additional research into this issue is imperative. Psychological perspectives During the first d ecades following Kanners de scription of infantile autism, there was tremendous focus on the parental role in th e psychogenesis of autism. It was wrongly believed
31 that autism was a psychiatric diso rder resulting when parents did not provide adequate love, or failed to create a supportive environment for their children (Volkmar & Klin, 2005). Early treatment therefore, involved working with children and their parents (Silverman & Brosco, 2007). This theoretical framework was challe nged in the 1960s by Bernard Rimland (1964), who more accurately alluded to th e genetic component and neurolog ical nature of the disorder (Rimland, 1964; Silverman & Brosco, 2007). Investigation into the etiology of ASD has now crossed a myriad of disciplines including those that take a more biomedical and physio logical approach. The historical roots in psychology and psychiatry however, have resulted in a long tradition of successf ul research into ASD. Much of what is currently realized a bout the cognitive nature of ASD derives from psychological literature, and ma ny effective treatment protocol s are rooted in cognitive and behavioral therapy (i.e., Applied Behavioral Analysis (ABA)). Similar to the debated issues encountered in neurobiology, developmental psyc hology houses a variety of theoretical positions on the etiology of the core features found in individuals on the autism spectrum. Three commonly referenced theories include Weak Central Coherence (Frith, 1989) Executive Dysfunction (Ozonoff, 1995), and Mindblindedness (Baron-Cohen, 1995). Weak central coherence. Central coherence refers to ones ability to process information in its context; to create a highe r-level meaning from a variety of sources of input (Hill & Frith, 2003). Weak central coherence would manifest itself in a display of behaviors where an individual focuses primarily on pieces of information. Several behavi ors displayed in ASD support the notion that they have weak central coherence. For example, tests involving homographs (words with more than one definition i.e., read or present) reveal that normal children pronounce words dependent on sentence c ontext, while verbal children with ASD do not
32 attend to the context of the sentence (Hill & Frith, 2003). Francesca Happe (1996) has extended these sentence findings and shows that struggles with weak coherence are manifested in lowlevel visual tasks as well. She poses that there is a bell-shaped distribution of coherence ability within the population of individu als with ASD, and that this might account for some of the variability seen regarding this cognitive ability. The weak central cohere nce theory is unique in that it not only explains several of the impairment s seen in individuals with ASD, but it also has explanatory power regarding the unusual streng ths present in this population. The embedded figures test (where a child must pick out a small piece of a global picture) is a test where children with ASD have been shown to excel. They ofte n perform better than typi cal individuals (Hill & Frith, 2003). Researchers hypothesize that a child on the spectrum is not distracted by the overall picture, and thus succeeds rather easily in these types of tasks. A number of tests and tasks have been developed to explore central cohe rence abilities in childre n with ASD. There is merit to this hypothesis in that it has the ability to explain some of the deficits and strengths witnessed in this population. Regarding the pain process, experts have id entified a number of ps ychological influences that may worsen, or minimize, a normal individuals pain experience. Part of interpreting a pain stimulus involves attributing higher-level meaning to the situation. If an individual is afraid, anxious, sees blood, or anticipates something is wrong, his or her percepti on of the experience might worsen. If children with autism have w eak central coherence, their attention to one specific perceptual process wit hout the added contextual information, might influence what is communicated to, or decoded by, a nearby observer. In fact, it might diminish the childs interpretation of the painful stimulus, and subsequently minimize what is communicated to, and thus interpreted by, others.
33 Executive dysfunction. Weak central coherence seems a plausible explanation for several of the behaviors displayed by individuals with ASD. It does not, however, account for all of them. Nor is there a clear link between weak central coherence and other proposed causal theories. Similarly, executive dysfunction has explanatory power for several other behaviors witnessed in individual s in this population. Executive function refers to th e ability to perform such f unctions as planning, working memory, impulse control, shifting set, and th e initiation and monitori ng of action. (Hill & Frith, 2003, p. 285). The investigation of executi ve function in young children with ASD has historically been limited by the minimal number of ta sks available to test these cognitive skills at such a young developmental age. Recent attempts have been made to increase researchers abilities to examine executive function in young children (Hughes, 1998). Tasks appropriate for older children, such as the Towe r of Hanoi (measuring planning ab ilities), the detour reaching task (measuring inhibition of prepotent res ponse), and the Wisconsin card sorting task (measuring mental flexibility) reve al that individuals with ASD display marked disabilities in these areas (Hughes, 1998; Ozonoff & Jensen, 1999). The inability to plan accurately or resist internal impulses, and the me ntal rigidity evident in AS D supports the postulation that impairment in executive functioning might explai n the behaviors seen in this population. The severe resistance to change and repetitiveness of actions both also appear to be dysregulations in executive functioning. Mindblindedness. Mindblindedness, or more commonly referred to as the lack of a theory of mind, is the theory most developed in the autism literature (Baron-Cohen, 1995; BaronCohen, Tager-Flusber, & Cohen, 1993). Its wide acceptance is related to mounds of empirical work supporting the main assumption that individuals with ASD lack a theory of mind. The
34 most significant deficits in chil dren with ASD involve social im pairment. Frequently, abnormal social interaction is the hallmar k characteristic that initiates a suspicion of autism (second to abnormal social interaction is language delay). Parents notice isolated be haviors early in their childs development. Thus, work has been done to explore the root of this social impairment, and to identify the aberrant cognitive structure responsible fo r the displayed behaviors. Theory of mind refers to the ability to unde rstand the mental states of others, and to appreciate the relationship between mental states and actions. Th eory of mind has traditionally been measured by false belief tests. Here, the pr esence of a theory of mind is reflected in an individuals ability to monitor a scenario and separate ones own thinking from the thought process of another, when both subjects have re ceived different information. For example, a box of crayons is presented to a child. Next, the box is opened to reveal that inside the box is, not crayons, but popcorn. The box is closed. Then the child is asked what Buddy (who has not seen inside the box) will think is inside. Individuals with autism commonly fail this type of task. They fail to realize that Buddy, who has not se en inside the box, cannot possibly already know that it does not contain crayons. Empirical work continues to support the exploration of theory of mind abilities in children with ASD, and yet many questions remain. The absence of a theory of mind is hypothesized to explain many of the so cial and communicative ch aracteristics seen in ASD. Because traditional theory of mind tasks eval uate the presence of theory of mind around four years old, researchers have sought to id entify cognitive structures which may act as precursors to theory of mind. Uta Frith (2003) describes theory of mind using the term mentalizing. She poses that the brain has an innate mechanism for processing other peoples mental states (p. 95). If mentaliz ing is innate, precursors to the th eory of mind measured at four
35 years old might be identifiable. Suggested precu rsors include pretend play (metarepresentation), and joint attention (reflected in eye contact and gaze monitoring) (Charman, 2003; Morgan, Maybery, & Durkin, 2003). Joint attention is posited to influence the development of a theory of mind, and also the emergence of language (C harman, 2003), which may indicate that joint attention represents a more primary cognitive stru cture. Its impairment results in an absent theory of mind. Joint attention behaviors are id entifiable between six and twelve months of age. They involve dyadic or triadic sharing of attention between the infa nt, another person, and an object or event (Charman, 2003). Joint attent ion abilities have been empiri cally associated with language and social development. While associations ha ve been identified, less is understood about the exact neurodevelopmental mechanisms that might e xplain how deficits in joint attention evolve into theory of mind and language impairments. R ecall also, that early does not always indicate cognitive structures are innate. Joint attention may be a direct precursor to theory of mind development, but it is not well understood how join t attention develops, wh ether directly out of eye contact, or by some othe r unidentified mechanism. Deficits in these proposed precu rsors to theory of mind (joint attention and eye contact) might influence the development of appropriate pain understanding and expression in children with ASD. The following illustration adds clarity to this proposition. When a normal young child falls downassuming that the tumble was painful for a brief momentmost often the child glances upward in search of a bystander. Once identified, the child jointly attends to the recent event (the fall). The childs reaction is heavily influenced by the other persons reaction to his or her fall. Fear in the face of the mother might instigate a cry. The childs own embarrassment might result in crying. A smile on th e face of the mother, or the offer of comfort
36 might diminish the desire to cry. Crying is an overt indicator of pain to many caregivers. The cry in this situation however, reflects an intr icate system of psychological processing. The psychological processing is blunted if the child fails to jointly attend to th e event of falling down. There may be no communicative interaction at all, and this may subsequently be misinterpreted that the child felt no pain during the fall. Only when it becomes unbelieva ble that a child could feel no pain (i.e., discovering the child has a br oken limb) might a care giver become overtly concerned. The exploration of this notion was th e premise for this research, and it is now clear that there may be a relationship between the so cial and communication impairments in children with ASD, and their ability to express whether they are in pain. Interrelatedness of psychological theories. Initially, each of th ese three described psychological theories originated to explain a single causal mechanism responsible for the clinical picture of ASD. However, it is becoming more apparent that no one theory can account for all phenomena seen in individuals with ASD. No single theory has successfully explained causality. Theory of mind approaches are limited in their ability to simultaneously explain executive dysfunction. Hughes (1998) asserts that execu tive functioning is crucial and primary to a childs development of a theory of mind. Ch allenging this contention, is the reminder that there are other clinical subpopul ations who suffer from executive dysfunction, and who do not have difficulty understanding others mental states (for example, schizophrenia, attention deficit hyperactivity disorder [ADHD], and frontal lobe syndrome) (Baron-Cohen & Swettenham, 1997). Several attempts have been made to integrate weak central coherence with the absence of a theory of mind. Jarrold (2000) asserts that weak central cohere nce is crucial and primary to a childs development of a theory of mind. Mo rgan, Maybery, and Durkin (2003) tested this
37 relationship, and conclude that weak central coherence, joint atten tion, and verbal ability independently discriminate children with auti sm. Weak central coherence alone could not predict autism group membership. The debate remains unresolved. There are thr ee well-established theoretical approaches in psychology, with researchers from each position i nvested in uncovering the causal mechanisms of autism. Newer ideas aim to incorporate th e established theories, and seek to unify the different approaches. Future exploration might reveal explanations for the variability in symptom type and severity seen across the spectrum of autism disorders. In this light, it might be possible to explain why a variety of contradictory pain experiences are reported within this population. Clinicians, parents, and experts in autism have reason to be encouraged by the vast number of research initiatives aimed at understanding the cause(s) of AS D. Understanding the various theoretical positions regarding etiology allows one to consider how pain processing might be disrupted in this population. Core Features Exhibited by Children with ASD All autism spectrum disorders are characterized by three defining features: impairments in socialization, limitations in ve rbal and nonverbal communication, a nd the presence of restricted and repetitive behavior s and/or interests (DSM-IV-TR 2000; NIMH, 2007). Currently, general practitioners in clinic al practice utilize the Diagnostic and Statistica l Manual of Mental Disorders (DSM-IV-TR 2000) and the 10th edition of the International Classification of Diseases ( ICD-10; World Health Organization [WHO ], 1992) diagnostic criteria as formal guidelines for diagnosis of autism, Aspergers disorder, and ot her spectrum disorders not meeting full criteria for autism. The current criteria in the DSM-IV-TR and the ICD-10 classification systems allow for extreme latitude in interpretation on the pa rt of the clinician (V olkmar & Klin, 2005).
38 Despite diagnostic limitations, it is important to discuss different manifestations of the core features and deficits that are associated with each of the autism spectrum disorders. Autism Typically developing infants are social beings (NIMH, 2007). Gazing at others, attuning to voices, holding fingers, and sm iling ar e natural social behaviors in infancy. Social impairments in autism therefore, often manifest themselves in poor eye contact, and the inability to utilize nonverbal gestures to manipulate the social envir onment, such as pointing or sharing gaze with others ( DSM-IV-TR, 2000; Dover & Le Couteur, 2007; NIMH, 2007; Volkmar & Klin, 2005). Younger children and infants do not imitate or play the same way as typically developing children. They often lack pretend play, and might s how little interest in things. As they grow, children with autism often do not develop peer re lationships appropriately and do not appear to respond emotionally to others. Ho wlin (1998) summarizes in detail the deficits seen in autism and discusses how older children lack social aw areness and thus may seem isolated from the world. They appear to not have feelings, and do not always seem sensitive to others. They may also act socially inappropriate at times. For example, verbal children with autism may ask an obese woman why she is fat. Communication impairments in autism manifest as either a delay in, or complete lack of, the development of spoken language. In individu als with verbal abilities conversation skills are limited, and they may demonstrat e repetitive use of language, or idiosyncratic language use ( DSM-IV-TR, 2000; Volkmar & Klin, 2005). Children with autism often repeat things or ask the same question over and over again. Parents of younger children ma y initially report a lack of babbling, or odd speech patterns in cluding echolalia (repeating or echoing others), and unusual tone or pitch. Many children with autism, if th ey do acquire some basic language abilities, have
39 difficulty telling others what they want or need, and this can lead to frustration, anxiety, and even tantrum-like behavior. Restricted patterns of behavior interests, and activities co mprise the third category of impairment in children with autism. Child ren with autism might be preoccupied with a particular interest that is of a bnormal intensity or focus. They often demonstrate repetitive play, such as in excessively lining up cars. Many ch ildren with autism have motor mannerisms, such as hand flapping, which are repetitive in nature. Ch ange is often a stressful occurrence for them. They may have routines or patterns fr om which they cannot deviate. Deficits associated with true autism permeat e all aspects of the lives of the child and family members. Special consideration needs to be taken when working with these children and families, especially in healthcare settings, in order to be sensitive to the way children with autism experience the world. While the specifications fo r the diagnosis of autism have been delineated and widely accepted, there is controversy about how to accommodate all children who have a spectrum disorder, but do not meet the criteria for true autism. These children include those with Aspergers disorder, and PDD-NOS. Aspergers disorder (AD) The characterization of Aspergers disorder (A D) continues to create confusion and controversy among the autism community. While related to autism and PDD-NOS diagnostically (it is included in the DSM-IV-TR and its diagnostic criteria are identical to autism, with the exclusion of the communication criteria), the definition of AD is argued to be unique (Filipek et al., 1999). Some have referred to AD as autism without mental retardation, high functioning autism (HFA), or milder forms of autism marked by higher cognitive and linguistic abilities (Klin, McPartlan d, & Volkmar, 2005, p. 88). Despite the DSM-IV-TR diagnostic implications that there are no language impairments in AD, it is clear that language in
40 children with AD disorder is not typical. Examples of current research include attempts to define whether there are distinctions between AD and HFA, and the consideration that perhaps a broader phenotype of AD might in clude relatively successful indivi duals who are not necessarily disabled (Klin et al., 2005). Children with AD sh are many of the same features as children with autism, but do not have a history of language delay and usually have aver age or above average intellectual abilities (Klin et al., 2005; Pratt & Buckmann, 2006; Sunil, 2006). Socially, children with AD may not appear as withdrawn as children with autism, but they tend to approach others in inappropriate or eccen tric ways (Klin et al., 2005). Children with AD do befriend others, but inherent in their friendships are difficu lties related to awkwardness and perceived insensitivity on the part of others. They participate in conversation, but have a tendency to only discuss topics they are interested in and fail to banter back and forth as in typical two-way conversation. Their inability to understand the rules of interaction, and poor comprehension of jokes and metaphor, can lead to feelings of isolation and embarrassment. Children with AD may have flat and emotionl ess speech. They are often obsessed with particular topics, may ask repetitive questions, and display c oncrete, literal thinking. Their circumscribed interests lead them to learn volume s of information about a particular topic in a very intense nature. For example, a child might know or want to learn absolutely everything about vacuums, from how they are built, the hist ory of when they were invented, down to the make and model of every vacuum in existen ce. Some children with AD display eccentric behaviors (Belschner, 2007). Children with AD usually have a history of poor motor coordination. Parents often describe them as clumsy. The intentional focus on AD and its relation to the other spectrum disorders has led to a revitalization of research examin ing this particular population. Th e heightened attention has led
41 to discussion about definition, causality, detailing its relationshi p to other spectrum disorders, and targeting interventions specifically de signed to work w ith these children. Pervasive developmental disorder-n ot otherw ise specified (PDD-NOS) PDD-NOS, also referred to as atypical autism, is referenced when there are clinically significant autistic symptoms, but not quite enough to meet the full criteria for true autism. It is not a separate disorder with di fferent manifestations, and ther efore, many of the previously described manifestations are applicable to this population. Children with PDD-NOS are sometimes considered to demonstrate a mild er symptomatology. Diagnosing PDD-NOS is a diagnosis by exclusion, when a child cannot be ca tegorized into one of th e other autism spectrum diagnoses. One report describes children with PDD-NOS as being midway between the autism and AD groups on IQ, measures of adaptive behavi or, and language milestones (Walker et al., 2004, p. 178). They are on the spectrum, are hi gher functioning than autism, have fewer repetitive behaviors than autism and AD, and have more delayed language than children with AD (Walker et al., 2004). Communication Impairment in ASD From a very young age children learn that they live in an interactive world. The world stimulates them, it responds to them, and they quic kly realize how to inter act with the world in order to elicit those responses from others. These basic realizati ons foster a sense of connectedness to the environment. To some who study the nature and development of communication, these early infant realizations are presumed to lay the foundation for the emergence of the ability to communicate. So mething goes awry, howev er, in children with ASD, rendering them unable to communicate in the same manner as other children. Impairments in communication are the focus of many different studies found throughout the autism literature, and have the potential to significantly alter and/or limit pain expression. If
42 children with autism are known to have social and communicative impairments, it is unlikely thataccording to The Communications Mode l of Pain (Hadjistravopoulos & Craig, 2002) their pain experience would go uninfluenced. Theoretically, th ese children have inherent difficulties when it comes to expressing pai n. In order to thor oughly understand how the impairments seen in children with ASD might influence their understanding and expression of physical pain (bumps, bruises, burns, etc.), it is necessary to examine communication, and specifically, the development of th e ability to communicate in ASD. This section aims to describe several cu rrent theoretical perspectives found in the communication and language literature, in an effo rt to begin to understand how impairments in this specific domain might influence a childs ability to express pain. Several of the longstanding debatable issues that are commonly addr essed in the language development literature are presented. Theories that attempt to explain the impairments seen in children with autism are explored. Finally, a more in-depth examin ation of specific communication and language impairments seen in children with ASD, as th ey might influence pain expression, is presented. The pain experience is a complex, multidimensional process involving different components of the brain and centr al nervous system. Similarly, pain expression is a complex, multidimensional process. Rather than cons ider how children with autism may uniquely experience pain, some individuals assume that the impairments inherent in children with ASD actually diminish or negate pain processing altogether. This is perhaps a result of the fact that many children with ASD do not express their physic al discomfort in a typical manner (i.e., cry, moan, seek comfort, etc.). To date, many case studies and personal testimonies published in the literature on autism describe sensory and per ceptual abnormalities (Bursch, Ingman, Vitti, Hyman, & Zeltzer, 2004; ONeill & Jones, 1997). Some individuals with ASD do experience
43 alteration in auditory, tactile, and/or visual sensory processi ng. Caregivers describe the difficulties they encounter when their children with autism are upset secondary to altered sensory perception (Zeltzer & Schlank, 2005). In addition to sensory alterations, caregivers describe unusual, or absent, responses to painful stimuli in their children with au tism. Others however, can describe unique behaviors that alert them to their childs being in pain. Altered pain expression is not universally observed in ASD, bu t most experts agree that the pain experience appears different in this population. It is possible that impairme nts in communication influence how pain is expressed in these individuals. Overview of Communication Theory Frank E. X. Dance (197 0) reported that there are around 100 published definitions of communication. The conceptualization of comm unication has been addressed from a wide range, and diverse set, of disciplines. For these reasons, it is nearly impo ssible to construct one single, comprehensive, working definition of communication. The definition must take on the form most appropriate for the particular purpose or inquiry. Despite the diversity, many communication experts identify and a ddress the various properties cen tral to most definitions of communication. In his text, Wood (1982) describes five premises that are foundational to a working definition of communication. He emphas izes that communication is a dynamic process, and that it is systemic. He points out that the w hole is greater than the sum of its parts, that communication is symbolic intera ction, and finally, that meaning in communication is personally constructed. The model he believes capture s these described premises is a Symbolic Interactionist Model of Co mmunication, which emphasizes dynamism and also the shared experience crucial to the overall function of communication. Th is trend from a more actionoriented model of communication towards an interactional, and ultimately transactional model of communication has been noted in other texts on communication theory as well (Budwig,
44 Wertsch, & Uzgiris, 2000; Leeds-Hurwitz, 1989; Smith & Williamson, 1977; Watzlawick, Beavin & Jackson, 1967; Wood, 1982). Bowers and Bradac (1982), similar to Wood (1982), published a number of competing axioms they believe comprise a foundation for communication theory. They list seven sets of axioms, which include, as one ex ample, the notion that while a) communication is the transmission and reception of information, it is also b) the generation of meaning. While a number of communication theorists aim to better define communication, and their work reflects the evolution and adaptation of mo dels attempting to define communication, other theorists explore a variety of other aspects of communication. Different disciplines generate different research programs, each aimed at addre ssing a particular resear ch question related to communication. Communication re search encompasses all issu es related to how and why humans communicate, from interpersonal comm unication, communicatio n methods used in educational settings, to mass media communicati on. Many of these theoretical frameworks are beyond the scope of this review, as they are not direct ly related to the re search question these studies addressed. The particul ar aspects of communication theo ry most applicable to this review include those foundations which addre ss the social aspects of communication, and additionally, how language is used (or not used) as a means of co mmunication. In an attempt to understand the deficits seen in children with au tism, and to discuss how different impairments might influence the verbal and nonverbal expressi on of pain, it is necessary to examine the closely related, but somewhat separate, body of literature focused on language development and use. Language and Communication Language and communication are intim ately re lated. The nature of the relationship however, is one of the controvers ial issues articulated throughout the language literature. For
45 example, Hoff (2005) points out th at there are different camps re garding whether the desire to communicate is the motivation for acquiring language. Several theo rists focus on the childs desire to share experience and express themse lves as the impetus for language emergence. Others counter argue that the complexities of grammar cannot be learned simply by the motivation to communicate, and they pose that these theorists give childr en too much credit when it comes to assuming they have such a sound social understanding (Hoff, 2005). This debate is also sometimes referred to as part of the formalis m vs. functionalism debate. Formalists regard language acquisition as a separate module of development, completely independent of communication (and other social or cognitive processes for that matter). This modular conceptualization derives historically from Chomskys articulation of language as a mental organ, and was further explored by Fodor (1983). Functionalists, on the other hand, tend to describe the process of language acquisition in light of the other cognitive capacities that are developing simultaneously, including the desire to socially in teract and communicate. They articulate how the interacti on between the use of language and the environment (other individuals), fosters the con tinued development of language and communicative capacity. Interestingly, some children with autism develop language skills, but fail to utilize language to successfully communicate, lendi ng support to the notion that language and communication are perhaps in some unique way distinctly separate. In examining the relationship between language and general cognition, Marcus and Rabagliati (2006) address the modul arity vs. domain-general deba te and provide insight into a possible compromise. They propose a theory whereby cognition and language share a common background from an evolutionary standpoint, an d have only recently diverged due to small changes. Thus, while distinct in some ways, they are also related, and research may ultimately
