<%BANNER%>

Perceptions of Wellness and Burnout among Certified Athletic Trainers

Permanent Link: http://ufdc.ufl.edu/UFE0024387/00001

Material Information

Title: Perceptions of Wellness and Burnout among Certified Athletic Trainers Contributions of the Wellness Domains
Physical Description: 1 online resource (153 p.)
Language: english
Creator: Naugle, Keith
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: athletic, burnout, district, physical, trainer, wellness
Applied Physiology and Kinesiology -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Burnout increases the prevalence of disease and dictates the severity of illness. Stressful job settings, long hours, and an increased number of stressors are mechanisms of professional burnout. These demands may cause a decline in health as time commitments generate imbalances between personal and professional lives. Furthermore, no information regarding the domains of wellness and their role in preventing burnout in AT is available. To better understand how the domains of wellness can reduce burnout and improve the health of the AT profession, the following hypotheses have been proposed; (Hypothesis 1): That burnout susceptible AT professionals (high burnout scores) will have decreased social, mental, and physical health. This result would suggest that these domains of wellness are part of a multi-factorial mechanism that leads to burnout. (Hypothesis 2): That disconnect (lack of a relationship) between the wellness domains and perceived wellness exists. A cross sectional design was used to compare perceived wellness, burnout, and wellness domain (social support, mental health, and physical activity) scores among certified athletic trainers employed in the southeastern district of the NATA (District 9). Variables include: demographics (hours worked, years of experience), Copenhagen Burnout Inventory (CBI), Perceived Wellness Survey (PWS), Baecke Physical Activity Questionnaire, Mental Health Inventory-5, and the Medical Outcomes Study (MOS) Social Support Survey. The mean scores for the CBI and the PWS were 41.81plus or minus17.44 and 16.5plus or minus2.8 respectively. A burnout score of 50 is considered high, and a PWS score closer to 29 is considered healthy. Four hundred and twelve members responded (response rate 26.4 %) of which 59 % were male. Pearson correlations showed significant negative relationships between burnout and years of job experience (r=-0.173, p < 0.001), social support (r=-0.265, p < 0.001), perceived wellness (r=-0.515, p < 0.001), mental health (r=-0.265, p < 0.001), and physical activity (r=-0.123, p < 0.001). Additionally, a significantly positive correlation was revealed between burnout and hours worked (r=0.124, p < 0.01), between perceived wellness and social support (r=0.388, p < 0.001), mental health (r=0.486 p < 0.001) and physical activity (r=0.200, p < 0.001). A regression analysis revealed that mental health, physical activity and social support directly contributed to perceived wellness (r2=0.579 p < 0.001). A regression analysis also revealed that perceived wellness, hours worked per week and mental health contributed to burnout (r2=0.32 p < 0.01). Based on these results, increased levels of social support, physical activity and mental health are associated with perceived wellness. Higher perceived wellness and fewer hours worked per week could lower burnout. Path analysis models were used to determine direct and indirect effects of the previous variables. The path model provided adequate fit and shows that hours worked per week, years of experience, physical activity, and mental health were causes of burnout. The three domains of health, social support, mental health and physical activity were causes of perceived wellness. By knowing potential causes of burnout and decreased wellness in athletic trainers, future research can work to decrease burnout by educating ATCs and employers about reducing work hours, increasing levels of physical activity and mental health, possibly causing less diseases and severity of illnesses.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Keith Naugle.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Borsa, Paul A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024387:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024387/00001