46 shed light on how language builds on and depart s from particular aspects of domain-general cognition (p. 1228). Marcus and Rabagliati poin t out that developmental disorders such as ASD, might be better accommodated by such an approach, and thus would explain why some children develop linguistic ability while remaining unable to communicate effectively. Language Development When theorists m ove beyond examining the nature of communication and how it relates to language, and attempt to specifically explain the development of linguistic competence in the early years of life, several vividl y different research enterprises can be identified. Despite ones approach towards the study of language, there are se veral major issues that every theory attempts to address and which continue to be pertinent issues in the field. These issues involve the initial state of knowledge regarding language, the proce ss of developmental change, and the role of external input influencing the development of language. The fo llow paragraphs highlight the most debatable issues that have historically laid the foundation for different camps within language development theory and research. Innateness One of the most long-standing debates invo lves the issue of innateness. Rapid developm ent regarding the study of childrens language followed a public ation in the 1950s by Noam Chomsky. The debate that ensued betwee n Chomsky (a nativist) and B. F. Skinner (a learning theorist) facilitated the emergence of two divergent positions regarding the origins of language. Chomsky focused on the grammar of ch ildren, and posited that there is an innate Universal Grammar (UG) whereby interaction with this UG (input from environment), and the presence of innate constraints, allows for the growth of a knowle dge of language. Nativists point to the ease and rapidity with which children acquire language, and also argue that even without implicit instruction children develop language abilities (Meltzoff, 1999). Skinner however,
47 espoused that the environment shapes the deve lopment and emergence of language without the original presence of anythi ng innate. The current state of this framework is usually referred to as connectionism. This initial debate ignited and fu eled language research, br inging it ultimately to the body of knowledge that is e xplored and expanded today. Current theoretical positions ha ve, in a sense, rejected th e pure behaviorist approach. Almost everyone agrees that there is something innate about the acquisition of language, and that it is the interaction of the innate capabilities with the environment th at best describe its emergence. The debate currently ensues re garding the nature of the innateness and the interaction, and it is in this realm that most theories currently differ. Developmental changes In order to understand w hat is innate in language, theorists must describe the process by which infants and children take that innate cap ability and develop language. Explanations for the process by which individuals l earn language vary significantly. Several particular issues are commonly addressed when referring to the deve lopment of language. These include explaining the process by which children acquire language, cons idering whether there ar e particular critical periods during the development of language, and examining the influence of input from the world on language development. Processes A variety of models attempt to detail the m echanism by which language abilities emerge. Modern research programs utilize newer tec hnology and methodology and have been able to gain detailed insight into the processes by wh ich children learn languag e. One segment of theoretical work focuses on babies abilities to identify patterns, learn rules, and even utilize statistics during language development. Work by Saffran, Aslin, and Newport (1996) emphasizes the unique ability of babies to iden tify patterns and use a form of statistics to
48 organize the sounds they hear and develop langu age. Pinker has writtenin light of learning rules and setting parameterst hat babies however are not ac quiring dozens or hundreds of rules; they are just setting a few mental switches (Akhtar, 1999, p. 340). Pinker also focuses on how children link rules based on an innate knowledge of word categ ories. Patricia Kuhl (2000) accounts for all concepts: rules, patterns, and statis tics; in addition she identifies other influences, such as experience, that likely shape language development. She concludes that they use inherent learning strategies that were not expected, ones thought to be too complex and difficult for infants to use (p. 11856). Recently, Kuhl examined speech and linguistic processing in children with autism (Kuhl, 2005). Conclusions from this work indicate that there is a link between social interaction and language learning. This notion th at social interaction is pr imal to the development of communication and language is historically based in writings by Vygotsky (Budwig et al., 2000). The process by which social functioning infl uences the development of communication and language is theorized by severa l different well-known resear chers (Bruner, 1983; Snow, 1999; Tomasello, 1992). These frameworks are often referred to as so cial-pragmatic approaches. Tomasellos approach however, has been referred to as cognitive-functional linguistics, and most recently has also been called usage-based lingui stics (Tomasello, 2003). The details of his theory will be discussed later in light of e xplaining particular f eatures of autism. Critical periods and continuity One issue commonly addressed by communica tion and language theo rists involves the conjecture that there are critic al periods in language developm ent, where children are able to acquire and use language rapidly. Beyond these critical periods, it is proposed that language processing is different. The critical period hypothesis is supported by several different arguments. One argument highlights the fact that children learn language at an extremely rapid
49 pace, and also points out that children are able to acquire a second-language with greater ease than adults. Additionally, isolated case reports of children rais ed without exposure to language, highlight how the children remain mute and acquire relatively little language ability if intervention occurs after a certain point in development. Those who disagree with the conceptuali zation of a critical period emphasize the continuous nature of communica tion and language development, and present counter arguments to the points mentioned above. They discu ss how adults learning second languages have a longer history of utilizing and working with their first language, which interferes with the ability to learn the second language. They also point out that children in general, across a variety of skills and cognitive capacities are more flexible le arners, aiding their abi lity to acquire language faster than adults. Patricia Kuhl (2000) highlights the learning strategies she has found employed by infants and reinterprets the criti cal period for language (p. 11850), stating that the critical period for language depe nds on experience, not just ti me (p. 11855). By experience, she refers to the neural commitment that is a result of experience. Each theoretical perspective on communication and language theory generally accounts for the previously discussed issues: the nature of what is inna te, and how the innateness translates via different learning processes and mechan ismsinto a functional language used to communicate. Additionally, in describing proposed processes and mechanisms, theorists address historically debated issues th at have divided communication a nd language researchers such as continuity vs. discontinuity, and the critical period hypothesis. The following section will explore how theories that emphasize the social-cognitive nature of communication and language development explain the deficits seen in children with autism.
50 Communication Impairment and Its Infl uence on Pain Expression in ASD Social and comm unicative deficits are core features of autism spectrum disorders. Many theorists attempt to explain the unique features se en in these children in light of the theoretical frameworks previously described. Those inte rested primarily in communicative impairments will re-visit what is believed about normal pa tterns of language development, and speculate where deviations in autism occur. Despite va rious theoretical perspect ives on language origin and development, there are several widely ac cepted beliefs about communication and language in ASD. As theories are explained in greater detail however, differen ces frequently emerge. It is generally agreed that an initial impai rment most likely occurs early in development (Lord & Paul, 1997). The impairment must lend to a range of consequences, because in autism, language abilities are highly variable. Language impairment is related t o, but not necessarily a result of, social and cognitive function (Lord & Paul, 1997). Despite deficits, progress within each domain of language usually follows a similar pathway as that of typically developing children. Lord and Paul (1997) refer to two general orientations towards autism: those with the modular approach (theory of mind, central cohere nce, and metarepresentation frameworks), and those with an information processing approach (executive function). They resolve that both orientations address how communi cation development is altered, and do so differently. Martin and McDonald (2003) address pragmatic language disorder and explore how it relates to weak coherence, theory of mind, and executive dysfunc tion. They conclude that all three processes can predict pragmatic language disorder, but do not specific ally address the underlying mechanisms responsible for the communication impairment. In examining the course of communicativ e competence and language development, theorists attempt to identify early indicators of communicative impairment and have identified preverbal behaviors that are directly related to the subsequent development of more complex
51 communicative, and eventual linguistic abilities. Mitchell et al (2006) found that communication and language delays can be pred icted by monitoring for delays in specific gestures. Other preverbal and so cial abilities strongly linked to the development of language and other communicative abilities include joint attention skills. Examples of joint attentional skills that children with autism either lack, or ha ve difficulty with, include referential looking, declarative pointing and showi ng, looking where others point a nd look, and social referencing (Carpenter & Tomasello, 2000). Joint attention is a precursor fo r other capabilities, such as engaging in role-reversal imitation, and unde rstanding communicative intentions. Tomasello (2003) published a usage-based th eory of language acquisition and thus produced one of the most comprehensive theo ries of communication from the functionalist perspective. He focuses on defining how one understands the communi cative intentions of others, and desires to share or di rect the attentional st ates of others. This, according to him, facilitates the emergence of language. He empha sizes the communicative na ture of interactions, even if there is minimal language exchange, a nd focuses on how the interaction itself can teach additional communicative skills. He also links gestures, such as pointing, to the emergence of an intersubjective experience from which lingui stic symbols are used to communicate. These basic, foundational social-cognitive skills are altered in individuals with ASD. What remains unclear is exactly how, or to what extent, these abilities are impaired. One of the most perplexing concerns to those who work in the field of autism is the extreme variability found along the spectrum, including variability in the communicat ive and language abilities of individuals. Most theories are un able to completely address this pa rticular issue. As a result, it is difficult to generalize research findings. Interventions and tr eatments therefore, must be tailored to each specific chil d. Additionally, the variability implies that each childs pain
52 experience, and thus ability to express his or her pain, may be a highly individualized phenomenon. This review of communication impairments in autism explores how particular deficits in social-cognitive and communicative skills might relate to the pain experience, and allows one to begin to approach the pain experience in children with ASD thoughtfully. Pain Assessment in Children The body of literature regarding the pain experience in typica l, healthy children has grown rapid ly over the past two decades. Significan t advances have improved the assessment, measurement, and treatment of pain in children of all ages. National orga nizations dedicated to the care of children have begun to acknowledge the prevalence of pain (acute, chronic, procedural, etc.) in this populati on. International efforts to exam ine the prevalence of pain in children, and attempts to characterize their pain have allowed investigator s to begin to quantify the presence of pain in children and adolescents. A recent article by Roth-Isigkeit, Thyen, Stoven, Schwarzenberger, & Schmucker (2005) fro m Germany published pain prevalence rates as high as 83% (n=739) when children were asked to report if they had e xperienced pain in the previous three months. Of children in the United States, 20% (aged 5-17) suffer chronic headaches (Zeltzer & Schlank, 2005). Zeltzer & Schlank (2005), experts in managing chronic pain in children, also report that juvenile arthri tis is one of the most co mmon chronic diseases in childhood, and that the prevalence of fibromyalgia is at least as high as 6% in school-age children. It is estimated that up to 20% of children suffer from some sort of chronic pain condition (Bassett, 2005). As the understanding of th e prevalence and nature of pain in typically developing children increases, additional work into assessing and managing pain in subpopulations of vulnerable child ren becomes imperative. Currently, there is ample research devoted to developing, validating, and modifying tools for assessment and measurement of pain across all ages in typically developing children. There
53 are behavioraland physiologically-based tools for neonates and infants (Johnston, Stevens, Boyer, & Porter, 2003), and ther e are now developmentally appropriate self-report tools that have been modified for young and school-age children (Champion, Goodenough, von Baeyer, & Thomas, 1998; Gaffney, McGrath, & Dick, 2003). These instruments have been adapted and validated for use in a variety of settings (i.e., inpatient, outpatient, schoo l) and under a variety of circumstances (i.e., children with cancer, or children after surgery) There are tools available to characterize the quality of different types of pain (i.e., acute, chronic, neuropathic), and also to monitor the effectiveness of diffe rent treatments (i.e., measuring for changes in pain scores). Additional bodies of literature focus on chronic pain management in children, while some even focus on the pain associated with specific dis ease processes. The na tional recognition of the importance of pain assessment and management in children has led to the establishment of pain clinics, where entire teams of professionals are dedicated to helping children in pain. It is unreasonable to think that children w ith disabilities or deve lopmental disorders are exempt from the pain prevalence reports describe d earlier. Alternativel y, several pain experts posit that the prevalence rates of pain would logically be higher in these children (McGrath, 1998). Stallard, Williams, Lenton, & Velleman (2001) report that everyday pain in cognitively impaired, noncommunicating children is commo n, yet rarely treated. When specifically considering children on the autism spectrum, one needs to recall the commonly reported hypersensitivities and the variety of behavioral tendencies these children exhibit. Zeltzer & Schlank (2005) propose that children with alread y altered neurological systems will experience, and thus respond to, basic pain se nsations differently. Bursch et al. (2004) posit that abnormal sensory processing will not eliminate the ability to feel pain, but it might affect the cognitive interpretation of the sensory experience. Additionally, in children with ASD, several
54 investigators propose that the anxiety often experienced by this population might actually contribute to the presence of pain, es pecially in those whose muscles remain tense for long periods of time (Zeltzer & Schlank, 2005). Many children in these vulnerable populati ons endure medical procedures and everyday painful events similar to typi cally developing children, if not in greater frequency. It is imperative that the current movement to examin e and better understand pain expression and management in these groups continue. In preparation for discussing the development of an instrument for use in ASD, the remainder of this literature review closel y examines the research that informs what is known about pain assessme nt in some similar, but distinctly different, populations of childrenthose wi th cognitive, neuromuscular, communication, and/or other developmental disabilities. Pain Assessment in Vulnerable Pediatric Populations Despite the wealth of information available on pediatric pain assessmen t, there is a paucity of research focused on pain assessment in children with cognitive, neuromuscular, communication, and/or developmental disabilities While children with ASD display some similar characteristics, and it might appear on the surface that a single pain assessment tool would be appropriate for all of these subpopulati ons of children, there are significant differences warranting an investigation of whet her children with ASD require a distinct pain assessment tool. Children with ASD are not always cognitively impa ired, and the combination of their social and communicative impairment is un ique to this population. Initial studies examining pain expression in ch ildren with disabilities reveal that it is inappropriate to use pain asse ssment tools currently available for typically developing children with these populations (Oberla nder & Craig, 2003). Children with disabilities co mmonly exhibit cognitive, motor, neurological and communicative impairment s which confound traditional
55 assessment tools and lend to inaccurate measurem ent of pain. For example, facial and spastic movements, guarding, groans, and tears may be associated with pain expression in the typical, healthy child. Some children with disabilities ho wever, display these behaviors as part of their daily, nonpain repertoire of expre ssions (Oberlander & Craig, 2003). It is therefore impractical to consider these behaviors as indicative of pain in all children with developmental disabilities (Breau et al., 2000). Regarding self-report m easures of pain, many children with cognitive and communicative impairments are unable to look at a se ries of faces (i.e., the Faces Scale) and selfreport which face reflects how they feel at the time. In the mid-90s, after realizing the limitations of available pe diatric pain assessment tools, several researchers focused on developing assessm ent tools applicable to specific groups of children; namely, those children with various deve lopmental disabilities. Behavioral assessment tools have subsequently emerged in the literature, and are now available for use in children with a) cerebral palsy (CP), with or without severe neurological impairment, b) infants and neonates withor at risk forneurological impairment, c) children with cognitive impairment and/or developmental delay, and d) children with cognitive impairment and the inability to communicate verbally. While thes e populations are somewhat simila r in nature, and researchers acknowledge overlap, they also emphasize the importance of c ontinuing to validate pain assessment tools in each specifi c population. The following sections review pain assessment tools currently available for these vulne rable populations of children. Cerebral Palsy and Neurological Impairment Giusiano, Jim eno, Collignon, and Chau (1995) first examined pain expressive behaviors in children with CP and neurological impairment. They developed a pain assessment tool comprised of 22 items and called it the Evaluation Scale for Pain in CP. They derived their items from behaviors that physicians identified as indicative of pain in children with CP during a
56 physical exam. The initial publ ication of this scale did not provide information on its psychometric properties. A follow up study by Collignon and Giusiano (2001) however, indicates that they have further developed and te sted the scale, creating a 10-item list of pain expressive behaviors for use in children with CP. They discuss validation of this newest version of the scale, and report that it demonstrates good psychometric propertie s. The ten identified pain indicators include: crying du ring manipulation, painful e xpression, int erest in the surrounding, increased tonus problems, coor dinated defensive reaction, moaning or inaudible cries, inability to communicate with nurse, increase in spontaneous movement, protection of painful areas, and observing pain-easing post uring. The final ten items reported by this group are similar to findings by McGrath, Rosmus, Canfield, Campbell, and Hennigar (1998). These researcher s were developing a pain assessm ent scale at the same time. Their scale, and its subsequent developm ent, is discussed in a later section. Hadden & von Baeyer (2002) extended the findi ngs of Giusiano et al. (1995) and McGrath et al. (1998) and worked to furt her explore the previ ously identified pain indicators in children with CPspecifically looking at those children with CP and severe neurological impairment. They sent surveys (The Caregiver Pain Survey) to parents of children with CP and asked 27 probing questions about pain expres sive behaviors in their childre n. They incorporated 24 of the indicators included in the a ssessment tool developed by McGr ath et al. (1998). The most frequently reported pain indica tors include: crying, less active, seeks comfort, moaning, not cooperating, irritable, stiff/sp astic/tense/rigid, decrease in sleep, difficult to satisfy or pacify, flinches or moves body pa rt away, and agitated/fidgety. Neonates at Risk for Neurological Impairment Several publications have em erged from a gr oup of pain researchers describing the pain experience in neonates at risk for severe neurological im pairment. Stevens et al. (2006) describe
57 a process whereby 14 pediatric pain experts determined the usefulness of 39 pain indicators used to assess pain in the neonatal population. The original list of 39 indicators was developed by the investigators. After two Delphi rounds with e xperts, they identified seven indicators that had high ratings of usefulness and importance. Thes e include: brow bulge, facial grimace, eye squeeze, inconsolability, increased and/or fluctuations in heart rate (from baseline), and reduced oxygen saturation. These identified behaviors support what has already been determined about pain assessment in neonates that assessment relies heavily on facial and physiologic indicators. Their st udy reveals that even with neur ological impairment, or other neurodevelopmental disorder diagnoses, pain assessment involves observing for specific facial responses and monitoring for changes in physiologic parameters. Children with Cognitive Impairment Not all children with profound cognitive im pair ment have simultaneous motor difficulties (like those seen in CP). Child ren with cognitive impairmentwhose daily functioning is typical in regards to movement and activitycomprise another population where preliminary pain assessment work has been conducted. Terstegen, Koot, de Boer, & Tibboe l (2003) developed an item pool of 138 potential pain indicators for this population (based on a quasi-qualitative study). While all children had profound cognitive impairment, not all children suffered from motor impairment. They subsequently observed (by video) children post-surgically, to watch for how often the potential pain indicators were present. Twenty-three reliable, observable indicators showed sensitivity to pre and post surgical observ ations. The most sensitive indicators were from the category Facial Expressions, which includes 10 behaviors. Other categories include Motor Behaviors (i.e., tensed up), Soc ial Behavior/Mood (i.e., panics, accepts comfort, and rebellious), Vocalizations, and Physiological Signs. For the first time, Terstegen et al. (2003) speak to the social-c ommunicative impairments sometimes seen in
58 children with cognitive impairment, and thus first identify another dimension important to consider in pain assessment. Non-Communicating Children with Cognitive Impairment The developm ent, and subsequent publication, of these various pain assessment tools has been somewhat simultaneous. Thus, there is overl ap within the research, and it is difficult to describe how each assessment tool builds upon the others. The benefit however, of having tools emerge around the same time for similar populations, is that comparison of each instrument reveals stark similarities and differences. There are solid commonalities amongst the tools available for each population, but ther e are also subtle differences that speak to the nature of each specific population. This premise supports the argument that a pain assessment tool specifically for use in ASD is warranted. One of the more widely available, well deve loped and validated instruments available for use in children with cognitive impairment and communication impairment is the previously mentioned assessment tool developed by McGrat h et al. (1998). McGrath and colleagues utilized caregivers of children with CP to generate a list of behaviors they typically use to discern the presence of pain in their children. Of note, none of the caregivers interviewed had a child with ASD. They identified 30 behaviors and put them into 7 categories, which include Vocal, Social, Facial, Activity, Body and Limbs, Physiological, and Eating/Sleeping. In a subsequent study, Breau et al. (2000) further tested this ch ecklist, now titled the NonCommunicating Childrens Pain Checklist, utilizing a group of 32 caregivers. Eight of these caregivers had a child with autism/PDD, however no children with ASD are reported to have been included in any of the subsequent validati on studies (Breau, Finl ey, McGrath, & Camfield, 2002; Breau, McGrath, Camfield, & Finley, 2002). Preliminary validation indicates that the checklist exhibits internal cons istency, and can detect pain and is reliable over time. They note
59 that the checklist needs furt her evaluation, specifically in other groups of children with impairments, and that it should be determined whether the checklist would be helpful to providers in the clinic or inpatient setti ng who might not know the children intimately. A revised version, The Non-Co mmunicating Childrens Pain Checklist Revised, (Breau, McGrath, Camfield, & Finley, 2002) and a post operative version, The Non-Communicating Childrens Pain Checklist Postoperative Version (Breau, Finley, McGrath, & Camfield, 2002) have been subsequently validated and are now ava ilable. The revision did not change any of the thirty pain expressive behavior s, but simply re-worded several of the items, and changed the response format for each item from present/absent, to a 4-point Likert scale. The postoperative version is similar to the original checklist; how ever it does not contain the Eating/Sleeping items because these are not direct ly relevant to postoperative pain in the clinical setting. The postoperative version demonstrates good psychometric properties and it is reported that this scale is valuable when assessing pain in children with intellectual impairments postoperatively. Additional development from this initial work with the Non-Communicating Childrens Pain Checklist has resulted in a) identification of a subset of seven items (not cooperating, cranky, irritable, unhappy, seek s comfort or physical closene ss, change in eyes, including squinching, eyes wide open, eye fr own, not moving, less active, qu iet, gestures to or touches part of the body that hurt s, tears, and gasping) that can successfully predict pain in children with cognitive impairment who cannot comm unicate (Breau, Camfield, McGrath, Rosmus, & Finley, 2001), and b) another attempt to develop a checklist aimed at assessing chronic pain in children without CPbut who have a wider range of cogn itive impairment and the inability to communicate (Stallard, Williams, Velleman, Lenton, & McGrath, 2002).
60 This development by Stallard et al. (2002) i nvolved caregiver interviews whereby 203 pain cues were identified and subsequently examined by experts. They identi fied six cues used by 90% of caregivers as indicators of pain. These include: crying, screaming/yelling, distressed face, tense body, difficult to comfort, and flinches if touched. The investigators report that despite asking about chronic pain, these cu es tend to identify acute pain responses, and they highlight the difficulties encountered when a ssessing chronic pain in these populations. They also mention that even though this scale was develope d utilizing parents of children with a wider range of cognitive and comm unicative impairments, its usefulness in some groups of childrenspecifically children with ASDis still questionable and needs to be explored. Stallard et al. continued work with these six identified cues and tested the instrument they developed and titled the Pain Indicator for Communicatively Impair ed Children (PICIC). They used 67 caregivers of children with central nerv ous system disorders (no children with ASD). They found that 5 cues were correlated with care giver judgments of pain, and that the presence of one observable cue (screwed up or distressed looking face) co rrectly identified almost 72% of the children in pain, and 93% of the children not in pain. Interestingly, crying was not a useful indicator of pain, given that many pare nts acknowledged the presence of crying in a number of distressing situations other than pain. One of th e more novel findings in this particular work was the addition of pain cues by caregivers. Ca regivers noted that he or she sometimes pushes me, bites bottom lip, m oves from one place to another, claps his hands, bangs head, laughs, rubs head, and/or bites, when in pain. The PICIC might be a useful caregiver tool, but may not be helpful to individuals unfamiliar with the child, and the additional items noted by caregivers speak to a) the intimate knowledge caregivers have
61 regarding pain in their children, and b) the fact that pain expression in children with cognitive and communicative impairment may be quite individualized. Acknowledging the individualized nature of pa in expression in these special populations, Solodiuk and Curley (2003) developed The Indivi dualized Numeric Rating Scale (INRS) to help assess pain in children with cognitive impair ment and communication difficulties. Their assessment scale is not a check list, but rather resembles a process by which a childs personal pain profile might be generated. It asks caregivers to think a bout their childs past painful events, provides specific indi cators they might take note of regarding facial expression, body movements, activity, soci al interaction, cry, and consolabilit y, and then asks caregivers to describe what their child does when in: no pain, moderate pain, and severe pain. The resulting picture of the child in pain is then provided to individuals i nvolved in his or her care (i.e., nurses, physicians, teachers, therapists, etc.). This novel approach speak s to the uniqueness of each childs individual expression of pai n, however it requires that a caregiver know what his or her child does when in pain. Finally, two additional qualitative studies dire ctly address the experience of pain in children with special needs and severe neurologi cal impairment (Carter, McArthur, & Cunliffe, 2002; Hunt, Mastroyannopoulou, Goldman, & Seers, 2003). These qualitative studies report that there are three forms of knowle dge required for optimal pain assessment and management: knowing the child, knowing the population, and knowi ng the science (Hunt et al., 2002). Carter et al. (2002) highlight issues im portant to caregivers of childre n with special needs regarding pain management. They report four emerging them es: learning to live with pain, dealing with uncertainty, expression of pain, and maki ng decisions. None of the parents had been taught
62 about pain management in their children; they had to learn. Parents discussed the lack of information available and reinforced the need for individualized pain assessment tools. Information regarding the pain experience in a few other pop ulations is available, but scarce, and does not involve development of pain assessment tools. For example, there are reports of children with Downs syndrome having delayed processing of pain and a latent behavioral response to painful stimulusbut th ere is no pain assessmen t tool specifically developed for use in children with Downs syndrome (Hennequin, Morin, & Feine, 2000). As previously discussed in Chapter 1, there is on e study indicating that children with ASD have a significant facial response to painful stimuli, but their means of assessing the acute pain event involved utilizing an intricate a nd detailed facial coding schemeone that is impractical to use on a regular basis for pain assessm ent in ASD (Nader et al., 2004). Challenges to assessing pain in children with ASD involve the sometimes held perception that children with ASD cannot feel pain, and thus this research aimed to determine whether caregivers could assess pain in th eir children with ASD. From ther e, it can be determined what type of pain assessment tool would best benefit this population. This review of available pain assessment tools, and the examination of resear ch looking at pain assessment in vulnerable populations of children, highlights the importance of continued pain assessment research in these unique populations. It is apparent that pain e xpression in children with cognitive impairment, communication difficulties, developmental and neuromuscular disorders is somewhat variable. Several core pain expressive behaviors have been identified for th ese populations, however additional work is required to ensure that pain assessment is valid in each of the different groups of children. While children with ASD might exhi bit several of the already identified pain expressive behaviors, close exam ination of this population is warra nted. The available literature
63 provides a foundation from which additional research into pain assessment in children with ASD may stem. Ethical Considerations This next discussion illum inates several ethi cal issues specifica lly relevant to the conducted research. It brie fly presents the ethical principles pertinent to these issues, and describes measures taken to minimize risk to study participants. The first issue relates to the fact that this research involved vulnerable populations (childre n, children with developmental disabilities, and caregivers of children with developmental disabili ties). The second issue relates to the nature of the research question which involved the study of pain. Vulnerable Populations Autonom y refers to recognizing an individual s capacities and perspective, including his or her right to hold certain view s, to make certain choices, and to take certain actions based on personal values and beliefs (Beauchamp & Walters 1999, p. 19). In an effort to preserve a participants autonomy, most rese arch involving human subjects requires that the participants provide informed consent. The ability to give in formed consent requires that a) participants be given all information regarding th eir participation, b) pa rticipants realize th eir participation is voluntary, c) participants understand all the information given to them, and d) participants have the capacity for decision-making (Ascension Healt h, 2007). Special consider ations were made to ensure these requirements in the conducted studies While the first stages of this research involved primarily the caregivers of children with ASD, future studies might involve more active participation by the children themselves, and thus i ssues pertinent to these future studies are also reviewed in this section. Whenever possible, children need to be in cluded in the informed consent process (American Academy of Pediatrics [AAP], 1995; Helseth & Slettebo, 20 04). Legally, minors
64 cannot enter into a contract, and thus they requ ire parental proxy. Id eally, even though minors cannot give consent, it is useful to obtain assent Assent is the moral re quirement to acquire the closest approximation of consent one can achieve within the childs capacity to understand (Helseth & Slettebo, 2004, p. 302). Research involving children w ith ASD poses an additional challenge to assent when one considers the li mited communication abiliti es of the population. Because ASD is a heterogeneous disorder, some children may be more capable of understanding their role in participation than ot hers. It is important to work w ith parents to ensure that their proxy consent is informed, and that if it is possi ble to obtain assent, this is done. The first study conducted in this research utilized surveys, which were filled out solely by caregivers. An informational letter was presented with the survey, describing the voluntary nature of filling out the form and returning it. According to the Inst itutional Review Board (IRB) at the University of Florida, consent may be implied if the caregiver fills out the form and then voluntarily sends it back. The requirement for documentation of informed consent may be waived (University of Florida Institutional Review Board, 2003/2004) and was approved for this research. Working with caregivers of children with ASD was also a unique situatio n. Parents of children with developmental disabilities need to know that their particip ation in research is completely voluntary, and that the care of thei r child in no way depends on participation in studies. If there is any sense that they must participate in order to receive quality care for their child, ethical principles involving ju stice may be violated. There ar e several principles of justice which are typically contextual ized when discussed (Beauchamp & Walters, 2003). Beauchamp and Walters (1999) point out the minimal, beginni ng principle as being concerned with ensuring that like cases should be treated alike, or, to use the language of equality, equals ought to be treated equally and unequals unequally (p. 22). Research with parents of vulnerable children
65 requires that when recruited, these parents are ensu red that the care of their child does not change with or without their participation. Util ization of institutional re view boards can help the researcher examine the process by which participants (parents and their children) are protected. The Study of Pain When working in fields involvi ng the clinical treatm ent and research of pain, there is an ethical im petus to relieve pain. Hadjistavropoul os (2004) points out that regardless of ones ethical perspective, ensuring that vulnerable popul ations have access to adequate pain assessment and management is not usually a controversial issue. Walco, Burns, and Cassidy (2003) point out that numerous papers assert that the alleviatio n of pain is an ethical imperative in nursing and medicine. According to Ferrell and colleagues (2001), the undertreatment of pain in vulnerable populations (children, cognitively impaired indivi duals, elderly), is a top ethical concern among physicians and psychologists who manage pain. The ethical principle of beneficence, and its relationship to nonmaleficence is pertinent to examining the phe nomenon of pain. According to Beauchamp and Walters (2003), beneficence is a complex principle, involving ones obligation to 1) actively promote good, 2) prevent evil or ha rm, 3) remove evil or harm and 4) do no harm (nonmaleficence). In an attempt to ensure that children with communicative and social impairments are not experiencing unrelieved pain, this research could be considered an ethical imperative. The studies conducted did not involve purposefully inflicting pain upon participants, and because of the vulnerability of this populat ion, this would never be an ac ceptable research method. Any examination of children in pain would have to oc cur in a natural setting, whereby the child might be in pain from some everyday occurrence, or perh aps because of a necessary medical procedure. Similar to pain research in the elderly with dementia, prescribed protocols ensuring that no unnecessary harm is inflicted on participants will n eed to be developed. Effective forms of pain
66 relief must be provided if pain is detected at any time (Hadjistavropoulos, 2004). There are current ethical codes available to guide research in the domain of pain, published by a variety of regulatory organizations, and se veral organized groups focuse d on the study of painthe American Pain Society [APS], for example. This research was guided by ethical codes, and all research involving parents and children underwent review by the IRB, whose sole purpose is to protect research participants. Rationale and Need for Instrument Development The literature does not report th e exis tence of a valid, reliab le instrument to assess for and subsequently measure and managepain in children on the autism spectrum. Any attempt to utilize an existing tool, or adapt a tool for this population is confounded by the considerable confusion surrounding whether children with autism ar e even capable of feeling pain. It is not known what specific parts of the pain process might be impaired or influenced by their inherent limitations. Because of the confusion that exis ts regarding how children with ASD experience pain, it was inappropriate to attempt to design an assessment tool without a better understanding of the phenomenon at hand. The studies conducted in this research were necessary in order to determine whether it is appropriate to continue instrument development. They a) explored caregiver perspectives regarding pain expression in children with ASD, b) generated a number of potential pain expressive behavi ors appropriate for an initial item pool, and c) reduced the initial list of behaviors to a subset of items that can now be used to further develop a tool for assessing pain in this population. Summary A review of autism (its prevalence, theo retical positions rega rding etiology, and description of core features asso ciated with the disorder) prepares the reader to consider how pain expression might be different in this population. Additionally, a focused theoretical
67 discussion on the nature of communica tion, communication development, and the communication impairments specific to autism allows one to realize that se lf-report of pain is limited in this population. Detailed examination of work that has been done to examine pain expression in similar and related populations of children provides an understanding of the foundation from which this current research ex tends. Finally, disc ussion of the ethical implications related to a) working with vulnera ble populations, and b) the study of pain, ensures the reader that this Investigator minimized risk to study participants when conducting this research. This review describes pain assessment t ools available for children with cognitive impairment, children with neuromuscular disord ers (i.e., CP), and children with communication impairment. These assessment tools provide valu able insight into how families and healthcare providers assess and evaluate pain in unique ch ildren. None of the aforementioned instruments however, focus solely on children with autism spectrum disorder. While children with ASD do have limitations in communication, the brea dth of their limitations extends beyond communication, to include the social and behavioral impairments previously de scribed in detail. It is inappropriate to use existing instru ments to assess pain in this population. Furthermore, several of the indicators included in the more recently developed instruments for children with communication impairment may not be applicable to children with autism (i.e., seeking comfort or physical closen ess). If one watches for thes e specific indicators, and does not see them, it may be wrongly assumed that a ch ild with ASD is pain free. This review highlights the theoretical uniquene ss of the pain experi ence in ASD, and suppor ts the premise of this research: that it was necessary to develop a pa in assessment tool specifically for use in this population.
68 CHAPTER 3 METHODS AND PROCEDURES The purpose of this research was to explor e how child ren with autism spectrum disorder (ASD) express pain, and to begin development of a pain assessment tool for use with this population. A series of two studi es were conducted to a) descri be how caregivers of children with ASD determine when their children are in pain, b) understand what experts believe about pain expressive behaviors in children with ASD, and c) establish the content validity of a set of pain expressive indicators that may be used to assess pain in children with ASD. Instrument Development DeVellis (2 003) articulates eight steps nece ssary to develop a psychometrically sound instrument. They include a) determine what is to be measured, b) generate an item pool, c) determine the format for measurement, d) have th e initial pool reviewed by experts, e) consider the inclusion of validation items, f) administer items to a development sample, g) evaluate the items, and h) optimize the scale length. M easurement experts (Allen & Yen, 1979; Crocker & Algina, 1986; Portney & Watkins, 2000) emphasize the extreme importance of the initial steps in instrument development. This research aimed to complete the first four steps in developing a pain assessment tool for use in ASD. These st eps are critical for future research, which will continue the development and testing of a pain assessment tool for use in this population. Thoughtfulness early in the development process ensures that an optimally valid instrument results. Study 1: Caregiver Perspectives of the Pain Experience in ASD The first study conducted for this research was titled Understanding the P ain Experience in Children with Autism: Caregiver Perspectives and asked caregivers to provide information
69 about how they determine when their children wi th ASD are in pain. From this point on, this first study will be referred to as Study 1 and the second study will be referred to as Study 2. Purpose The purpose of this exploratory descript ive s tudy was to characterize caregiver perspectives and concerns regardi ng pain in their children with au tism, and to begin to identify pain expressive indicators and behaviors unique to children with ASD. This was achieved by examining caregiver responses to a mailed surv ey and by conducting semi-structured interviews with a random subset of caregivers. The resul ting descriptive data provided insight into how children on the autism spectrum experience a nd communicate pain to the world around them. Study Design This two-phase study was approved by the Univ ersity of Florida H ealth Science Center, Institutional Review Board (IRB). During Phase I, a survey was developed by the author and m ailed to four hundred families of children with a diagnosed autism spectrum disorder. During Phase II, ten caregivers participated in a semi-str uctured phone interview in order to discuss their childs pain expression in more detail. Participants: Recruitment, Consent, and Anonymity Participants included caregivers of children with a diagnosis of autism PDD-NOS, or Aspergers disorder (AD). All caregivers were identified as primary caregivers, living in the home with their child with ASD, and all were ab le to write and speak in English. Surveys were distributed, by mail, to 400 caregivers whose children utilized se rvices and resources provided by the Center for Autism and Related Disabilities (CARD) at the University of Florida. Response was voluntary. A detailed cover letter provided information regarding the study aims and caregiver rights as participants (see A ppendix A). This study was conducted with IRB
70 permission to waive documentation of informed consent, und er the assumption that by voluntarily responding, it could be safely assumed that participan ts implied their consent. Interested caregivers anonymously responded by completing the mailed survey and returning it (also by mail) in a pre-addressed pre-stamped envelope, which had been included with the survey. Caregivers were given the option to provide their first name and telephone number, if they were interested in participating in the semi-structured interview via a follow up phone call. A subset of caregivers were randomly selected, and then invited to participate in Phase II. Telephone interviews were then c onducted. Caregiver responses and thoughts were hand-written as the interview was conducted, and then typed by the primary investigator. All surveys and response sheets were de-identified prior to analysis. Measures/Instruments Two instrum ents were developed for this e xploratory descriptive study, the Caregiver Survey on Pain Expression in Children with ASD and the Semi-Structured Interview Guide to Discuss Pain in ASD (see Table 3-1). Caregiver Survey on pain expressio n in children with ASD The Caregiver Survey was developed in orde r to elicit perceptions caregivers have regarding their ch ilds pain expression. In addition to basic demographic information, participants were asked three yes or no que stions, and were prompted to provide additional descriptive information re garding their child. Semi-structured interview guide to discuss pain in ASD Al most 80% of survey respondents provided a first name and phone number in order to receive a follow-up call. A random set of ten caregivers were calle d to conduct the semistructured interview. The interview involved asking ten ope n-ended questions with each
71 participant. Caregivers were allowed to elaborate on those issues that they felt were most important and concerning to them. Data Synthesis and Analysis Caregiver survey responses were com piled. Answers to yes or no questions were quantified, and descriptive information was examined to identify common themes and issues. All pain expressive indicators (or behaviors) described by caregivers were noted, and used to generate a growing list of potential pain indicator s in this population. This list of items was utilized in Study 2 in order to de termine whether it is possible to develop a pain assessment tool for use in this population. Caregiver challenges re garding pain assessment in their children were noted. Details from the semi-structured interviews were used to better qualify the beliefs and perspectives held by caregivers re garding pain expression in their children. Several case studies were drafted to inform others about pa in expression in children with ASD. Study 2: Expert Appraisal of Potential Pa in Expressive Indicators in ASD The second study conducted in this research was titled Expert-Rated Relevancy of Potential Pain Expressive Indica tors in ASD: Establishing Conten t Validity. From this point on, this second study will be referred to as Study 2. Purpose The purpose of this study was to identify a co re set of indicators relevant to pain assessm ent in ASD. This was accomplished by a) generating a comprehensive list of potential pain expressive indicators in children with ASD, b) having this list reviewed by autism experts, and c) establishing the content validity of each ite m in order to reduce the list to a number of items achieving an acceptable scale-level content validity.