Material Information

Title: Perceptions of Wellness and Burnout among Certified Athletic Trainers Contributions of the Wellness Domains
Physical Description: 1 online resource (153 p.)
Language: english
Creator: Naugle, Keith
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: athletic, burnout, district, physical, trainer, wellness
Applied Physiology and Kinesiology -- Dissertations, Academic -- UF
Genre: Health and Human Performance thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Burnout increases the prevalence of disease and dictates the severity of illness. Stressful job settings, long hours, and an increased number of stressors are mechanisms of professional burnout. These demands may cause a decline in health as time commitments generate imbalances between personal and professional lives. Furthermore, no information regarding the domains of wellness and their role in preventing burnout in AT is available. To better understand how the domains of wellness can reduce burnout and improve the health of the AT profession, the following hypotheses have been proposed; (Hypothesis 1): That burnout susceptible AT professionals (high burnout scores) will have decreased social, mental, and physical health. This result would suggest that these domains of wellness are part of a multi-factorial mechanism that leads to burnout. (Hypothesis 2): That disconnect (lack of a relationship) between the wellness domains and perceived wellness exists. A cross sectional design was used to compare perceived wellness, burnout, and wellness domain (social support, mental health, and physical activity) scores among certified athletic trainers employed in the southeastern district of the NATA (District 9). Variables include: demographics (hours worked, years of experience), Copenhagen Burnout Inventory (CBI), Perceived Wellness Survey (PWS), Baecke Physical Activity Questionnaire, Mental Health Inventory-5, and the Medical Outcomes Study (MOS) Social Support Survey. The mean scores for the CBI and the PWS were 41.81plus or minus17.44 and 16.5plus or minus2.8 respectively. A burnout score of 50 is considered high, and a PWS score closer to 29 is considered healthy. Four hundred and twelve members responded (response rate 26.4 %) of which 59 % were male. Pearson correlations showed significant negative relationships between burnout and years of job experience (r=-0.173, p < 0.001), social support (r=-0.265, p < 0.001), perceived wellness (r=-0.515, p < 0.001), mental health (r=-0.265, p < 0.001), and physical activity (r=-0.123, p < 0.001). Additionally, a significantly positive correlation was revealed between burnout and hours worked (r=0.124, p < 0.01), between perceived wellness and social support (r=0.388, p < 0.001), mental health (r=0.486 p < 0.001) and physical activity (r=0.200, p < 0.001). A regression analysis revealed that mental health, physical activity and social support directly contributed to perceived wellness (r2=0.579 p < 0.001). A regression analysis also revealed that perceived wellness, hours worked per week and mental health contributed to burnout (r2=0.32 p < 0.01). Based on these results, increased levels of social support, physical activity and mental health are associated with perceived wellness. Higher perceived wellness and fewer hours worked per week could lower burnout. Path analysis models were used to determine direct and indirect effects of the previous variables. The path model provided adequate fit and shows that hours worked per week, years of experience, physical activity, and mental health were causes of burnout. The three domains of health, social support, mental health and physical activity were causes of perceived wellness. By knowing potential causes of burnout and decreased wellness in athletic trainers, future research can work to decrease burnout by educating ATCs and employers about reducing work hours, increasing levels of physical activity and mental health, possibly causing less diseases and severity of illnesses.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Keith Naugle.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: Borsa, Paul A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024387:00001


This item has the following downloads:


Full Text

PAGE 1

1 PERCEPTIONS OF WELLNESS AND BURNOUT AMONG CERTIFIED ATHLETIC TRAINERS: CONTRIBUTIONS OF THE WELLNESS DOMAINS By KEITH E. NAUGLE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFIL LMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Keith Naugle

PAGE 3

3 To my family and friends for all their prayers and support

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank God for the abilities I have been granted and the faith he has bestowed in me to complete this long journey. I would like to acknowledge the committee for their hard work and help in this unique and wo n derful research experience. I appreciate Dr Borsa giving me the autonomy to pursue a topic that has always been a passion of mine and that will help every athletic trainer in the profession. I thank Dr Tillman for being there for me every time I had a questio n that I needed clarification or guidance with in this creative process known as a dissertation I thank Dr Dodd for her excellent guidance in all the aspects of health and wellness on this project and in creating the best work I am capable of. I thank Dr Behar Horenstein for helping fine tune this topic and for her guidance and support in this journey I would like to thank Dr Garvan for the many meetings and much guidance as I pursued all the advanced statistical methods for this work. I would like to thank Dr Erik Wikstrom for all of his help and guidance in purs u ing the grant and for helping in the early stages of my doctoral career, functioning not only as a colleague but also as a dear friend I would like to thank Lexi Douglas for the abundant amount of time spent looking at excel data poin ts from all the survey items. I would like to thank Christy White for her editing and proofing of this project and for her unbelievable quickness in returning me the final product. Finally, I would like to thank Kelly Gamble for her continued assistance i n interpreting the multitude of SAS outputs and for her steadfast patience during the last two years as I juggled my responsibilities as a student, a f aculty member and the various other hats I wore in my life.