72 Participants: Expert Reviewers Fifteen experts were invited to participate in this study. Expert s were identified first by the investigator, and then discussed with a seas oned autism expertwhose role was to further examine and validate the background and experience of the identified experts. Experts needed to be currently working with children diagnosed with ASD. They needed to have a history of ongoing clinical and/or research projects in the fields of auti sm and/or pain assessment in vulnerable populations of children. Of the fifteen e xperts invited to participate, it was estimated that between six and nine would be able to review the list of poten tial pain indicators, which is the number necessary to establish content validity for individual items and the resulting scale. Measures/Instruments: Item List for Expert Review It is universally recommended that a larg e item pool be generated when developing instruments, in order to allow an investigator to include all possible items tapping into a construct, and thus devise a psychometrically stro ng instrument. The definition of large varies from one and a half times as large as the anticipate d final scale (Allen & Yen, 1979) to three or four times as large as the final version (DeVellis, 2003). Current pain assessment tools for use in specialized populations, such as the Non-Communicating Childrens Pain Checklist, are about 30 items long. Preliminary reports indicate that this tool with 30 items is easy to use by caregivers, takes little time, and accurately discriminates pain in populations in which it has been validated (Breau, Finley, McGrath, & Camf ield, 2002; Breau, McGrath, Camf ield, & Finley, 2002). Given that, theoretically, several of th e items in this already develope d tool might be appropriate for children with ASD, while other items may need to be removed, or replaced, it was anticipated that the final version of the t ool in development might be between 20 and 30 items. Review of the pain and autism literature, examination of other observational pain assessment tools,
73 caregiver surveys and interviews from Study 1, and prelim inary discussions with autism expertsnone of whom participated in Study 2generated 77 initial items. An item list was developed that cons isted of these 77 behaviors that might indicate pain in children with ASD (see Appendix B). This total number of initial items was appropriate, given the recommendations that the item pool be some where between one and a half (45) and four times (120) larger than the anticipated final instrument. These behaviors were put into 6 categories: Vocal/Verbal Indicators, Physio logical Indicators, Activity/Movement, Behavioral Indicators, Daily Activities, and Facial Indicators. Procedures and Data Analysis An initial iteration of the content validity pr ocess was conducted with experts in the fields of autism and pain to closely examine the curre nt item pool. Items were revised, removed, or retained based on expert input regarding the re ported relevance to pain assessment in children with ASD. Establishing content validity Measurem ent experts emphasize the importance of establishing content validity. Content validity represents the degree to which elements of an assessment instrument are relevant to and representative of the targeted construct for a particular assess ment purpose (Haynes, Richard, & Kubany, 1995, p. 238). Throughout the literature, researchers co mmonly report that content validity was established, however the process by which content validity was established is not always articulated. It is impor tant to point out that establ ishing content validity is a process. As articulated by Polit, Beck and Owen (2007), it ma y be necessary to undergo several iterations of item review, especially if the first review reveal s that a number of item revisions, clarifications, or additions are warranted. This research aime d to complete the first iteration of item review, and implications for a second round with experts are discussed in Chapter 5.
74 Before beginning this study, various methods for establishing content validity were examined. The process for establishing content validity has been described in several nursing texts, and is further outlined in a number of articles (Polit & Beck, 2006; Polit, Beck, & Owen, 2007). One of the commonly reported quantitative measures of content validity is the content validity index (CVI). Lynn (1986; as cited in Polit & Beck, 2006) recommends that between three and ten experts participate in the content va lidity process. Each ex pert rates every item on a 4-point ordinal scale labeled: 1 ( not relevant ), 2 ( somewhat relevant ), 3 ( quite relevant ), and 4 ( highly relevant ). Following review by all experts, an item-level CVI (I-CVI) is calculated for each item. For this study, a cover letter desc ribing the research accompanied the item list (see Appendix C) and instructed the reviewers to rate each of the 77 items as either not relevant, somewhat relevant, quite releva nt, or highly relevant to pain assessment in children with ASD. Item-level content validity indexes (I-CVI) I-CVI is calculated by figuring the num ber of experts giving a rating of either three or fourdivided by the total number of experts (Polit & Beck, 2006). To accommodate for chance agreement, Lynn (1986; as ci ted in Polit & Beck, 2006) advocat es that if there are five or fewer judges, the I-CVI should be 1.00 in order to retain an item as relevant. For six or more judges, she recommends an I-CVI of .78 or above. Others have calculated the I-CVI with five judges and accepted any item with an I-CVI abov e .80 (Decker & Perry, 2003). Polit, Beck, & Owen (2007) state that because establishing content validity may require several expert reviews, in the initial iteration it is wise to retain it ems with an I-CVI of .78 or above, however it may be acceptable to consider items with an I-CVI some what lower than .78 as candidates for revision rather than discard them too hurriedly. L ynns (1986) methods are most commonly reported throughout the nursing literature, and have b een adopted by many researchers embarking on
75 instrument development. Her methods are sometimes critiqued because they do not statistically take into account the possibility for chance ag reement (Wynd, Schmidt, & Schaefer, 2003). It is pointed out that other indica tors are available to calculate interrater agreement (kappa, for example), and do account for chance agreement (Polit, Beck, & Owen, 2007; Wynd, Schmidt, & Schaefer, 2003). Supporters of the CVI counter argue that kappa accounts for chance agreement and chance disagreement. Kappa also does not provide information at the item-levelw hich is often necessary when examining each items relevan ce (or irrelevance) to a particular assessment tool. I-CVIs provide information that decision-makers need in order to construct an assessment scale. Polit, Beck, and Owen (2007) argue that the CVI is appropriate for item-level analysis, and to address the debate regarding chance agreement, they calculated a modified kappa statistic (taking into consid eration both chance agreement and the proportion of expert agreement on relevance) to allow an investigator to compare each calculated I-CVI with what would translate into an excellent, good, or fair modified kappa coefficient. To articulate excellent, good, or fair, interr ater agreement, they utilize th e kappa coefficient criteria set by Cicchetti and Sparrow (1981). In a sense, thei r calculations indicate that items with high ICVIs often have adequate modified kappa coeffi cients, therefore indicating that the I-CVI is useful even with the described limitations regarding chance ag reement (and disagreement). This current research utilized the table developed by Polit et al. (2007) to determine whether all calculated I-CV Is were excellent, good, or fai r in relation to their modified kappa. For this study, it was establ ished a priori that six experts would participate in the initial expert review, and a cut-off of .78 would be utilized to determine whether items should be
76 retained. Items with an I-CVI just below .78 w ould be considered for revision, and those with the lowest I-CVIs would be removed from the list. Scale-level content validity index Despite use of the I-CVI to inform scale deve lopers whether to dis card or retain items, many published articles describing various scales development only report the scale-level CVI (S-CVI). Because there are several ways to compute the S-CVI, it is important for authors to provide information on how the S-CVI was com puted (Polit & Beck, 2006). Polit and Beck (2006) recommend calculating the S-CVI by reporting the average of all the I-CVIs in a final scale. For example, a scale with five items, whose I-CVIs are 1.0, 1.0, 1.0, .83, and .83, would have a calculated S-CVI of (1+1+1+.83+.83)/5 = .93. This calculation is referred to as the SCVI/Ave by Polit & Beck (2006). There is deba te regarding whether the acceptable S-CVI should be set at .80 or .90 (Polit & Beck, 2006). This initial item review with experts reveal ed that it was possible to remove some items, retain some items, and that there were some items that needed to be revised. As previously discussed, Polit, Beck, and Owen (2007) point out that it may be necessary to undergo multiple iterations in an effort to establish content valid ity. Because the initial list of items needed revision, an S-CVI was not yet ca lculated. If it is determin ed, following another review by experts, that item revision is no longer necessa ry; the S-CVI/Ave will be calculated according to specifications outlined by Polit and Beck (2006). An a priori determination will be made to accept .80 as an acceptable scal e-level content validity. Content validity procedures guarantee that the theoretical conceptualization of the investigator is supported by those co nsidered to be experts in the fi eld. This process ensures that items remaining in the tool following these co ntent validity procedures are a) supported
77 theoretically by the lite rature, b) supported by qualitative and/or quantitative study findings and/or clinical experience, and c) have systematically undergone expert review. Expert comments and suggestions Expert suggestions, questions and comments were solicited following completion of the item s list review. Experts were given the opport unity to comment on specific items, or perhaps to discuss the overall pro cess of this instruments development. Many researchers embarking on instrument development underutilize the expert a dvice that is available (Grant & Davis, 1997). Careful thought was given in order to select e xperts that would be inte rested in the current research, so that their advice a nd expertise would be readily available throughout the process. Expert comments and suggestions resulting fr om the item review have been thoughtfully considered and are reported in Chapter 5. The imp lications of their thoughts on this instruments further development will also be discussed. Summary The m ethods presented here describe an initia l attempt to think about the long-term nature of understanding how to assess pain in children with ASD. It is the intent of this discussion to ensure readers that the writer is adequately famili ar with the various issues related to instrument developmentdesigning studies that will ensu re the development of psychometrically sound instruments, understanding reliab ility and validity, and how to es tablish such measures, etc. Thoughtful preparation for every step in the proc ess of instrument development, including the two studies conducted for this research, ensure s the development of a psychometrically sound tool for assessing pain in childre n with autism spectrum disorder.
78 Table 3-1. Study 1: Caregiver survey and semi-structured interview questions Caregiver survey Semi-structured interview guide Compared to other children your childs age, do you feel your child responds differently to painful events? Please describe. Do you find it difficult to tell when your child might be in pain? Please describe. Do you feel healthcare providers are able to assess and treat your childs pain well? Please describe. Compared to other ch ildren your childs age, describe your general impression regarding how your child responds to painful events. Describe a specific time, if any, when your childs response to pain appeared unusual or different from other children. How do you determine if your child is in pain? What type of comfort measures might you use to help your child when he/she is in pain? Describe how your child communicates with you. How do you usually determine what your child wants or needs? Describe a time, if any, when your child got unusually upset or irritated by specific sounds, smells, or touch. Does your child ever deliberately injure himself/herself, for example by biting him/herself, or banging his/her head? Discuss the behaviors. Compared to other children your childs age, describe the range of facial expressions they display. Do you feel your childs pain is adequately managed by healthcare providers? If not, what improvements could be made?
79 CHAPTER 4 RESULTS The findings from two studies are presented in this chapter. First presented are results from Study 1, including demographic informa tion, and quantitative data for each question. Then, descriptive information provided by car egiversfrom surveys and semi-structured interviewsis presented. Next, results of Stud y 2 are described, beginning with item-level analysis. Items achieving acceptable content validity are presented. Those items not achieving an I-CVI of .78 are also discussed. Thoughts and comments provided by expert reviewers are reported, and anticipated item list revisions are introduced, but are more thoroughly discussed in Chapter 5. Study 1: Caregiver Survey Results Eighty-eight surveys were com pleted by caregiv ers of children with ASD and returned by mail. All surveys were included in data analysis. This is equivalent to a return rate of 22%. Given this low response rate, it is important to consider how a high nonresponse rate might influence findings and subsequent conclusions. This is discussed further in the Limitations section of Chapter 5. Demographic Information Caregivers reported that their children had a diagnosis of either autism (n = 32), PDD-NOS (n = 19), Aspergers disorder (n = 12) or likely autism (n = 25). For the purposes of this study, a category likely autistic was used because so me children exhibited enough autistic behaviors to receive services from the Center for Autism and Related Disabilities, but the childs official ASD diagnosis may not have been conclusive ly established at the time the survey was completed. A majority of the children described in Study 1 were autistic or likely autistic (n = 57). The population of children described by care givers consisted of 62 boys, 20 girls, and 6
80 children whose gender was not reported. These numbers reflect re ported trends that ASD occurs three times more often in boys than girls. For this sample, the male female ratio was 3.3:1 (62 boys, 19 girls). Quantitative Results from Caregiver Surveys Table 4-1 summarizes the quantitative responses to the three survey questions. Eighty-six percen t (n = 76) of caregivers felt their child resp onded differently to pai n. Seventy percent (n = 62) stated they find it difficult to determine when their child might be in pain, and fifty-six caregivers (64%) respon ded that healthcare providers ei ther did notor only sometimes adequately assessed and treated their childs pain. Descriptive Information Provided by Caregivers Question 1 : Compared to other children your childs age, do you feel your child responds differently to painfu l events? Please describe Written caregiver responses provided descriptive information regarding pa in expression in children with ASD. The responses to Question 1 were examined, and themes were identified by the author and an autism expert. Themes common to those caregivers who do feel their child responds differently to pain are reported. These themes highlight the mo st commonly reported differences articulated by caregivers, and were identified by looking at ho w frequently each issues was mentioned. The themes, along with a description/definition of each, are summarized in Table 4-2, and include exhibits altered sensitivity to pain, displays unusual/unique pain expression, demonstrates pain seeking behavior, pain expression varies by type, and pain expression varies with development. The theme, exhibits altered sensitivity to pai n, refers to a variety of caregiver reports describing either hyper-sensitivit y or hypo-sensitivity to pain. Several caregiver quotations illuminating this commonly reported phenomenon are presented in Table 4-3. The theme,
81 displays unusual/unique pain e xpression, refers to those res ponses where caregivers correlate specific behaviors and/or indicators to pain in their children. They openly acknowledged that these reported pain indicators are not typically associated with pain in children. Finally, examples of caregiver quotes from which the themes demonstrates pa in seeking behavior, pain expression varies by type, and pain expression changes w ith development were derived are also provided in Table 4-3. Question 2: Do you find it difficult to tell when your child might be in pain? Please Describe. Caregivers were asked whether they had difficulty determining when their child might be in pain. Seventy percent (n = 62) stated they find it difficult to determine when their child might be in pain. Written descriptions (n = 39, 44%) of the difficulties associated with pain assessment primarily reflected chal lenges related to working with nonverbal children (and thus children who do not verbally complain). Othe r caregivers (n = 27, 31%) reported that because their children do not cry, or because they appear stoic, it is difficult to determine if they are in pain. The thirty percent (n = 26) of caregivers reporting that it is not difficult to determine when their children are in pain stated that their childre n were verbal (n = 10), or they could identify specific pain expressive behavior s unique to their child. These reported behaviors are listed in Table 4-4, and were referenced when generati ng the comprehensive list of potential pain expressive behaviors for review in Study 2. Ot her caregivers (n = 4) stated they can tell when something is wrong, but get frustrated when thei r child cannot help them determine where or how serious the pain is. Question 3: Do you feel health care providers are able to assess and treat your childs pain well? Please describe. Fifty-six caregivers (64%) reported that healthcare
82 providers either did not, or onl y sometimes, adequately assessed and treated their childs pain. Written responses highlighted that many parents believe pain is an indicator that something is wrong in the body. Because they have a difficult tim e determining when their child is in pain, they are forced to depend on other signals (i.e., fever, gait disturbances ) to alert them that something might be physically wrong. The prevalent caregiver concerns involved their ability, as caregivers, to advocate for their children. They reported not bei ng taken seriously at healthcare facilities when they knew something was wrong, but their child was not exhibiting any typical pain expressive behavi ors. A number of parents reported that they have worked hard to find primary care physicians who understand ASD, and who will advocate for them and their children within the healthcare system. Study 2: Expert-Rated Relevance o f Potential Pain Indicators Information gleaned from Study 1 characterized caregiver perceptions regarding the pain experience in children with ASD, and highlight ed challenges associated with attempting to assess pain in children with so cial and communicative impairment s. Caregiver-identified pain indicators were added to a growing list of expr essive/observable behavi ors that might indicate pain in children with ASD. In Study 2, this list was reviewed by autism and pain experts in order to a) gain additional insight into pain expression in children with ASD, and b) identify behaviors considered highly relevant to pain assessment in ASD Examination of Item-Level Conten t Validity Indexes An item-level content validity index (I-CV I) was computed for each of the 77 items reviewed by the expert panel. I-CVIs ranged from 0 (where not one expert believed the item was either quite or highly relevant, to 1.0 (whe re ALL experts believed the item was quite or highly relevant to pain assessment in ASD. The calculated I-CVI for each item is displayed in Table 4-5. In summary: 4 items had an I-CVI of 0, 13 items had an I-CVI of .17, 13 items had
83 an I-CVI of .33, 14 items had an I-CVI of .5, 19 items had an I-CVI of .67, 13 items had an ICVI of .83, and 1 item had an I-CVI of 1.0. For this study, it was established a priori that six experts would par ticipate in the expert review, and an I-CVI cut-off of .78 would be utilized to determine whether items should be retained, revised, or removed from the list. Using these guidelines, 14 items achieved an I-CVI greater than .78. These items are referred to as the Primary Pain Indicators and achieved a high enough I-CVI to remain in the developing pain assessment tool. These items include: rocking unusually, lashing/striking out being less active, bangi ng his/her head, injuring oneself, crankiness (being more cranky than usual), lying ar ound/acting lethargic, increased heart rate, furrowing brow, s creaming/yelling/throwing tantrums, guarding (protecting) or favoring a part icular part of the body when moving, inconsolability (being unable to console), tensing face, and grimaci ng. Five of these items were originally categorized as Activity/Movement pain indicator s. Three items were from the Behavioral category, three from the Facial category, one from Vocal/Verbal, one from Daily Activities, and one from Physio logical. This highlights that experts considered a range of indictors to be relevant to pain assessment, from each of the aforementioned categories. As recommended by Polit, Beck, and Owen (2007) the group of indicators that achieved an I-CVI of .67 (just below the pre-established ICVI = .78) were not completely removed at this point in the instruments development, and will be considered for revision. They are referred to as the Secondary Pain Indicators (n = 19), a nd the implications of revising, removing, or retaining each of these items are discussed in Chapter 5. These items include: acting off according to caregiver, not acting like him/her self, crying, moaning, whining, whimpering (softly), stating/repeating a specific pain-relat ed word, seeking comfort or physical touch,
84 touching rubbing, picking and/or tapping a specific part of the body, changing color (flushing/redness), produc ing tears (without overt crying or vo calizations), increased rate of breathing, sleeping less, gri nding teeth, clenching jaw do wn, frowning, turning down mouth, pouting, quivering, puckering lips, widening eyes, becoming more aggressive (hitting, pinching, biting others, acting angry, and holding his/ her head for prolonged periods of time. Calculation of a Scale-Level Content Validity Index Because th is expert review resulted in the n eed to revise itemsand perhaps re-categorize the items rated most relevant to pain assessment in ASDthe final form of this instrument was not yet realized. It was also ther efore, not appropriate to calculat e the S-CVI at this point in the instruments development. This decision is a ddressed again in the Chapter 5. Once a set of relevant items are finalizedand prior to additional administration of the instrument to the sample populationthe S-CVI will be calculated. Expert Comments and Suggestions Expert com ments are briefly mentioned here, and the implications of these thoughts and suggestions are discussed in Ch apter 5. Comments generally in cluded a) suggestions regarding how to re-word particular items, b) requests th at better/additional operational definitions for particular items be provided, c) suggestions on how to more clearl y categorize the relevant pain indicators, d) information on how they (the particular expert) rate d whether items were considered relevant or not, and e) suggestions for the next steps in this instruments development. Summary These results highlight the perceptions held by caregivers regarding the pain experience in children with ASD. The ir insights regarding th e indicators they use to determine when their
85 children are in pain demonstrate the unique nature of pain expression in this population. These indicators were added to the list of potential pain i ndicators that had been already derived from the pain and autism literature. Finally, expert review of the entire list identified key pain expressive behaviors that will be used in the further development of an instrument to assess pain in children with ASD.
86 Table 4-1. Study 1: Quantitative results from caregiver surveys Question Yes No Sometimes Dont know Compared to other child ren your childs age, do you feel your child responds differently to painful events? 76 (86%)12 (14%)0 0 Do you find it difficult to determine when your child is in pain? 62 (70%)26 (30%)0 0 Do you feel healthcare providers are able to assess and treat your childs pain well? 19 (22%)40 (45%)16 (18%) 13 (15%) Note: n=88
87Table 4-2. Caregiver survey (question 1) results: Identification and description of frequently reported themes Identified themes Detailed description/definition Exhibits altered se nsitivity to pain: Hyper-sensitive to pain Hypo-sensitive to pain/has a high tolerance for pain Hyper-hypo-se nsitive phenomenon Child exhibits exaggerated response to minimal or seemingly non-painful events Child exhibits a minimal or lack of response to seemingly painful event Child oscillates between over-reacting and under-reacting to seemingly painful events Displays unusual/unique pain expression Child demonstrates delayed response to pain, or exhibits unusual/atypical pain expressive behaviors as identified by caregiver (i.e., laughing) Demonstrates pain seeking behavior Child engages in self-injurious behavior or seeks out painful Stimuli Pain expression varies by type Child reacts differently to various types (acute vs. chronic) or degrees of pain Pain expression changes with development As children grow and develop, pain expr ession and the ability to communicate pain changes
88Table 4-3. Caregiver survey (quest ion 1) results: Identified themes and sample caregiver quotations Caregiver quote Identified theme Benjamin* is very demonstrative he seems to feel things more intensely. He has extreme dramatic reactions to unexpected pain. Loud screaming and tantrums. It takes some time to calm him down. He makes a big thing about somethi ng small, like cutting his finger a little bit. Exhibits altered sensitivit y to pain: Hyper-sensitivity It takes a while for her to feel pa in. Sometimes has no painshe can get hit or bump into things and not realize she is hurt or bleeding. My child only responds to severe pa in and then he only rubs the place that has been injured and briefly fusses. Most times he doesnt even acknowledge his injuries. Exhibits altered sensitivity to pain: Hypo-sensitivity or high tolerance for pain A light touch will sometimes cause a painful reaction. Whereas, a deeper touch will not. Minor pa ins will cause crying. More moderate to severe pains will sometimes cause laughter! My daughter will either overreact or show no signs that it was painfulfor example I noticed sh e had half of one toenail gonebut she never reacted to it at all Other times she will go on and on about something as small as a little scratch. Exhibits altered sens itivity to pain: (Hyper-hypo-sens itive phenomenon) Generally, my child does not compla in of minor pains. He will punch himself in the stomach and that i ndicates he is a little constipated. He doesnt talk. If he is in pain he might stomp his foot or clap his hands loud. Just sort of off and less active and cheerful. She will say odd things like it tickles. Often giggles instead of saying ouch. Displays unusual/unique pain expression
89Table 4-3. Continued Caregiver quote Identified theme Jon* enjoys putting hands on hot things, such as light bulbs and oven burners. She will moan and bite her hand. He under and over reactsand he often seeks out ways to hurt himselfthrowing himself on the floor, biting his own arms. Demonstrates pain seeking behavior Blunt pain isnt as noticeable to hi m. However a splinter or scrape to the skin is very excessive in his expression. Yes to trauma types, i.e., needle pricksno to gas headaches, small cuts, blisters on his feet Pain expression varies by type When he was younger, he did not re spond to pain. For example, one day he had new shoes on all daya nd I didnt know they had the paper stuffed in them! Or hed bang his head and bite his hands. Now hes eight and can talk well, but now it seems he over-reacts to pain. The tiniest little thing gets him very upset. He reacts and seems to feel pain more now than when he was younger. He has improved greatly in his ability to recognize and describe painonly in the past two years has he been able to tell me he has a headache. Pain expression changes with development Note: indicates names have been changed.
90Table 4-4. Pain indicators identifie d by caregivers of children with ASD Behavioral indicators Activit y/Movement Vocal/Verbal indi cators Changes in routine Lashes out/becomes aggressive His/her body language Changes Cries Lays around more Uses actions Bangs his/her head Screams/throws tantrums Sleeps more Stomps his foot Touches whatever hurts Gets quiet Something is just off Claps his hands loud Holds his/her head U tters/states a specific wo rd Is less cheerful Bites his hand Hunches over when walking Laughs Is less active Rocks violently Hurts himself Moans Is listless Hides Gets still He tells me Note: Indicators in quotation were taken directly from caregiver written responses. Indicators not in quotation were derived by the investigator to summarize concisely what several caregivers reported.
91Table 4-5. Study 2: Calculated item-l evel content validity indexes (I-CV I)* for each item reviewed by experts I-CVI = 0 I-CVI = .17 I-CVI = .33 I-CVI = .5 I-CVI = .67 I-CVI = .83 I-CVI = 1.0 Humming/ singing Interacting more than normal with others Decreasing number and/or frequency of repetitive behaviors Sleeping more Sleeping less Lashing/striking out Rocking unusually Laughing Rolling eyes Sitting still Feeling cool or cold to touch Irritability (being more irritable than usual) Being less active Eating more than usual Raising eyebrows Stomping foot Lip biting Seeking comfort/ physical touch Banging his/her head Stating/ repeating a specific non pain word or phrase (i.e., bat, toe) Decreased rate of breathing Perseverating more on a particular item Fever Becoming more aggressive (i.e., hitting, pinching, biting others) Injuring oneself Seeming less sensitive to touch Making up stories or reenacting a painful circumstance Scratching a specific part of the body Acting angry Crankiness (being more cranky than usual) Becoming floppy, or goes limp Lip curling Interacting less than normal with others Widening eyes Laying around, acting lethargic Eating less, not interested in food Fidgeting, becoming more active, jumping around (motor restlessness) Isolating oneself more, hiding Frowning, turning down mouth Increased heart rate
92Table 4-5. Continued I-CVI = 0 I-CVI = .17 I-CVI = .33 I-CVI = .5 I-CVI = .67 I-CVI = .83 I-CVI = 1.0 Holding breath Cooperating less than normal Loosing interest in things that are usually extremely important or interesting to child Pouting, quivering, puckering lips Furrowing brow Chewing, moving tongue around and/or out, smacking lips Facial restlessness, tics Increasing number and/or frequency of repetitive behaviors Producing tears (without overt crying or vocalizations) Screaming/ yelling/ throwing tantrums Babbling (saying something incomprehensible) Vocalizing less than normal, getting quiet Flaring nostrils Increased rate of breathing Guarding (protecting) or favoring a particular part of the body when moving Clapping hands loudly Seeming hypersensitive (overly sensitive) to touch Muscle tensing, rigidity Acting off according to caregiver; not acting like him/herself Inconsolability (being unable to console) Continuous eye blinking Shivering, Shaking Unhappiness Grinding teeth, clenching jaw down Tensing face Moving to, or staying in, a specific location or safe space Changing his/her usual body language Sweating, perspiring Squinching, squinting eyes Grimacing
93 Table 4-5. Continued I-CVI = 0 I-CVI = .17 I-CVI = .33 I-CVI = .5 I-CVI = .67 I-CVI = .83 I-CVI = 1.0 Changing color (paleness) Crying (with or without tears) Moaning, whining, whimpering (softly) Changing color (flushing, redness) Touching, rubbing, picking, and/or tapping a specific part of the body Holding his/her head with both hands for prolonged periods of time Stating or repeating a specific pain -related word (i.e., hurt, ouch) Note: I-CVI was calculated by looking at th e proportion of expert reviewers rating an items relevancy as either quite or high. For example, if five of six experts rated an item as either quite or highly releva nt to pain assessment in ASD, the calcula ted I-CVI would be 5/6 = .83. All items in the I-CVI = .83 column had a calculated I-CVI of .83.
94 CHAPTER 5 DISCUSSION This research explored pain expression in chil dren with autism spectrum disorder (ASD). These two studies add to what is known about how vulnerable pediatric populations specifically, those unable to accurately self-report painexpress their discomfort to those around them. This research focused on children with autism spectrum disorder, whose pain experience has hardly been investigated. The literature review conducted fo r this research revealed one empirical study examining pain in this populatio n (Nader et al., 2004). Nader et al. (2004) demonstrated that children with ASD reflexively (involuntarily) respond to painful stimuli, indicating that perhaps th e perceptions of pain insensitivity in this population are inaccurately founded. This dissertation resear ch extended those findings, to be tter describe what caregivers and autism expertspresumably, individuals w ho know this population bestbelieve about how these individuals experience and express pain. Finally, this re search facilitated the initial development of a pain assessment tool for use in this population, by identifying specific pain indicators relevant to children with ASD. Pain Expression in ASD Compared to Typically Developing Children Nader et al. (2004) reported th at a) caregivers of children with ASD m ight be confused about pain in their children, and that b) parent s perceive insensitivity to pain in their children when compared to nonimpaired children. Findings in this research are consistent with these reports. Caregivers stated that they do believe their children respond differently to pain when compared to other developing children. When as ked to qualify these differences, however, they did not always refer to their children as having a high tolerance for pai n, or insensitivity to pain. In this way, these fi ndings are new and unique, adding to what is known about how caregivers perceive the pain experience in their children.
95 Several caregivers did refer to their childre n as having a minimal reaction to everyday painful occurrences; however a majority of caregivers desc ribed a phenomenon whereby their child would attend to various types of pain differently They described how a deep cut on the back of the childs head would cause minimal discomfort, but a tiny scratch on the hand would result in him or her being hyste rical. This phenomenon seems to be somewhat unique to this population. Bursch, Ingman, Vitti, Hyman, & Zeltz er (2004) posit that children with ASD may have intact acuity to sensor y input; however their processing of incoming information might be distorted. Thus rain might sound like gunfire, cl othing might feel like sandpaper, or fingers shampooing a scalp might feel like sharp meta l (p. 290). These sensory abnormalities have potential to distort the processing of pain stimuli as well, speaking to the fact that they feel the stimulus, but they may not attend to it as other children do. Thus, the reports by caregivers that their children seem hypersensitive to some pain, and then hyposensitive to other pain, might be explained by this overall abnormal sensory processing. This unusual responsiveness to the environm ent can be explained in part by stimulus overselectivity, the tendencyto respond to only a very limited amount of relevant sensory information (Bursch et al., 2004, p. 291). One car egiver described a situation where her child would grab a hot curling iron and not respond as if he were in pa in, but yet he would scream and yell in pain from the tag in his shirt. Sensory abnormalities are not seen in every child on the autism spectrum, but it may be that those child ren with sensory abnormalities processfocus on different aspects ofpain sensat ions differently. Additional i nvestigation into subgroups of children with ASD, for example those w ith sensory abnormalities, is warranted. Caregiver comments on how their children, when they did report pai n, would perseverate on the pain (or source of pain) al so highlight differences in the pain experience in this population
96 compared to typically developing children. The pe rseverative tendencies unique to children with ASD permeate all types of experiences, including their pain experience. Bursch et al. (2004) stated that once focused on pain, difficulties in shifting attentional focus can serve to increase pain and associated distress (p. 294). Interrupt ing the perseveration then might lead to pain relief. Zeltzer and Schlank (2005) describe a situation in their pain clinic where a child with ASD presented to them yelling repeatedly, Ow! Ow! Ow! He had been shouting for the past two months, since sustaining a leg injury. Zeltzers immediate suggestio n that he squeeze a ball instead of shouting Ow! resulted in a transfer of his expression, and he actually reported feeling better because he was no longer embarrass ed about shouting Ow! all the time. While this particular persevera tive response was clearly asso ciated with pain (i.e., it is easy to infer the child was in some sort of pain because the word Ow! is typically associated with pain), one has to consider that other children may also engage in repetitive behaviors when in painhowever they may select (especially if they are unable to verbally communicate) other behaviors not typically associated with pain (i.e., rocking unus ually). These articulate d differences indicate that caregivers do not always believe their children are incapable of feeling pain. Determining When Children with ASD are in Pain Caregive rs reported that they find it difficult to determine when their children are in pain. Of those caregivers who can dete rmine when their children are in pain, many are frustrated because they can never figure out why, or where, or how bad it hurts. This inability to determine when children with ASD are in pain could be attributed to the abnormal sensory processing previously describedwhich confuses caregivers, because it often contributes to a minimal (or lack of) response to seemingly painful events. A lternatively, it could be related to what parents look for in trying to determine whether their children are in pain.
97 Interestingly, a majority of the caregivers that stated their child doesnt feel pain, can handle pain more, or has a high tolerance for pain, alluded to how they make this inference. They mentioned that their child doesnt cry, doesnt say what is bothering him, is pain seeking, only cries a littl e, is nonverbal, or tells you in a calm manner that it really hurts mommy. One needs to examine what these ca regivers consider to be pain expressive behaviors in this popula tion. If caregivers are expecting their children to cry, or respond dramatically to pain, they may wrongly assume that their children have a high tolerance for pain, simply because they do not display these mo re typical pain expressive behaviors. Several caregivers acknowledged that their child ren experience pain differently, but they referred to this difference as being in the way they express their pain. They stated that they can determine when their child is in pain because he tells me, but tends to be quite stoic, he will punch himself in the stomach if it hurts, he wi ll lay around and sleep a lot, he will make up stories about getting away in a time machin e, or because he never communicates that something hurts, you have to pay close attention to behavioral cues. These caregivers have identified behaviors unique to their children that usually indicate they are in pain. One can see though, how some caregivers might interpret he will punch himself in the stomach as pain seeking, or having a high tolerance for painwhile other caregivers might identify that specific behavior as indicating pain in their particular child. Some caregivers also reported that they ha ve no trouble determining when their children are in pain because they are verbal, and can communicate when they are hurting. Comments regarding how they verbally communicate, however, indicate that those communication impairments typical of children with ASD permea te their ability to communicate pain. For example, they may be quite stoic, just scream just says several words and not express how
98 he feels about it at all, simply repeat a phrase he has heard, or reenact a story. Children with reported high functioning autism will more often tell their caregivers when they are in pain. These findings warrant investigation into whet her there are differences in pain expression between children with ASD of different diagnose s (i.e., Aspergers disorder vs. PDD-NOS) and varying degrees of communicative ability. Pain assessment in this population heavily reli es on the perceptions and beliefs held by caregivers. This research aimed to illuminate so me of the perceptions and beliefs that might be contributing to confusion regarding whether ch ildren with ASD feel pain. Clearly, the communication impairments described previously influence how children with ASD express pain. The theoretical model guiding this res earch proposes a relationship between observer bias, and an observers ability to decode pa in expressive behavior s. If caregivers are watching for more typical pain expressive behaviors, their bias might cause them to inaccurately assume that their children are not in pain. This research demonstr ates that there is a relationship between these two variables, and begins to illuminate what that relationship might be like. These caregiver surveys indicated that while children with ASD mi ght respond to painful events differently, caregivers do believe their children are capable of feeling pain. This conclusion is congruent with the findings report ed by Nader et al. (2004) that children with autism should be viewed as being as reactive to pa inful stimuli as children without autism and do not support beliefs about pain insensitivity in children with autism (p. 96). Interestingly, these surveys indicate that sensory processing abnorma lities in some children with ASD, and the perseverative nature of the disorder likely cont ribute to the confusion (especially experienced by caregivers) surrounding the pain experience in ch ildren with ASD. Astute caregivers have identified specific pain indicators (often non-typi cal pain indicators) that are uniquely exhibited
99 by their children. The second study conducted in this research aimed to investigate what autism experts believe about pain expression in ASD, and utilized the pain in dicators identified by caregivers to further explore whether development of a pain assessment tool specific to this population might be feasible. Communication and Social Impairme nts in ASD and Pai n Expression Findings from Study 1 and Study 2 further hi ghlight how the communication and social impairments seen in ASD might influence pain expression. Deficits in joint attention are theorized to impair a childs ability to develop social and communicative competencies. As previously described, joint attentional skills that children with ASD either lack (or have problems with) include referential looking, pointing and show ing, following ones gaze to look at an object with them, and visually following where others point. Tomasellos (2003) theory of language acquisition describes how understand ing the communicative intentions of others facilitates the emergence of language. The communication and soci al impairments in children with ASD affect the way they experience, and thus express, pain. Study 1 findings emphasize that those children with greater communicative competency do outwardly express their pain. Caregiver comments that their children are often stoic when they describe their pain, or that they will verbally communicate their pain by repeating a phrase like ants hurt you, demonstrate that despite verbal abilities, communicative deficits in pragmatic language use and echolalia influence pain expression. Children with Aspergers disorder might communicate their pain by using flat and emotionless speech, thus making it difficult to determine how severely they are hurting. Study 2 findings highlight that the pain experi ence in children with AS D might be more of an intrapersonal experience, rather than an interpersonal experience. This is consistent with the core social deficitsalso related to a lack of jo int attention skillsseen in children with ASD.
100 Several expert-identified pain indicators that we re unique to children with ASD seem to reflect an intrapersonal experience, for example: r ocking unusually, injuring oneself, banging his/her head, laying around, actin g lethargic, and holding his/ her head with both hands. Several pain indicators helpful in assessing children with cognitive impairment and/or communicative impairments that were not considered relevant to ch ildren with ASD reflect an interpersonal experience. These include: disinterest in surroundings, an d less interaction with others, withdrawn. These findings support the postulated signi ficance of the social deficits seen in children with ASD to pain expression. The Communications Model of Pain proposed by Hadjistavropoulos and Craig (2002) posits that social and communicative competenci es influence the pain experience and pain expression. These research findings suppor t the models proposed relationships, and preliminarily demonstrate that the communication a nd social impairments specifically seen in the ASD population appear to uniquely influence pain expression. When determining how to assess pain in ASD, it will be important to continue to test the proposed model, and consider how social and communicative features commonly seen in this population affect their abili ty to express pain. The studies conducted in this research establis hed a foundation from whic h future studies may extend. Items Determined to be Relevan t to Pain Assessment in ASD The second study in this research examined 77 potential pain indicato rs in children with ASD and aimed to determine which indicators are most relevant to pain assessment in this population. Expert review incor porated into instrument develo pment ensures that items are added, removed, and revised, via an informed pr ocess. As previously mentioned, it is not uncommon for item lists utilized in instrument development to undergo several iterations of expert review, in order to ensure that the fina l set of items validly represent the construct of
101 interest. This section presents the Primary Pain Indicators identified as being most relevant to pain assessment in this populat ion. Additionally, the subset of items that were considered relevant by a majority of the experts, howev er their calculated I-CVI fell below the preestablished cut-off (the Secondary Pain Indicators) are discussed. The Secondary Pain Indicators are evaluated, in light of this particular conten t validity process, and implications for leaving these items in the list for another expert review are presented. Finally, comments and suggestions provided by expe rts are reviewed. Primary Pain Indicators: Retained Items Fourteen item s had a calculated I-CVI greater than .78, which indicated that they achieved content validity for this construct of interest (pain). To review, these included: rocking unusually, lashing/striking out being less active, bangi ng his/her head, injuring oneself, crankiness (being more cranky than usual), lying ar ound/acting lethargic, increased heart rate, furrowing brow, s creaming/yelling/throwing tantrums, guarding (protecting) or favoring a part icular part of the body when moving, inconsolability (being unable to console), tensing face, and grimacing. When these items are compared to items in pain assessment tools currently available for use in related populations, there are similarities and differences The Primary and Secondary Pain Indicators derived from Study 2 are compared to three existing pain assessment tools in Table 5-1. The existing assessment tools chosen for comparison are the Non-Communicating Childrens Pain Checklist (NCCPC), the Pain Indicator for Communicativ ely Impaired Children (PICIC), and a Pain Evaluation Scale developed for children with CP. These three assessment tools were chosen for comparison because they have good psychometric properties and have demonstrated validity in the populations for which they were developed. When this Investigator
102 generated the original item list, indicators from each of these existing tools were included so as to identify whether those indictor s were relevant, or not releva nt to children with ASD. Significant differences imply that using existing pain assessment tools with individuals having an ASD would not be valid, and might lend to inaccurate pain assessment. For example, pain indicators identified by careg ivers in Study 1 were added to the item list developed by this Investigator. Several of these items were subs equently validated by expe rts in Study 2, as being relevant to pain assessment in ASD (i.e., rocking unusually). These indicators however, are not seen in the existing pain assessment tools ava ilable for use in similar and related populations. Furthermore, several of the indicators that are utilized to assess pain in cognitively impaired, non-communicating children were not considered relevant to pain assessment in ASD. These included, for example: disinterest in surrounding s, less interaction w ith others, withdrawn, flinching or moving away, sensitive to touch, and jumping around, agitated, fidgety. Given the underlying nature of ASD, these behaviors are commonly exhibited due to their disorder (i.e., less interaction with others, withdrawn), and th us would not be useful in pain assessment. Other pain expressive indicator s included in existing assessment tools, but not determined to be relevant to pain assessment in ASD, involved behaviors more applicable to children with CPbecause of their associated motor impairment ssuch as stiff, spastic, tense and rigid, floppy, or searches for an unusual position that calms. Additionally, several physiological indicators and changes in daily activity that were included in other assessment scales were not considered relevant to pain assessment in ASD (shivering, sweating, perspiring, eating less, not interested in food, and increase in sleep). A number of pain indicators identified by careg ivers were subsequently determined to be relevant to pain assessment in ASD by pain and autism experts. These indicators included
103 rocking unusually, lashing/st riking out, injuring oneself and banging his/her head. Other unique changes likely to indicate pain in this population, according to Study 2, include increased heart rate, and lying around/acting lethargic. One other finding that consistently arose throughout this res earch is that while ther e may be some unique pain indicators applicable to most children with ASD, there will also be those pain indicators identified by caregivers that are completely unique to the one child. Despite attempts to identify common pain indicators in children with ASD, pain assessment in this popul ation also seems to rely heavily on deviations from normal. Secondary Pain Indicators: Item Retention and Revision Several experts explained how they conceptualiz ed whether a n item was relevant to pain assessment in ASD. They indicated that they de termined an items relevance based on whether a particular indicator applied to all children with ASD (thus rating it as highly relevant), most children with ASD (quite relevant) some children with ASD (somewhat relevant) or not occurring at all in children with ASD (not relevant). As a result, a number of items (n = 19) were rated somewhat relevant by experts, but b ecause the I-CVI is calculated using only those ratings of quite and highly relevant, these items fell below the pre-established I-CVI cut-off of .78. Given that these items were close to the desired I-CVI of .78, and according to recommendations by Polit, Beck, and Owen (2007), it was determined that they should be closely examined before determining whether they should be discarded. Ta ble 5-2 lists the items with a calculated I-CVI = .67. Th is I-CVI (.67) was obtained when four of six experts rated the item as quite or highly relevant, and the other two rated the item as either not or somewhat relevant. When listing each item with an I-CVI of .67, as presented in Table 5-2, the differences in how experts rated each of these items can be more closely examined. Conceptually, there is a
104 difference between the item where ALL experts believed it was somewhat relevant or higher, and the item where two experts believed it was not relevant at all, and the other four believed it was quite or highly relevant. These tw o items however, given the way the I-CVI is calculated, would have an identica l I-CVI. This is clearer when one looks at a calculated sum score for each item. If 1 = not relevant, 2 = somewhat relevant, 3 = quite relevant, and 4 = quite relevant, a summed score can be gene rated for each item, depending on how relevant each expert rated the item. The items can th en be rank-ordered and one can look at the relationships between them. Despite all items having and I-CVI of .67, the variability becomes evident when one looks at the range of sum scores (15 19) amongst all items with an I-CVI or .67. For example, experts rated sleeping less as not rele vant (n = 1), somewhat relevant (n = 1), and quite relevant (n = 4). The sum score for this item is 15 (1+2 +3+3+3+3 = 15). Stating/Repeating a specific pain word was rated somewhat relevant (n = 2), quite relevant (n = 1), and highly relevant (n = 3). The sum score for this item is 19 (2+2+3+4+4+4 = 19). There is a difference between the item that has one expert ra ting it not relevant and does not have any experts rating it as highly relevant, and the item that was never rated not relevant and has three experts rating it highly relevant. Th e methodology used to generate the I-CVI does not take this variability amongst the items with the same I-CVI into consider ation, and by dichotomizing the 4-point Likert scale, information about each item is lost, and might be to the detriment of the scales development. A number of items that might be relevant to pain assessment in ASD, which is a highly heterogeneous population, would be removed if the strict .78 cut-off is used, without examining those items just below the cut-off.