PAGE 5

5 TABLE OF CONTENTS Page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 9 LIST OF FIGURES ............................................................................................................................ 11 LIST OF ABBREVIATIONS ............................................................................................................ 12 ABSTRACT ........................................................................................................................................ 14 CHAPTER 1 INTRODUCTION ....................................................................................................................... 16 Significance ................................................................................................................................. 17 Statement of the Problem ............................................................................................................ 20 Research Questions ..................................................................................................................... 20 Operational Definitions ............................................................................................................... 21 Interventions ................................................................................................................................ 22 Assumptions ................................................................................................................................ 22 Delimitations ............................................................................................................................... 22 Methodological Limitations ....................................................................................................... 22 2 LITERATURE REVIEW ........................................................................................................... 23 Burnout ........................................................................................................................................ 23 Social Support and Burnout ........................................................................................................ 30 Health and Burnout ..................................................................................................................... 31 Health and Wellness .................................................................................................................... 31 Allied Healthcare Providers Burnout ....................................................................................... 32 Physician Burnout ................................................................................................................ 33 Physician Assistant Burnout ............................................................................................... 34 Physical Therapist Burnout ................................................................................................. 35 Athletic Trainers Burnout and Wellnes s .................................................................................. 36 Burnout in the Athletic Trainer ........................................................................................... 36 Mental Health in Athletic Trainers ..................................................................................... 39 Physical Activity in the Athletic Trainer ............................................................................ 40 Barriers to Physical Health in Athletic Trainers ................................................................ 42 Social Support Le vels of the Athletic Trainer ................................................................... 42 Domains of Health and Wellness ............................................................................................... 42 Physical Health .................................................................................................................... 42 Mental Health ....................................................................................................................... 43 Social Health ........................................................................................................................ 44 Theoretical Framework ............................................................................................................... 45

PAGE 6

6 Summary ...................................................................................................................................... 45 3 METHODS .................................................................................................................................. 47 Study Setting and Participants .................................................................................................... 47 Study Design ............................................................................................................................... 47 Materials ...................................................................................................................................... 48 Instrumentation ............................................................................................................................ 48 Demogr aphics .............................................................................................................................. 49 Physical Activity ......................................................................................................................... 49 Mental Health .............................................................................................................................. 51 Social Support ............................................................................................................................. 51 Burnout (Professional and Personal) .......................................................................................... 52 Perceived Wellness ..................................................................................................................... 53 S coring of Instruments ................................................................................................................ 54 Pilot Data Collection ................................................................................................................... 54 Data Collection ............................................................................................................................ 54 Data Analysis ............................................................................................................................... 55 Descriptive Statistics ........................................................................................................... 56 Coefficient Alpha ................................................................................................................. 56 Pearson Correlations ............................................................................................................ 56 Multiple Regressions ........................................................................................................... 57 Path Analysis (SEM) ........................................................................................................... 57 T test/ANOVA ..................................................................................................................... 58 Permission Notes ......................................................................................................................... 58 4 RESULTS .................................................................................................................................... 61 Pil ot Data ..................................................................................................................................... 61 Internal Consistency .................................................................................................................... 61 Data Management ....................................................................................................................... 62 Demographi cs and Descriptive Analysis ................................................................................... 62 Primary Analysis ......................................................................................................................... 65 Correlations Between Variables ......................................................................................... 65 Regression Analysis ............................................................................................................ 66 Path Analysis for Burnout and Perceived Wellness .......................................................... 67 Perceived wellness ....................................................................................................... 69 Burnout ......................................................................................................................... 70 Mental health, social support, and physical activity: ................................................. 70 Second ary Analysis ..................................................................................................................... 71 Bivariate Analysis ................................................................................................................ 71 One -way ANOVA ............................................................................................................... 71 Tables an d Charts for Results ..................................................................................................... 74 Tables and Charts for Results ..................................................................................................... 74 5 DISCUSSION .............................................................................................................................. 96

PAGE 7

7 Burnout and Perceived Wellness Path Model ........................................................................... 96 Burnout ........................................................................................................................................ 97 Athletic Trainers .................................................................................................................. 97 Physicians ........................................................................................................................... 102 Physical Therapist .............................................................................................................. 103 Physician Assistant ............................................................................................................ 103 Physical Activity and Burnout in Athletic Trainers ................................................................ 104 Mental Health and Burnout in Athletic Trainers ..................................................................... 1 05 So cial Support and Burnout in Athletic Trainers .................................................................... 106 Perceived Wellness and Burnout .............................................................................................. 107 Perceived Wellness and the Health Domains i n Athletic Trainers ........................................ 108 Conclusion ................................................................................................................................. 109 Limitations ................................................................................................................................. 109 Implicati ons ............................................................................................................................... 110 Practical Applications ............................................................................................................... 111 Future Research ......................................................................................................................... 111 APPENDIX A PSYCHOMETRIC DATA FOR INSTRUMENTS ................................................................ 113 B INSTRUMENT ......................................................................................................................... 116 Demographics .................................................................................................................... 116 Perceived Wellness Survey ............................................................................................... 117 Copenhagen Burnout Inventory ........................................................................................ 121 Mental Health Inventory 5 ................................................................................................ 123 Baecke Physical Activity Questionnaire .......................................................................... 124 MOS Social Support Survey ............................................................................................. 126 C PERMISSION TO USE INSTRUMENTS .............................................................................. 127 D INSTITUTIONAL REVIEW BOARD ................................................................................... 130 E LETTER TO PARTICIPANT .................................................................................................. 133 F CONTACT LIST REQUEST FORM ...................................................................................... 134 G VARIABLE TABLES .............................................................................................................. 136 Perceived Wellness Survey ............................................................................................... 136 Copenhagen Burnout Inventory ........................................................................................ 140 Mental Health Inventory 5 ............................................................................................. 143 Baecke Physical Activity Questionnaire .......................................................................... 144 MOS Social Support .......................................................................................................... 146 LIST OF REFERENCES ................................................................................................................. 147

PAGE 8

8 BIOGRAPHICAL SKETCH ........................................................................................................... 153

PAGE 9

9 LIST OF TABLES Table page 4 1 Pilot data test -retest correlations ........................................................................................... 74 4 2 Internal Consistency/Cronbachs Alpha (reliability) ........................................................... 75 4 3 Job title frequency .................................................................................................................. 76 4 4 Job setting frequency ............................................................................................................. 77 4 5 Job title frequency collapsed ................................................................................................. 78 4 6 Job setting frequency collapsed ............................................................................................. 79 4 7 Means, standard deviations, Chronbachs Alpha scales and subscales .............................. 80 4 8 Means and standard deviations for each gender ................................................................... 81 4 9 Means and standard deviations of scales by job setting and job title ................................. 82 4 10 Correlations between variables ............................................................................................. 84 4 11 Re gression model for burnout ............................................................................................... 85 4 12 Regression model for perceived wellness ............................................................................. 86 4 13 Covariance table ..................................................................................................................... 87 4 14 Path model Beta estimates and significance ......................................................................... 88 4 15 Factor loading table ................................................................................................................ 89 4 16 Erro r variances for path model. ............................................................................................. 90 4 17 Independent t -test gender ....................................................................................................... 91 4 18 Job setting ............................................................................................................................... 92 4 19 Job title ANOVA .................................................................................................................... 93 A 1 Table of psychometric data .................................................................................................. 113 A 2 Scoring of instruments table ................................................................................................ 115 G 1 Perceived W ellness Survey ................................................................................................. 136 G 2 Copenhagen Burnout Inventory .......................................................................................... 140