105 Several of the Secondary Pain Indicators also need to be revised, as suggested by the expert panel. For example, crying should be combined with moaning, whining, whimpering, to generate one pain expressive behavior that captures what th e two separate indicators are reflecting: different degrees of vocalization. So the new item once the response format is determined, might be Vocalization. Caregivers would then rate thei r childrens vocalization on a Likert scale, choosing either: ( quiet ), 1 ( soft vocalization like moaning, whining, or groaning), 2 ( crying ), or ( screaming, yelling ). In doing this, the indicator screaming/yelling/throwing tantru ms would also be revised, and subsequently separated into screaming/yelling and throwing tantrums. Throwing ta ntrums would be taken out, resulting in a new items creationit was men tioned by one expert th at throwing a tantrum seems different than perhaps screaming or yelling in pain. Whenever items are revised, it is suggested th at they remain in the item list and undergo and second expert review to determ ine if revision made them more relevant to the instruments construct of interestin this case, assessing pain in children with ASD. All items that underwent some type of revision were retained in the item list for future examination of the items to further establish the content va lidity of a final set of items. The question arises in this research, what should be done with the Secondary Pain Indicators? Given a) the early stage of this instruments development, b) the methodological challenges associated with dichotomizing the I-CVI c) challenges associated with assessing pain in such a heterogeneous population, a nd d) the fact that this initial iteration revealed a need for several Secondary Item revisi ons, it was determined that th e Primary and Secondary Pain Indicators would both be utilized to generate the next it em list, and that an additional expert review is likely warranted. Item s subsequently not meeting Lynns (1986) criteria for adequate
106 content validity will be removed after the next review by experts, and a final set of items will include only those indicators deemed most releva nt to pain assessment in ASD by experts. The findings from Study 1 highlight the variabil ity seen in children with ASD as reported by caregivers. Children with ASD present with va rious patterns of strengths and weakness, and thus it may be impractical to develop one assessment tool appropriate for all children with ASD. The Primary Pain Indicators highly endorsed by expe rts indicate that there may be several pain expressive behaviors unique to this population and somewhat common across all children with ASD. The Secondary Pain Indicators might be re levant to subpopulations of children with ASD, or simply not all children with ASD. Experts believed them to be somewhat relevant but simply not highly relevant. These items need to be further explored, systematically, in another round with expertsor perhaps in a subsequent study where the assessment tool is actually utilized by caregivers and providers of children with ASD. Items Removed from the Item List Following Expert Review. All item s with an I-CVI less than .67 (n = 44) were removed. Least relevant items included laughing, humming/si nging, becoming floppy, or limp, interacting more with others, eating more than usual, and ch ewing, moving tongue around and/or out, smacking lips. A number of items may be re levant to pain assessment in in dividual children with ASD, or perhaps in small subgroups of children with AS D. These items not commonly seen in ASD reflect a more individualized pain process. The next section discusses how to accommodate these individually significant (but less common) pain indicators, in light of several expert comments. Expert Comments and Suggestions The Investigator solicited experts who could address th e relevance of item s to pain assessment in ASD and offer insight regarding th e overall development of this pain assessment
107 tool. Grant and Davis (1997) suggest using experts to help ensure items are clear, concise, and comprehensive. Expert suggestions regarding item revision were consid ered, and input on how to categorize the resulting pain indicators in an assessment tool was well received by this investigator. DeVellis (2003) points out however, th at the final decision to accept or reject the advice of your experts is your re sponsibility as the scale deve loper (p. 87). Given that suggestion, expert comments are presented here, but are considered in light of the overall goals and plans for this instruments development. A Need for Individualized Pain Assessmen t in this Population. Contrary to what experts believed to be relevant to pain assessment, several caregivers reported in Study 1 that laughing was sometimes an indication of pain in their children. This discord speaks to the fact that there will like ly continue to be unique and individual pain expressive indictors exhibited by children in this population. Expert suggestions that this instrument include a section focused on developmen t of an individual pa in profile for use in children with ASD were considered. This is similar to how Solodiuk and Curley (2003), as previously described, approach pain assessmen t in nonverbal children with severe cognitive impairment. Their instrument provides examples of pain expressive behaviors commonly seen in nonverbal children with cognitive impairment, and then asks caregivers to work from the suggested list to generate a pain expression profile unique to their child at various intensities of pain. Given the heterogeneity seen in children with ASD, it may be helpful to incorporate some of what Solodiuk and Curley (2003) have found us eful. Perhaps this assessment tool could be comprised of two parts. An initial part (Par t A) might recommend that caregivers watch for particular (those most common) pain expressive behaviors. This would be especially helpful for newly diagnosed children, or to caregivers who are watching for more t ypical pain expressive
108 indicatorsand thus concluding th at their children do not feel pain. A second part (Part B) will allow caregivers to begin to determine their childs own pain profile. Once a comprehensive picture of how the child expresses physical pain is established, th is tool can be distributed to other individuals involved in his or her care. In this way, the tw o methods of pain assessment are merged. Suggestions of what to watch for are of fered, however it is also acknowledged that each child is different and unique in his or her pain expression. The Role of Direct Behavioral Obs ervation in Instrument Development The original item list generated for Study 2 ai med to include a wide variety of possible pain indicators in children with ASD. When the items were reviewed by experts, the role of direct observation in the instruments developm ent, and subsequent use with children was questioned and is therefore addressed here. It was acknowledged that several of the pain expressive indicators under review by experts are not directly obser vable, and require one to rely on caregiver reports of deviation from normal. These items include: unhappiness, acting off according to caregiver, and crankiness. Two were considered highly relevant to pain assessment in this population, and therefore clarification regardi ng the definition and operationalization of these items is necessary before re-evaluati ng the item list. This research illuminated the heavy emphasis on deviations from normal when assessing whether this population of children is in pain. While there we re several objective, observable, and measurable indicators presumed to be relevant to pain assessment in ASD (i.e., furrowed brow, banging his/her head, i njuring oneself, grimacing, guarding, and increased heart rate), many other identified indicators were subjective, and require a) knowledge of the childs baseline, and b) monitoring for changes from normal (i.e., crankiness, being less active, rocking unusually, acting off, irritability). Given this, there might need to be sections of the pain assessment tool that are objective a nd quantifiablethese coul d be used by someone
109 unfamiliar with the childwhile ot her sections of the final assessment tool might be based on caregiver judgments regarding wh at is normal in their child. Caregiver information regarding their childs baseline, and thus deviations from normal will be critical to comprehensive pain assessment in this population. Future research in to whether caregivers subjective information correlates with objective, measurab le indicators is needed. In this way, direct observation of children would be extremely valuable. When beginning to explore pain assessment in children with ASD, this Investigator considered study designs that inco rporated direct observation of ch ildren presumed to be in pain, in order to inductively generate a list of observable beha viors potentially relevant to pain assessment in this population. The confusion that surrounded pain assessment in children with ASD however, made it difficult to theoretically determine what to watch for when observing this population. The reports that these children do not react to pain in the same manner as typically developing children hi ghlighted the need for initial pilot work aimed to simply characterize the issues en countered when assessing pain in this population. Thus the studies conducted in this research gathered descriptive information in order to deductively determine what types of pain expressive behaviors one would watch for when directly observing children. The next steps in this instruments development will certainly incorporate direct observation, and attempt to correlat e caregiver reports with certain pain expressive behaviors that are directly observable to someone unfamiliar wi th the child. Direct observation without any informed information regarding how children with ASD express pain would have been limited in its usefulness. Next Steps in this Instruments Development This instrument developm ent process began by a ttempting to articulate a) if children with ASD experience and express pain differently than typically developing child ren and b) whether it
110 is possible to identify a set of pain expressi ve behaviors unique to this population. Study 1 findings indicate that children with ASD are capable of experiencing and expressing pain, however their pain expression is unique. Study 2 validates this uniqueness by identifying pain indicators specific to children with ASD, and by de monstrating that indicato rs relevant to other, similar, populations of children are not considered to be relevant to this population. Following Study 2, the resulting list of pain expressive behaviors deemed rele vant to children in ASD needs to be further examined. Therefore, the next step s in this instruments development involve a) the revision of particular items, b) re-evaluation of the new set of items (having been revised and recategorized), c) determination of how each item s hould be scaled, and d) additional testing of the final set of items to determine whether the resulting pain assessment tool is valid and reliable in children with ASD. In formalizing the final se t of items into a scale, each pain indicator will need to have its operational definition further articulated. Future item analysis involves examining th e final instruments applicability to the population of interest. It will need to be determ ined whether caregivers can utilize the tool to discern pain in their children with ASD. The pur pose of follow up research will be to a) examine how the items relate to one anothe r, and how each item influences the overall statistics of the tool (internal consistency), b) conduct correlation analyses to examin e whether total-item (summed) scores are different for children with various characteristics within the ASD population, c) examine the endorsement frequencythe propor tion of caregivers w ho give each response option to an itemof each item, and d) analyze th e caregivers mean usefulness ratings. It will be helpful to know which items caregivers endorse, and how useful the items are in determining whether their children are in pain.
111 The next step following item analysis and adjust ment will be to further test the reliability and validity of the existing tool. Establishing construct validity will require a series of studies. Each additional study demonstrati ng that the instrument is useful further validates it as a pain assessment tool for the specified population. Th e purpose of a validation study is to determine whether the tool maintains adequate internal cons istency, to establish intrarater reliability, to examine concurrent and discriminant validity, and to determine whether caregivers can use it to discriminate between episodes of pain and nonpain. The most significant analysis for this particular study design will be to determine if the scores on the assessment tool differ between pain and nonpain episodes (discriminant validi ty). This repeated measures design requires Repeated Measures Analysis of Variance (RANOVA) The same caregivers will fill out the checklist during two separate pain events, and th erefore it will possible to examine the tools stability over time. It can be determined if the number of items observed is consistent, and whether the same items are endorsed for a child during two separate pain events. Mixed design RANOVA will allow one to examine whether there are differe nces between groups of subjects regarding a specific factor (i.e ., communication ability, age, etc) Finally, because the same caregivers will utilize the tool on two different oc casions, it will be possible to examine intrarater reliability in terms of caregiver use of the tool. Kappa coefficients analyzing the caregivers agreements when util izing the tool at each pain event will establish intrarater reliability. Streiner and Norman (1995) propos e that acceptable reliabi lities reported for instruments need to be greater than .75. Future studies will also incorporate video observation of children with ASD experiencing pain in perhaps a more clinical setting (post-operative pain fo r example) to establish other required forms of reliability (interrater) between different observers (observer, caregiver, staff
112 nurse, researcher). Additionally, correlating instru ment scores with perhaps facial coding could help further establish concurrent validitygiven that facial codi ng is a well-established valid method for assessing for pain in children as so mewhat of a gold standard. There are no behavioral observation protocol s well validated for assessing pain in children with ASD, although there are specific behavior s that have been validated as observable pain indicators in other populations (elders with dementia, for exam ple). Capturing pain behavior on video would allow for additional exploration into how children with ASD exhibit pain from the perspective of pain experts, rather than caregiver report. This will be useful if the caregivers have difficulty discerning pain in their children with ASD, even after being given a re pertoire of behaviors thought to indicate pain in this population. Limitations Given the paucity of inform ation available to begin to investigat e the general question approached in this research, these studies were preliminary in nature, and aimed to explore and describe issues important to assessing pain in children with ASD. Several limitations were identified as each study was conducted, and thes e include the survey response rate, and the quasi-qualitative design u tilized in Study 1. Survey Response Rate A total of 400 surveys w ere mailed to caregiv ers of children with ASD in Study 1. Eightyeight surveys were completed and returned by mail, resulting in a 22% return rate. This lower response rate is problematic because it influen ces the validity of infe rring that the sample represents the population (H ulley, Newman, & Cummings, 2001, p. 31). However, a lower response rate might be expected in this par ticular population given th e tremendous challenges associated with caring for a child with this disorder. Many of these care givers are overwhelmed by the responsibilities required to ensure that th eir children receive the best intervention and care
113 possible. A better response rate might have b een achieved if an incentive was included in the study design. Of those caregivers who did resp ond, almost all include d written descriptive responses, and 80% provided their name and numb er for a follow up call. This willingness to provide detailed information highlighted the strong desire to discuss and share concerns regarding pain assessment in their children with others. Quasi-Qualitative Design Study 1 afforded caregivers the opportunity to provide written desc riptions of their thoughts regarding the pain experi ence in their children. These wr itten responses were exam ined by the investigator and an autism expert and co mmonly reported issues were tallied. The most frequently reported issues were ut ilized to identify a few themes amongst this descriptive data. If the focus of this research was to specifically describe how caregivers perceived the pain experience in their children, a purely qualitative study would have been informing. However, the goal of Study 1 was to simply establish if pain assessment in children with ASD was a problem. Nothing in the literatur e to date addressed whether pa in assessment in ASD warrants an investigation. Study 1 was originally designed to obtain a snapshot of whether caregivers and healthcare providers would benefit from research focused on pain assessment in ASD. The sheer volume of written descriptive information provi ded by caregivers was surprising. Given this volume of descriptive data, more rigorous and formal qualitative inquiry might have been even more revealing. Other studies attempting to develop pain asse ssment tools for unique populations utilize caregivers to draft an in itial item list, as done for this rese arch. Some common methods used to solicit caregiver perspectives incl ude having them participate in focus groups, fill out checklists, and answer survey questions. This study was de signed in a similar fashion, foregoing the option
114 to conduct an in-depth qualitative study, and c hoosing instead to complete two preliminary studies aimed at facilitating instrume nt development for this research. Nursing Implications and Clinical Significance As m embers of dynamic and inte rdisciplinary health care teams providing patient care, nurses are often directly responsible for assessi ng pain and initiating means to relieve patient suffering. In the absence of self-report, it is of ten the astute assessment of the subtleties in a patient that allow a good nurse to identify, interven e, and relieve pain. Th is awareness is even more critical when assessing and managing pain in several vulnerable populations of children. As the need for better pediatri c pain assessment emerged in the 90s, Ferrell, McCaffery, & Ropchan (1992) surveyed clinical nurses and identified barriers to adequate pain management. Top barriers included lack of knowledge on the part of the physicians, nurses, patient and family members, and system issues involving administration of pain medication. An effort has been made over the past several decades to increa se nursing knowledge of pain assessment and management in children, and to prioritize pain assessment and management in the clinical setting. This nursing research aimed to equip parent s and healthcare providers with the additional knowledge and understanding required to determine if and when a child with ASD might be experiencing pain. In pursuing the answers to the proposed research questions, it was the hope that many of the myths and controversial issu es surrounding the questi on of whether children with ASD are capable of feeling pain can be elim inated. Not only are nurs es directly involved in patient care management, but nur ses cross a myriad of settings and function at various practice levels. Dissemination of these research findings will equip nurses at the bedside, and nurses in the clinic setting, with the ability to better asse ss pain in children with ASD. Furthermore, the perceptions that children with ASD are incapabl e of feeling pain can be corrected. Nurses
115 function as clinicians, advocates, and educator s. An accurate understanding of how children with ASD experience pain will ensure that pare nts of newly diagnosed children, or parents having a difficult time determining when their chil dren with ASD are in pain, are supported and taught about pain assessment and management in their children. Future Directions Instrum ent development requires diligence and dedication. A series of studies are required to ensure that a valid and reliable means of assess ing the construct of intere st results. The studies conducted in this research lay a foundation from which future instrument development can build. Future research aims to continue refining pain expressive indicators specific to children with ASD. As these indicators are further articulated, additional investigation into the variability in pain expression found along the spectrum is warrante d. It is likely that there are subgroups of children with ASD who have par ticular features infl uencing their specific pain expressive abilities (i.e., those with seve re abnormal sensory processing). The Communications Model of Pain (Had jistravopoulos & Crai g, 2002) highlights a variety of relationships that need to be explored. This resear ch demonstrated that caregiver perceptions, and existing myths or biases, likely influence ones abil ity to decode pain expressive behaviors in children with ASD. By focusing on the nonverbal and involuntary mechanisms of pain expression, it is possible to determine how children with ASD behaviorally exhibit their physical discomfort. Additional investigation in to how children with ASD respond to different types of pain is still neede d. It is possible that acute pain elicits a more reflexive, involuntary response, while everyday bumps and bruises (or mo re chronic types of pa in) require a greater understanding of the social c ontext surrounding the pain expe rience. This requirement (understanding the social context) would be impair ed in a population of children with social and communicative impairments (i.e., ASD).
116 Additionally, future work will need to focus on caregivers abilities to accurately decode their childs pain expressive behavior, and examine whether their knowledge is appreciated and utilized by healthcare providers ac ross settings. Complaints that healthcare providers dismiss caregiver beliefs about their ch ildrens pain are alarming, and sp eak to a lack of partnership between some healthcare providers a nd caregivers in this population. Finally, work to disseminate findings follo wing the ultimate development of a pain assessment tool available for use in children wi th ASD will include examining the applicability of this tool to children in a va riety of settings. Hopefully, illumi nating the pain assessment issues specific to children with ASD will debunk circulating myths, and allow everyone involved in the care of a child with ASD to adequately assess, treat, and manage their pain.
117Table 5-1. Comparison of expert-identified pa in indicators for use in ASD with pain indicators from three existing pain assessm ent tools Study 2: Items identified as relevant to pain assessment in ASD by experts Non-Communicating Childrens Pain Checklist (NCCPC); 30 Items Pain Indicator for Communicatively Impaired Children (PICIC); 6 Items Pain Evaluation Scale for CP; 10 Items Primary pain indicators (I-CVI > .78) Rocking unusually Lashing/striking out Guarding (protecting) or favoring a part of the body when moving Protecting, favoring, guarding part of body that hurts Coordinated defensive reaction of an area supposed painful protection of painful areas Increased heart rate Tensing face Screwed up or distressed looking face Painful expression Furrowing brow Furrowed brow Screwed up or distressed looking face Painful expression Grimacing Screwed up or distressed looking face Painful expression Screaming/yelling/throwing tantrums Screaming/yelling Screaming, yelling, groaning or moaning Injuring oneself Banging his/her head Disinterest in surroundings Lying around/acting lethargic Being less active Not moving, less active, Quiet Crankiness (being more cranky than usual) Not cooperating, cranky, irritable, unhappy
118 Table 5-1. Continued Study 2: Items identified as relevant to pain assessment in ASD by experts Non-Communicating Childrens Pain Checklist (NCCPC) Pain Indicator for Communicatively Impaired Children (PICIC) Pain Evaluation Scale for CP Inconsolability (unable to c onsole) Difficult to distract, notable to satisfy or pacify Difficult to comfort or console Secondary pain indicators (I-CVI = .67) Acting off according to caregiver: not acting like him/her self Less interaction with others, withdrawn Flinching or moving body part away, sensitive to touch Flinches or moves away if touched Stiff, spastic, tense, rigid Body appears stiff or tense Tonic troubles: Stiffness Floppy Crying (with or without tears) Crying (moderately loud) Crying (with or without tears) Crying Moaning, whining, whimpering (softly) Moaning, whining Whimpering Moaning or inaudible cries Stating/repeating a specific pain-related word (i.e., hurt, ouch) Exhibits a specific word/sound for pain (or type of laugh) Does not communicate with others Moving body in a specific way to show pain Searches for an unusual position that calms Seeking comfort or physical touch Seeking comfort or physical closeness Touching, rubbing, picking, and/or tapping a specific part of the body Gesturing to or touching part of body that hurts Shivering
119 Table 5-1. Continued Study 2: Items identified as relevant to pain assessment in ASD by experts Non-Communicating Childrens Pain Checklist (NCCPC) Pain Indicator for Communicatively Impaired Children (PICIC) Pain Evaluation Scale for CP Changing color, (flushing/redness) Change in color, (pallor) Sweating, perspiring Producing tears (without overt crying or vocalizations) Tears Increased rate of breathing Sharp intake of breath, Gasping Breath holding Eating less, not interested in food Increase in sleep Sleeping less Decrease in sleep Grinding teeth, clenching jaw down Clenching or grinding teeth, chewing or thrusting out tongue Frowning, turning down mouth Tu rning down of mouth Pouting, quivering, puckering lip s Lips puckering up, tight, pouting or quivering Widening eyes Change in eyes Becoming more aggressive (hitting, pinching, biting others) Acting angry Holding his/her head with both hands for prolonged periods of time Jumping around, agitated, Fidgety Increase in spontaneous movement (voluntary/not) Note: Primary and Secondary Pain Indicators are compared to item s included in existing pain assessment tools for use in related, but different, populations of child ren. Italicized items indicate that items are either unique to children with ASD or not relevant to children with ASD (because they were not validated in Study 2 by autism experts for use in this population).