PAGE 10

10 G 3 Mental Health Inventory 5 .................................................................................................. 143 G 4 Baecke Physical Activity ..................................................................................................... 144 G 5 MOS Social Support Survey ............................................................................................... 146

PAGE 11

11 LIST OF FIGURES Figure page 3 1 Variables of Importance and outcome measures for Athletic Trainers Burnout. ............. 59 3 2 Causal Model: Factors that determine Professional Burnout and Perceived Wellness ..... 60 4 1 Full model with significant and non significant paths ......................................................... 94 4 2 Path model with significant paths only ................................................................................. 95

PAGE 12

12 LIST OF ABBREVIATION S Allied Health Care P rofessional Educated, certified or licensed professional who works within the American Medical Association medical field s American College of Sports Medicine (ACSM) Organization that researches and determines appropriate fitness and health standards in the United States. American Medical Association (AMA) Governing body of American Professionals Athletic Tra iner: (ATC): Allied health care professional that works with athletes and active individuals Athletic Training profession (AT) The profession itself will be referred to as AT in the writing of the paper Baecke P hysical Activity Question naire M easurement of habitual physical activity at work, recreational and leisure times. Body Mass Index (BMI) Ratio of weight to height commonly used to determine healthy levels of body weight Burnout P hysical and emotional/mental exhaustion from work environments that are physically and emotional demanding .* CAATE Commission on Accreditation of Athletic Training Education Copenhagen Burnout Inventory (CBI) Measurement of burnout for work, client a nd personal Medical Outcomes Study (MOS) Social Support Survey Measurement of social support Mental health Possession of self efficacy and emotional stability in daily living Mental Health Inventory (MHI) Measurement of mental healt h NATA National Athletic Training Association Perceived Wellness Survey (PWS) Measurement of perceived wellness Physical activity Activity that takes place either as leisure time activity or work related activity. **

PAGE 13

13 Physical health Physical properties of good body measurements (BP Cholesterol et.) due to a sufficient amount of physical activity Social support The availability of someone to provide or exchange resources that are perceived as a way to enha nce ones wellness. *** Southeast Athletic Training Association: (SEATA) SEATA is one geographic district of the National Athletic Training Associations and is also called District 9. Stressor Job characteristic or personal characteris tic that causes negative outcomes on a persons health or wellness Wellness Objective and subjective balance between the domains of physical, mental and social health Maslach (2001) ** Cuppett (2002) ***Sherbourne (1991) Callaghan (1993)

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PERCEPTIONS OF WELLNESS AND BURNOUT AMONG CERTIF IED ATHLETIC TRAINERS: CONTRIBUTIONS OF THE WELLNESS DOMAINS By KEITH E. NAUGLE May 2009 Chair: Paul Borsa Major: Health and Human Performance Burnout increases the prevalence of disease and dictates the severity of illness. Stressful job settings, lo ng hours, and an increased number of stress ors are mec hanisms of professional burnout. These demands may cause a decline in health as time commitments generate imbalances between personal and professional lives. Furthermore, no information regarding the domains of wellness and their role in preventing burnout in AT is available. To better u nderstand how the domains of wellness can reduce burnout and improve the health of the AT profession the following hypothes e s h ave been proposed; (Hypothesis 1) : Th at burnout susceptible AT professionals (high burnout scores) will have decreased social, mental, and physical health. This result would suggest that these domains of wellness are part of a multi factorial mechanism that leads to burnout (Hypothesis 2) : That disconnect (lack of a relationship) between the wellness domains and perceived wellness exists A cross sectional design was used t o compar e perceived wellness, burnout, and wellness domain (social support, mental health, and physical activity) score s among certified athletic trainers employed in the southeastern district of the NATA (D istrict 9). Variables include: demographics (hours worked, years of experience), Copenhagen Burnout Inventory (CBI), Perceived Wellness Survey (PWS), Baecke Physical Activity Questionnaire, Mental Health Inventory 5 and the Medical Outcomes