120Table 5-2. Study 2: Variability in expert-rated relevance amongs t all items with an I-CVI = .67 Pain behavior/indicator Not relevant Somewhat relevant Quite relevant Highly relevant Item CVISum Sleeping less 1 1 4 .67 15 Irritability (being more irritable than usual) 2 4 .67 16 Seeking comfort/physical touch 2 4 .67 16 Becoming more aggressive (hitting, pinching, biting others) 2 4 .67 16 Acting angry 2 4 .67 16 Widening eyes 1 1 3 1 .67 16 Frowning, turning down mouth 1 1 3 1 .67 16 Pouting, quivering, puckering lips 1 1 3 1 .67 16 Producing tears (without overt crying or vocalizations) 2 3 1 .67 17 Increased rate of breathing 2 3 1 .67 17 Acting off according to caregiver: not acting like him/herself 2 3 1 .67 17 Grinding teeth, clenching jaw down 2 3 1 .67 17 Squinching, squinting eyes 3 1 2 .67 17 Crying (with or without tears) 2 2 2 .67 18 Moaning, whining, whimpering (softly) 2 2 2 .67 18 Changing color (flushing, redness) 2 2 2 .67 18 Touching, rubbing, picking, and/or tapping a specific part of the body 2 2 2 .67 18 Holding his/her head with both hands for prolonged periods of time 2 2 2 .67 18 Stating/Repeating a specific pain-related word (i.e., hurt, ouch) 2 1 3 .67 19 Note: The numbers in each column indicate how many experts rated th e item as not, somewhat, quite, or highly relevant. The differences in sum score highlight the variability amongst a ll items with the same calculated I-CVI (.67).
121 APPENDIX A STUDY 1: CAREGIVER SU RVEY COVER LETTER EXPL AINING PARTICIPATION Dear Paren t or Guardian, I am a nursing student at the University of Florida, and I am conducting a research study exploring how children with auti sm experience and express pain. The purpose of the study is to understand the thoughts parents have about whether their child feels, and can communicate when he or she is in pain. Participation in this study invol ves completing the attached brief survey. It should only take about 5-10 minutes to answer the three questions. Your responses will be kept completely anonymous. Please do not write your name on the survey. All responses will be grouped together, and there will be no way to link your responses to your identity. The completed surveys can be placed in the sealed envelope and either dropped in the box located by the exit, or may be returned to the staff member who gave it to you. Sealing your survey in the envelope ensures that your responses will not be opened by anyone other than myself. Participation in this study is completely voluntar y. You do not have to fill out the survey. You may withdraw from the study at any time. If you choose not to participate, this will have no effect on the care or treatment of your child or fa mily. Only the investigators involved in this study have the legal right to review your res ponses, and I assure you that we will protect the confidentiality of your information. If results from the study are published, anonymity will be maintained. If you have questions about your rights as a resear ch participant, you may call the Institutional Review Board at 352-846-1494. I will not be able to pay you for participating, but it will not cost you anything to participate. Thank you for you time in reading this letter and thinking about completing the survey. If you have any questions, please call Melissa Dodd (352-392-9409), or ask the staff presenting you with this survey. Sincerely, Melissa Dodd, MSN, ARNP Doctoral Student College of Nursing, University of Florida
122 APPENDIX B STUDY 2: ITEM LIST FOR EXPERT REVIEW Instru ctions: Please review each item and determine whether it might be relevant for assessing pain in children with ASD (especial ly nonverbal children). Recall that these items will undergo further testing/validation, and do not reflect the final version of the pain assessment scale. You can simply place an x under the appropriate column below, re-save the co mpleted document with your name, and return to me via email. It is the goal of this process to obtain a number of items that might be re levant to this population by weeding out irrelevant indicators. If there are behaviors that need to be considered but are not on this list, feel free to suggest that they be added. Those items rated as relevant by the majority of expert reviewers will remain in the item pool for the next step in the development of this tool. The final list of items will be utilized to construct a more formal pain assessment tool which will undergo additional expert review fo r item style (wording, clarity), and then be distributed to a test sample to evaluate overall tool comprehens iveness (understanding instructions, response format, length, etc). Pain Behavior /Indicator NOT Relevant SOMEWHAT Relevant QUITE Relevant HIGHLY Relevant Vocal/Verbal Indicators Crying (with or without tears) Screaming/Yelling/Throwing Tantrums Stating/Repeating a specific painrelated word (i.e. hurt, ouch) Stating/Repeating a specific non-pain word or phrase (i.e. bat, toe) Babbling (Saying something incomprehensible) Moaning, whining, whimpering (softly) Humming/Singing Laughing Vocalizing less than normal, getting quiet Physiological Sweating, perspiring Producing tears (without overt crying or vocalizations) Shivering, shaking Increased rate of breathing Decreased rate of breathing Changing color (paleness)
123 Changing color (flushing, redness) Increased heart rate Fever Feeling cool or cold to touch Activity/Movement Sitting still Fidgeting, becoming more active, jumping around (motor restlessness) Muscle tensing, rigidity Guarding (protecting) or favoring a particular part of the body when moving Touching, rubbing, picking, and/or tapping a specific part of the body Seeming hypersensitive (overly sensitive) to touch Seeming less sensitive to touch Becoming floppy, or goes limp Being less active Lashing/Striking out Changing his/her usual body language Holding his/her head with both hands for prolonged periods of time Stomping foot Clapping hands loudly Rocking unusually Scratching a specific part of the body Banging his/her head Clapping hands loudly Behavioral Holding breath Injuring oneself Crankiness (being more cranky than usual) Cooperating less than normal Irritability (being more irritable than usual) Unhappiness Interacting less than normal with others Interacting more than normal with others Seeking comfort/physical touch Isolating oneself more, hiding Moving to, or staying in, a specific location or safe space
124 Perseverating more on a particular item Making up stories or reenacting a painful circumstance Acting off according to caregiver: not acting like him/herself Becoming more aggressive (hitting, pinching, biting others) Inconsolability (being unable to console) Loosing interest in things that are usually extremely important or interesting to child Increasing number and/or frequency of repetitive behaviors Acting angry Decreasing number and/or frequency of repetitive behaviors Daily Activities Sleeping less Sleeping more Eating less, not interested in food Eating more than usual Laying around, acting lethargic Facial Grinding teeth, clenching jaw down Chewing, moving tongue around and/or out, smacking lips Furrowing brow Squinching, squinting eyes Widening eyes Frowning, turning down mouth Pouting, quivering, puckering lips Tensing face Facial restlessness, tics Grimacing Lip biting Continuous eye blinking Rolling eyes Flaring nostrils Raising eyebrows Lip curling
125 Please feel free to offer comments, or recommend additional behavioral i ndictors that might not have been listed above. Again, thank you.
126 APPENDIX C STUDY 2: ITEM LIST COVER LETTER EXP LAINING STUDY AND PARTICIPATION Dear Participant, I am a doctoral candidate in Nursing, at the Univer sity of Florida, and am inviting you to participate because of your experience in caring for children w ith autism spectrum disorders. Thank you for taking the time to provide your expert insight regarding how children with autism spectrum disorder (ASD) behaviorally express physical pain. My research seeks to better understand how children with ASD express their physical pain, and aims to ultimately equip providers and caregivers with a tool that can be used to assess pain in this population, specifically in those children unable to ve rbally communicate their discomfort. Because of the lack of information regarding how children with ASD express pain, and because of the confusion that surrounds this issue (given various re ports of these children having a high tolerance for pain), it was determined that the development of a new pain assessment tool for this population is warranted; specifically, for those children with ASD who have severe communication and language impairments. Existing pain assessment tools have not been utilized and/or validated in this population, and there are limitations to using traditional pain assessment methods in children with ASD. Given the unique nature of this disorder, and the heterogeneity found in children on the spectrum, this population requires a novel approach to pain assessment. The development of this tool is in the early stages. Recent work aimed to generate a long list of items that might indicate pain in children with ASD who cannot verbally communicate well. The current item pool you will review presents this list of behaviors. Th is list has been thoughtfully derived from several sources: the pain and autism literature, one-to-one di scussion with several autism experts, semi-structured caregiver interviews, and caregiver surveys that asked about the pain experience in their children with autism. The next step in examining these theoretical items is to have a panel of experts review the potential indicators, and rate whether each one is releva nt to this population. This will ensure that the revised set of items (which will undergo further review via a large test sample in the next step of instrument development) achieves adequate content validity Your thought regarding whether each item should remain in the item pool at this stage in instru ment development is crucial to ensuring that all potential indicators have been considered, and included. Given the heterogeneous nature of this disorder, it is acknowledged that this (creating a universal pain assessment tool for this population) might not be pos sible, however it is believed that in attempting to gather information regarding how ch ildren with autism express pain, we can illuminate the fact that while some children might seemingly show an insensitivity to pain, many children on the spectrum do in fact feel and process pain, and therefore cannot be ig nored or dismissed regarding appropriate pain management. I sincerely appreciate your time, and am availabl e should you have additional comments, insights, or questions. Kindest Regards, Melissa Dodd Inglese, Do ctoral Candidate, ARNP University of Florida College of Nursing
127 LIST OF REFERENCES Akhtar, N. (1999). Acquiring basic word order: Evidence for data -driven learning of syntactic structure. Journal of Child Language, 26, 339. Allen, M. J., & Yen, W M. (1979). Introduction to measurement theory. Long Grove, IL: Waveland Press, Inc. Alligood, M. R., & Tomey, A. M. (2002). Introd uction to nursing theory : History, terminology, and analysis. In A. M. Tomey & M. R. Alligood (Eds.), Nursing theorists and their work (5th ed., pp. 3). St. Louis, MO: Mosby, Inc. American Academy of Pediatrics Committee on Bi oethics (1995). Informed consent, parental permission, and assent in pediatric practice. Pediatrics, 95 (2), 314. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (Revised 4th ed.). Washington, DC: Author. Anand, K. J. S. (1989). The neuroanatomy, neur ophysiology, and neurochemistry of pain, stress, and analgesia in newborns and children. Pediatric Clinics of North America, 36(4), 795 817. Anand, K. J. S., & Craig, K. D. (1996). New perspectives on the definition of pain. Pain, 67, 3 6. Ascension Health (2007). Healthca re ethics: Principle of informed consent. Retrieved July 14, 2007 from http://www.ascensionhealth.or g/ethics/public/key_principles/ informed_consent.asp Autism Society of America. (2006). Next steps: A guide for families new to autism. Retrieved June 13, 2007 from https://secure2.convio.net/asa/site/ Ecommerce?store_id=2221&JServS essionIdr011=km0qfshmu2.app26a Barbaresi, W. J., Katusic, S. K., Colligan, R. C., Weaver, A. L., & Jacobsen, S. J. (2005). The incidence of autism in Olmsted County, Minnesota, 1976: Results from a population-based study. Archives of Pediatric an d Adolescent Medicine, 159, 37. Barclay, L. (2005). Autism epidemic? A newsmaker interview with Morton Ann Gernsbacher, PhD, and Craig J. Newschaffer, PhD Retrieved June 12, 2007 from http://www.medscape.com/viewarticle/508429_print Baron-Cohen, S. (1995). Mindblindedness: An essay on autism and theory of mind. Boston: MIT Press/Bradforn Books.
128 Baron-Cohen, S., & Swettenham, J. (1997). Theo ry of mind in autism: Its relationship to executive function and central coherence. In D. J. Cohen & F. R. Volkmar (Eds.), Handbook of autism and pervasive developmental disorders (2nd ed., pp. 880). New York: John Wiley & Sons, INC. Baron-Cohen, S., Tager-Flusberg, H. & Cohen, D. J. (Eds.). (1993). Understanding other minds: Perspectives from autism. Oxford: Oxford University Press. Bassett, J. (2005). Young suffering. Advance for Directors in Rehabilitation, 14 (10), 40. Beauchamp, T. L., & Walters, L. (1999). Contemporary issues in bioethics (5th ed.). Belmont, CA: Wadsworth Publishing Company. Beauchamp, T. L., & Walters, L. (2003). Contemporary issues in bioethics (6th ed.). Belmont, CA: Wadsworth Publishing Company. Belschner, R. A. (2007). Stop, assess and motiv ate: The SAM approach to autism spectrum disorder. The American Journal for Nurse Practitioners, 11 (4), 43. Bowers, J. W., & Bradac, J. J. (1982). Issues in communication theory: A metatheoretical analysis. In M. Burgoon (Ed.), Communication yearbook 5 (pp. 1). New Brunswick, NJ: Transaction Books. Breau, L. M., Camfield, C., McGrath, P. J., Rosm us, C., & Finley, G. A. (2001). Measuring pain accurately in children with cognitive impairments: Refinement of a caregiver scale. The Journal of Pediatrics, 138, 721. Breau, L. M., Finley, G. A., McGrath, P. J., & Camfield, C. S. (2002). Validation of the noncommunicating childrens pain ch ecklist postoperative version. Anesthesiology, 96 528. Breau, L. M., McGrath, P. J., Camfield, C. S., & Finley, G. A. (2002). Psychometric properties of the non-communicating childrens pain checklist-revised. Pain, 99, 349. Breau, L. M., McGrath, P. J., Camfield, C. S., Rosmus, C., & Finley, G. A. (2000). Preliminary validation of an observational pain checklist for persons w ith cognitive impairments and inability to communicate verbally. Developmental Medicine & Child Neurology, 42, 609. Brown, N. J., Berkovic, S. F., & Scheffer, I. E. (2007). Vaccination seizures and vaccine damage. Current Opinion in Neurology, 20 181. Bruner, J. S. (1983). Childs talk: Learning to use language. NY: W. W. Norton.
129 Budwig, N., Wertsch, J. V., & Uzgiris, I. C. (2000). Introduction: Communication, meaning, and development: Interdisciplinary perspectives. In I. E. Sigel (Series Ed.) & N. Budwig, I. C. Uzgiris, & J. V. Wertsch (Vol. Eds.), Communication: An arena of development: Vol. 19. Advances in applied developmental psychology (pp. 1). Stamford, CT: Ablex Publishing Company. Bursch, B., Ingman, K., Vitti, L., Hyman, P., & Zeltz er, L. K. (2004). Chroni c pain in individuals with previously undiagnosed autistic spectrum disorders. The Journal of Pain, 5 (5), 290 295. Carpenter, M., & Tomasello, M. (2000). Jo int attention, cultural learning, and language acquisition: Implications for children with au tism. In S. F. Warren & J. Reichle (Series Eds.) & A. M. Wetherby & B. M. Prizant (Vol. Eds.), Communication and language intervention series: Vol. 9: Autism spectru m disorders: A transactional developmental perspective. (pp. 31). Carter, B., McArthur, E., & Cunliffe, M. (2002). Dealing with uncertainty: Parental assessment of pain in their children with profound special needs. Journal of Advanced Nursing, 38(5): 449 Centers for Disease Control and Prevention. (1999). Thimerosal in vaccines: A joint statement of the American Academy of Pediatrics and the Public Health Service. MMWR, 48, 563 565. Centers for Disease Control and Prevention. (2000). Notice to r eaders: Summary of the Joint Statement on Thimerosal in Vaccines. MMWR, 49, 622. Centers for Disease Contro l and Prevention. (2007). How common are autism spectrum disorders (ASD)? Retrieved March 3, 2007, from http://www.enotalone.com/article/6965.html Champion, G. D., Goodenough, B., von Baeyer, C. L., & Thomas, W. (1998). Measurement of pain by self-report. In Internat ional Association for the Study of Pain (Series Ed.) & G. A. Finley & P. J. McGrath (Vol. Eds.), Progress in pain research and management: Vol. 10. Measurement of pain in infants and children (pp. 123) Seattle, WA: International Association for the Study of Pain Press. Charman, T. (2003). Why is joint attention a pivotal skill in autism? Philosophical Transactions of the Royal Society of London. Se ries B, Biological Sciences, 358, 315. Charman, T., & Baird, G. (2002). Practitioner review : Diagnosis of autism spectrum disorder in 2and 3-year-old children. Journal of Child Psychology and Psychiatry, 43 (3), 289. Cicchetti, D. V., & Sparrow, S. (1981). Developing criteria for establishing interrater reliability of specific items: Application to assessment of adaptive behavior. American Journal of Mental Deficiency, 86, 127.
130 Collignon, P., & Giusiano, B. (2001). Validation of a pain evaluation scale for patients with severe cerebral palsy. European Journal of Pain, 5, 433. Craig, K. D. (1998). The facial display of pain. In International A ssociation for the Study of Pain (Series Ed.) & G. A. Finley & P. J. McGrath (Vol. Eds.), Progress in pain research and management: Vol. 10. Measurement of pain in infants and children (pp. 103) Seattle, WA: International Association for the Study of Pain Press. Crocker, L., & Algina, J. (1986). Introduction to classical and modern test theory Australia: Wadsworth Group. Dance, F. E. X. (1970). The concept of communication. Journal of Communication, 20, 201 210. Dawson, G., Toth, K., Abbott, R., Osterling, J., M unson, J., Estes, A. et al. (2004). Early social attention impairments in autism: Social orienting, joint atten tion, and attention to distress. Developmental Psychology, 40 (2), 271. Dawson, G., Webb, S., Schellenberg, G. D., Dager, S., Friedman, S., Aylward, E., et al. (2002). Defining the broader phenotype of autism: Ge netic, brain, and beha vioral perspectives. Development and Psychopathology, 14, 581. Decker, S. A., & Perry, A. G. (2003). The develo pment and testing of th e PATCOA to assess pain in confused older adults. Pain Management Nursing, 4 (2), 77. DeVellis, R. F. (2003). Scale development: Theory and applications (2nd ed.). Thousand Oaks, CA: Sage Publications. Deyo, K. S., Prkachin, K. M., & Mercer, S. R. (2004). Development of se nsitivity to facial expression of pain. Pain, 107, 16. Dover, C. J., & Le Couteur, A. (2007). How to diagnose autism. Archives of Disease in Childhood, 92, 540. Fawcett, J. (2005). Criteria for evaluation of theory. Nursing Science Quarterly, 18 (2), 131. Ferrell, B. R., McCaffery, M., & Ropchan, R. (1992) Pain management as a clinical challenge for nursing administration. Nursing Outlook, 40 (60), 263. Ferrel, B. R., Novy, D., Sullivan, M. D., Banja, J., Dubois, M. Y., Gitlin, M. C., et al. (2001). Ethical dilemmas in pain management. The Journal of Pain, 2 (3), 171. Filipek, P.A., Accardo, P. J., Baranek, G. T., Cook, E. H., Jr., Dawson, G., Gordon, B., et al. (1999). The screening and diagnosis of autistic spectrum disorders. Journal of Autism and Developmental Disorders, 29 (6), 439.
131 Finley, G. A., & McGrath, P. J. (1998). Intr oduction: The roles of measurement in pain management and research. In International A ssociation for the Study of Pain (Series Ed.) & G. A. Finley & P. J. McGrath (Vol. Eds.), Progress in pain research and management: Vol. 10. Measurement of pain in infants and children (pp. 1) Seattle, WA: International Association for the Study of Pain Press. Fodor, J. A. (1983). The modularity of mind. Cambridge, MA: MIT Press. Fombonne, E. (2001). Is there an epidemic of autism? Pediatrics, 107 411. Fombonne, E. (2003). Epidemiological surveys of autism and other pervasive developmental disorders: An update. Journal of autism and Developmental Disorders, 33, 365. Fombonne, E. (2005). Epidemiological studies of pervasive developmental disorders. In F. R. Volkmar, R. Paul, A. Klin, & D. Cohen (Eds.), Handbook of autism and pervasive developmental disorders: Vol. 1. Diagnosis development, neurobiology, and behavior (3rd ed., pp. 42). Hoboken, NJ: John Wiley & sons, Inc. Francois, G., Duclos, P., Margolis, H., Lavanchy, D., Siegrist, C., Meheus, A., et al. (2005). Vaccine safety controversies and th e future of vaccination programs. The Pediatric Infectious Disease Journal, 24 (11), 953. Freitag, C. M. (2007). The genetics of autistic diso rders and its clinical re levance: A review of the literature. Molecular Psychiatry, 12 2. Frith, U. (1989). Autism: Explaining the enigma. Oxford: Blackwell. Frith, U. (2003). Autism: Explaining the enigma (2nd ed.). United Kingdom: Blackwell Publishing. Gaffney, A., McGrath, P., & Dick, B. (2003). Meas uring pain in childr en: Developmental and instrument issues. In N. L. Schechte r, C. B. Berde & M. Yaster (Eds.), Pain in infants, children, and adolescents (2nd ed., pp. 128). Philadelphia, PA: Lippincott Williams & Wilkins. Gilbert-MacLeod, C. A., Craig, K. D., Rocha, E. M., & Mathias, M. D. (2000). Everyday pain responses in children with and without developmental delays. Journal of Pediatric Psychology, 25 (5), 301. Gernsbacher, M. A., Dawson, M., & Goldsmith, H. H. (2005). Three reasons not to believe in an autism epidemic. Current Directions in Ps ychological Science, 14 (2), 55. Giusiano, B., Jimeno, M. T., Collignon, P., & Chan, Y. (1995). Utilization of a neural network in the elaboration of an evaluation sc ale for pain in cerebral palsy. Methods of Information in Medicine, 34, 498. Grant, J. S., & Davis, L. L. (1997). Selecti on and use of content experts for instrument development. Research in Nursing & Health, 20, 269.
132 Hadden, K. L., & von Baeyer, C. L. (2002). Pain in children with cerebral palsy: Common triggers and expressive behaviors. Pain 99 (1), 281. Hadjistavropoulos, T. (2004). Ethics for psychologists who treat, a ssess, and/or study pain. In T. Hadjistavropoulos & K. D. Craig (Eds.), Pain: Psychological perspectives (pp. 327 344).Mahwah, NJ: Lawrence Erlbau m Associates, Publishers. Hadjistavropoulos, T., & Craig, K. D. (2002). A theoretical fr amework for understanding selfreport and observational measures of pain: A communications model. Behavior Research and Therapy, 40 551. Hadjistavropoulos, T., Craig, K. D., & Fuchs-L acelle, S. (2004). Social influences and the communication of pain. In T. Had jistavropoulos & K. D. Craig (Eds.), Pain: Psychological perspectives (pp. 87) Mahwah, NJ: Lawrence Erlbaum Associaties, Publishers. Happe, F. (1996). Studying weak central coherence at low levels: Children with autism do not succumb to visual illusions. Journal of Child Psychology and Psychiatry, 37 873. Happe, F., Ronald, A., & Plomin, R. (2006). Time to give up on a single explanation for autism. Nature Neuroscience, 9(10), 1218. Harrington, J. W., Patrick, P. A., Edwards, K. S ., & Brand, D. A. (2006). Parental beliefs about autism: Implications for the treating physician. Autism, 10(5), 452. Haynes, S. N., Richard, D. C. S., & Kubany, E. S. (1995). Content valid ity in psychological assessment: A functional approach to concepts and methods. Psychological Assessment 7(3), 238. Helseth, S., & Slettebo, A. (2004). Research involving children: Some ethical issues. Nursing Ethics, 11 (3), 298. Hennequin, M., Morin, C., & Feine, J. S. (2000). Pain expression and stimulus localization in individuals with Downs syndrome. Lancet, 356(9245), 1882. Herbert, M. R. (2005). Autism: A brain disorder, or a disorder that affects the brain? Clinical Neuropsychiatry, 2 (6), 354. Hill, E. L., & Frith, U. (2003). Understand ing autism: Insights from mind and brain. Philosophical Transactions of the Royal Society of London, 358, 281. Hoff, E. (2005). Language development (3rd ed.). Belmont, CA: Wadsworth Thomson Learning. Howlin, P. (1998). Practitioner review: Psychologi cal and educational treatments for autism. Journal of Child Psychology and Psychiatry, 39(3), 307. Hughes, C. (1998). Finding your marbles: Does preschoolers strategic be havior predict later understanding of mind? Developmental Psychology, 34 (6), 1326.
133 Hulley, S. B., Cummings, S. R., Browner, W. S., Grady, D., Hearst, N., & Newman, T. B. (2001). Designing clinical research (2nd ed.). Philadelphia, PAL Lippincott Williams & Wilkins. Hunt, A., Mastroyannopoulou, K., Goldman, A., & Seers, K. (2003). Not knowing: The problem of pain in children with severe neurological impairment. International Journal of Nursing Studies, 40 (2), 171. Institute of Medicine. (2001). Immunization safety review: me asles-mumps-rubella vaccine and autism Retrieved July 2, 2007 from http: //books.nap.edu/catalog.php?record_id=10101 Institue of Medicine. (2004). Immunization safety revi ew: vaccines and autism Retrieved July 2, 2007 from http://books.nap.edu/catalog/10997.html International Association for the Study of Pa in (2007). IASP pain terminology. Retrieved April 25, 2007 from http://www.iasp-pain.org/AM/T emplate.cfm?Section=Home&template=/ CM/HTMLDisplay.cfm&ContentID=3088#Pain Jarrold, C., Butler, D. W., Cottington, E. M., & Ji menez, F. (2000). Linking theory of mind and central coherence bias in autis m and in the general population. Developmental Psychology, 36(1), 126. Johnston, C. C., Stevens, B. J., Boyer, K., & Porter, F. L. (2003). Development of psychologic responses to pain and assessment of pain infant s and toddlers. In N. L. Schechter, C. B. Berde & M. Yaster (Eds.), Pain in infants, children, and adolescents (2nd ed., pp. 105 127). Philadelphia, PA: Lippincott Williams & Wilkins. Kanner, L. (1943). Autistic distur bances of affective contact. Nervous Child, 2, 217. King, I. M. (1981). A theory for nursing: Systems, concepts, process. Albany, NY: Delmar Publishers Inc. Klin, A., McPartland, J., & Volkmar, F. R. (2005). Asperger syndrome. In F. R. Volkmar, R. Paul, A. Klin, & D. Cohen (Eds.), Handbook of autism and perv asive developmental disorders: Vol. 1. Diagnosis, development neurobiology, and behavior ( 3rd ed., pp. 88 125). Hoboken, NJ: John Wiley & Sons, Inc. Kolevzon, A., Gross, R., & Reichenberg, A. (200 7). Prenatal and perina tal risk factors for autism: A review and integration of findings. Archives of Pediatric and Adolescent Medicine, 161 326. Kuhl, P. K. (2000). A new view of language acquisition. Proceedings of the National Academy of Sciences, 97(22), 11850. Kuhl, P. K. (2005). Links between social and linguistic processing of speech in preschool children with autism: Behavioral and electrophysiological measures. Developmental Sciences, 8 (1), F1F12.
134 Kutscher, M. L. (2006). Autistic spectrum disorders: Sorting it out. Retrieved June 28, 2007 from http://pediatricneurology.com/autism.htm#Asperger%E2%80%99s%20Syndrome Leeds-Hurwitz, W. (1989). Communicat ion as multichannel behavior. In Communication in everyday life (pp. 101). Norwood, NJ: Ablex Publishing Company. Lord, C., & Paul, R. (1997). Language and communica tion in autism. In D. J. Cohen & F. R. Volkmar (Eds.), Handbook of autism and pervasive developmental disorders (2nd ed., pp. 195). New York: John Wiley & Sons, INC. Lynn, M. R. (1986). Determination and quantification of content validity. Nursing Research, 35, 382. Marcus, G., & Rabagliati, H (2006) What developmental disorders can tell us about the nature and origins of language. Nature neuroscience, 9 (10), 1226. Martin, I., & McDonald, S. (2003). Weak coherence, no th eory of mind, or executive dysfunction? Solving the puzzle of pragmatic language disorders. Brain and Language, 85, 451. McCaffery, M., & Beebe, A. (1989). Pain: Clinical manual for nursing practice. Philadelphia, PA: Mosby. McGrath, P. J. (1998). Behavioral measures of pa in. In International Asso ciation for the Study of Pain (Series Ed.) & G. A. Finley & P. J. McGrath (Vol. Eds.), Progress in pain research and management: Vol. 10. Measurement of pain in infants and children (pp. 83) Seattle, WA: International Association for the Study of Pain Press. McGrath, P. J., Rosmus, C., Canfield, C., Cam pbell, M. A., & Hennigar, A. (1998). Behaviors caregivers use to determine pain in non-verbal, cognitively impaired individuals. Developmental Medicine & Child Neurology, 40 340. Meltzoff, A. N. (1999). Origins of theo ry of mind, cognition, and communication. Journal of Communication Disorders, 32 251. Melzack, R., & Wall, P. D. (1965). Pain mechanisms: A new theory. Science, 150, 971. Mercer, L., Creighton, S., Holden, J. J. A., & Lewi s, M. E. S. (2006). Parental perspectives on the causes of an autism spectrum disorder in their children. Journal of Genetic Counseling, 15(1), 41. Merskey, H. (1991). The definition of pain. European Journal of Psychiatry, 6, 153. Mitchell, S., Brian, J., Zwaigenbaum, L, Roberts, W., Szatmari, P., Smith, I., et al. (2005). Early language and communication development of infants later diagnosed with autism spectrum disorder. Developmental and Behav ioral Pediatrics, 27 (2), S69S78.
135 Morgan, B., Maybery, M., & Durkin, K. (2003). W eak central coherence, poor joint attention, and low verbal ability: Indepe ndent deficits in early autism Developmental Psychology, 39(4), 646. Muhle, R., Trentacoste, S. V., & Rapi n, I. (2004). The genetics of autism. Pediatrics, 113(5), e472e486. Retrieved June 30, 2007 from http://www.pediatrics.org/cgi/content/full/113/5/e472 Murch, S., Anthony, A., Casson., D., Malik, M., Be rlowitz, M., Dhillon, A. et al. (2004). Retraction of an interpretation. Lancet, 363 750. National Institute of Mental Health (2007). Autism spectrum disorders (pervasive developmental disorders). Retrieved June 11, 2007, from http://www.nimh.nih.gov/publicat/autism.cfm Nader, R., Oberlander, T. F., Chambers, C. T ., & Craig, K. D. (2004). Expression of pain in children with autism. Clinical Journal of Pain, 20 (2), 88. Neimark, J. (2007). Autism: Its not just in the head Retrieved June 30, 2007 from http://discovermagazine.com/2007/apr/autismit2019s-not-just-in-the -head/article_print Newschaffer, C. J., Falb, M. D., & Gurney, J. G. (2005). National autism prevalence trends from United States special education data. Pediatrics, 115 (3), e277e282. Retrieved June 12, 2007 from www.pediatrics.org /cgi/doi/10.1542/peds.2004-1958 Oberlander, T. F., & Craig, K. D. (2003). Pain an d children with developm ental disabilities. In N. L. Schechter, C. B. Berde, & M. Yaster (Eds.), Pain in infants, children, and adolescents ( 2nd ed., pp. 599). Philadelphia: Li ppincott Williams & Wilkins. ONeill, M., & Jones, R. S. P. (1997). Sensory-perceptual abnormalities in autism: A case for more research. Journal of Autism and Developmental Disorders, 27 (3), 283. Ozonoff, S. (1995). Executive functions in auti sm. In E. Schopler & G. B Mesibov (Eds.), Learning and cognition in autism (pp. 199). New York: Plenum Press. Ozonoff, S., & Jensen, J. (1999). Brief report: Specific executive func tion profiles in three neurodevelopmental disorders. Journal of Autism and Developmental Disorders, 29 (2), 171. Parse, R. R. (2000). Obfuscating: Th e persistent practice of misnaming. Nursing Science Quarterly, 13 (2), 91. Peterson, S. J. (2004). Introduction to the nature of nursing knowledge. In S. J. Peterson & T. S. Bredow (Eds.), Middle range theories: Application to nursing research (pp. 3). Philadelphia: Lippincott Williams & Wilkins. Polit, D. F., & Beck, C. T. (2006). The content validity index: Are you sure you know whats being reported? Critique and recommendations. Research in Nursing & Health, 29, 489 497.
136 Polit, D. F., Beck, C. T., & Owen, S. V. (2007). Is the CVI an acceptable indicator of content validity? Appraisal and recommendations. Research in Nursing & Health, 30, 459. Portney, L. G., & Watkins, M. P. (2000). Foundations of clinical re search: Applications to practice (2nd ed.). Upper Saddle River, NJ: Prentice Hall Health. Pratt, C., & Buckman, S. (2007). Supporting appropriate behavior in students with aspergers. Retrieved June 13, 2007 from https://secure2.convio.net/asa/site/ Ecommerce?store_id=2221&JServS essionIdr011=km0qfshmu2.app26a Rimland, B. (1964). Infantile autism: The syndrome and its im plications for a neural theory of behavior. Englewood Cliffs, NJ: Prentice-Hall Inc. Roth-Isigkeit, A., Thyen, U., Stoven, H., Schwar zenberger, J., & Schmucker, P. (2005). Pain among children and adolescents: Restrictions in daily living and triggering factors. Pediatrics 115 (2), e152e162. Retrieved March 19, 2007 from http://www.pediatrics.org/cgi/content/full/115/2/e152 Saffran, J. R., Aslin, R. N., & Newport, E. L. (1996). Statistical learning by 8-month-old infants. Science, 274, 1926. Silverman, C., & Brosco, J. P. (2007). Unders tanding autism: Parents and pediatricians in historical perspective. Archives of Pediatric and Adolescent Medicine, 161 392. Smith, D. R., & Williamson, L. K. (1977). Traditions. In Interpersonal communication (pp. 27 32). Dubuque, IA: William. C. Brown. Snow, C. E. (1999). Social pers pectives on the emergence of la nguage. In B. MacWhinney (Ed.), The emergence of language (pp. 257). Mahwah, NJ: Erlbaum. Solodiuk, J., & Curley, M. A. Q. (2003). Pain assessment in nonverbal children with severe cognitive impairments: Individualiz ed numeric rating scale (INRS). Journal of Pediatric Nursing, 18(4), 295. Stallard, P., Williams, L., Lenton, S., & Velleman, R. (2001). Pain in cognitively impaired, noncommunicating children. Archives of Disease in Childhood 85 460. Retrieved March 19, 2007 from http://adc.bmj.com/cgi/content/full/85/6/460 Stallard, P., Williams, L., Velleman, R., Lenton, S., & McGrath, P. J. (2002). Brief report: Behaviors identified by caregivers to de tect pain in noncommunicating children. Journal of Pediatric Psychology, 27 (2), 209. Stevens, B. (1998). Composite measures of pain In International Association for the Study of Pain (Series Ed.) & G. A. Finley & P. J. McGrath (Vol. Eds.), Progress in pain research and management: Vol. 10. Measurement of pain in infants and children (pp. 161) Seattle, WA: International Association for the Study of Pain Press.
137 Stevens, B. J., McGrath, P., Yamada, J., Gi bbins, S., Beyene, J., Br eau, L. et al. (2006). Identification of pain indicators for infants at risk for neurological impairment: A Delphi consensus study. BMC Pediatrics, 6 (1). Retrieved March 28, 2008, from http://www.biomedcentral.com/1471-2431/6/1 Streiner, D. L., & Norman, G. R. (1995). Health measurement scales: A practical guide to their development and use (2nd ed.). New York: Oxford University Press. Sunil, K. (2006). Autism: A review for family physicians. Indian Journal of Medical Sciences, 60(5), 205. Szpir, M. (2006). Tracing the origins of autism: A spectrum of new studies [Electronic version]. Environmental Health Perspectives, 114 (7), A412A418. Taylor, B. (2006). Vaccines and th e changing epidemiology of autism. Child: Care, health, and development, 32 (5), 511. Terstegen, C., Koot, H. M., de Boer, J. B., Tibboe l, D. (2003). Measuring pain in children with cognitive impairment: Pain response to surgical procedures. Pain, 103(1), 187. Tomasello, M. (1992). The social bases of language acquisition. Social Development, 1, 67. Tomasello, M. (2003). Constructi ng a language: A usage-based theory of language acquisition. Cambridge, MA: Harvard University Press. University of Florida Institutional Review Board (2003/2004, December/January) Waiver of informed consent. Health Center IRB InvestiGATOR 6(12), 1. Retrieved July 15, 2007 from http://irb.ufl.edu/edu cation/InvestiGATORS.htm Villarruel, A. M., Bishop, T. L., Simpson, E. M., Jemmott, L. S., & Fawcett, J. (2001). Borrowed theories, shared theories, and the advancement of nursing knowledge. Nursing Science Quarterly, 14 (2), 158. Volkmar, F. R., Chawarska, K., & Klin, A. (2005). Autism in infancy and early childhood. Annual Review of Psychology, 56, 315. Volkmar, F. R., & Klin, A. (2005). Issues in th e classification of autism and related conditions. In F. R. Volkmar, R. Paul, A. Klin, & D. Cohen (Eds.), Handbook of autism and pervasive developmental disorders: Vol. 1. Diagnosis, development, neurobiology, and behavior (3rd ed ., pp. 5). Hoboken, NJ: John Wiley & Sons, Inc. Volkmar, F. R., Lord, C., Bailey, A., Schultz, R. T., & Klin, A. (2004). Autism and pervasive developmental disorders. Journal of Child Psych ology and Psychiatry, 45 (1), 135. Wakefield, A. J., Murch, S. H., Anthony, A., Linnell, J., Casson, D. M. Malik, M. et al. (1998). Ileal-lymphoid-nodular hyperplasia, non-speci fic colitis, and pervasive developmental disorder in children. Lancet, 351, 637.
138 Walco, G. A., Burns, J. P., & Cassidy, R. C. (2003). The ethics of pain control in infants and children. In N. L. Schechter, C. B. Berde & M. Yaster (Eds.), Pain in infants, children, and adolescents (2nd ed., pp. 157). Philadelphia: Lippincott Williams & Wilkins. Walker, D. R., Thompson, A., Zwaigenbaum, L., Go ldberg, J., Bryson, S. E., Mahoney, W. J., et al. (2004). Specifying PDD-NOS: A comparis on of PDD-NOS, aspe rger syndrome, and autism. Journal of the American Academy of Child and Adolescent Psychiatry, 43 (2), 172. Watzlawick, P., Beavin, J. H., & Jackson, D. D. (1967). The organization of human interaction. In Pragmatics of human communication (pp. 118). NY: W. W. Norton & Company. Williams, J. G., Higgins, J. P. T., & Brayne, C. E. G. (2006). Systematic review of prevalence studies of autism spectrum disorders. Archives of Disease in Childhood, 91, 8. Wood, J. T. (1982). Theory as a f oundation of human communication. In Human communication: A symbolic interactionist perspective (pp. 13). NY: Holt. World Health Organization. (1992). The ICD 10 Classification of Mental and behavioral Disorders: Clinical descript ions and diagnostic guidelines. Geneva, Switzerland: Author. Wynd, C. A., Schmidt, B., & Schaefer, M. A. (2003). Two quantitative approaches for estimating content validity. Western Journal of Nursing Research 25(5), 508. Zafeiriou, D. I., Ververi, A., & Vargiami, E. (2007). Childhood autism and associated comorbidities. Brain & Development, 29, 257. Zeltzer, L. K., & Schlank, C. B. (2005). Conquering your childs chroni c pain: A pediatricians guide for reclaiming a normal childhood. New York: HarperCollins Publishers, Inc.
139 BIOGRAPHICAL SKETCH Melissa Kay Dodd Inglese was born on May 27, 1979 in Yokosuka, Japan. The daughter of a Navy dentist, Melissas trav els and early exposure to m edical practices and health concerns all over the world prompted her to pursue a career in nursing. Sh e graduated from the University of Florida in 2001, with a Bachelor of Science degree in nursing. Her developing passion for children led her to pursue adva nced degrees in nursing thro ugh the accelerated BSN-PhD program offered by the University of Florida. In 2002, Melissa received her Master of Science degree in nursing and began he r clinical work as a pediat ric nurse practitioner, while simultaneously continuing to pursue her doctoral degree. Melissas interests involve th e alleviation of suffering in vulnerable populations of children. Specifically, her clinical work focu ses on children with comp lex chronic conditions conditions requiring children and families to navi gate Americas vastly complex healthcare system. Her attention to proficient pain assessment and management in various pediatric populations led her to question how childre n with autism spectrum disorder (ASD)a population receiving national and international attentionexperience and express physical discomfort. Thus the focus of her doctoral wo rk was directed toward s this unique group of individuals. Melissa now continues her cl inical and academic work focused on serving vulnerable populations of children.