PAGE 15

15 Study (MOS) Social Support Survey. The mean scores for the CBI and the PWS were 41.8117.44 and 16.52.8 respectively. A burnout score of 50 is considered high, and a PWS score close r to 29 is considered healthy. Four hundred and twelve members responded (response rate 26.4 %) of which 59 % were male. Pearson correlations showed significant negative relationships between burnout and years of job experience ( r = 0 .173, p < 0 .001), social support (r = 0 .265, p < 0 .001), perceived wellness ( r = 0 .515, p < 0 .001), mental health ( r = 0 .265, p < 0 .001), and physical activity ( r = 0 .123, p < 0 .001). Additionally, a significantly positive correlation was revealed between burnout and hours worked ( r = 0 .124, p < 0 .01) between perceived wellness and social support ( r = 0 .388, p < 0 .001) mental health ( r = 0 .486 p < 0 .001) and physical activity ( r = 0 .200, p < 0 .001). A regression analysis revealed that mental health, physical activity and social support directly contribu ted to perceived wellness ( r2= 0 .579 p < 0 .001). A regression analysis also revealed that perceived wellness, hours worked per week and mental health contributed to burnout ( r2= 0 .32 p < 0 .01). Based on these results, increased levels of social support, physica l activity and mental health are associated with perceived wellness. Higher perceived wellness and fewer hours worked per week could lower burnout. Path analysis models were used to determine direct and indirect effects of the previo us variables. The path model provided adequate fit and shows that hours worked per week, years of experience, physical activity, and mental health were causes of burnout. The three domains of health, social support, mental health and physical activity were causes of perceived w ellness. By knowing potential causes of burnout and decreased wellness in athletic trainers, future research can work to decrease burnout by educating ATCs and employers about reducing work hours, increasing levels of physical activity and mental health, possibly causing less diseases and severity of illnesses

PAGE 16

16 CHAPTER 1 INTRODUCTION Athletic training is a service profession dedicated to the care and prevention of injuries for clients ranging from high school, college or professional athletes to elderl y patients and industrial workers.1 Long work hours and high job stress situations can lead to job burnout and a decline in health and the quality of life of athletic trainers ( ATC ).2, 3 ATCs typically put in long hours at work working with injured individuals spending large amounts of time traveling with sports teams or providing coverage to teams during practice and competition These work related factors often create an imbalance between work and personal life An imbalance of work and personal life can result in decreased healthy activities such as exercise or socialization with friends an d families and social support.2, 4 8 Research has shown that ATCs are susceptible to burnout .2, 4, 5, 9, 10 As a result of the intensity of their work roles, some athletic tra iners do not have the availability of leisure time as do other athletic trainers.2, 5, 6, 9 Other health care professionals such as physical therapist s nurses, social workers, physicians, physicians assistants, teachers and coaches are al so susceptible to high levels of stress and burnout.2, 4 6, 9, 1115 As a profession, athletic training has been in existence for over fifty years and may have began as early as 1917.1 The American Medical Association recognizes the athletic training profession as an allied health care profession.16 Athletic trainers (ATC) are certified by the Independent Board of Certification (B OC),16 licensed or registered in 44 states in the United States, and must maintain strict continuing education credits t o keep their certification and license.17 Educational levels among ATCs vary from a bachelors degree to those individuals holding an MD or PhD. More than 70 % of ATCs hold at least a master s degree.17

PAGE 17

17 Certified athletic trainers specialize in the prevention, recognition, management and rehabilitation of injuries result ing from physical activity.1 The ATC works in a variety of settings that include: high school, college and professional sports as well as in corpo rate hospital and clinical settings. Certified athletic trainers work with other health care professionals including physical therapist s and physicians. They also interact with parents, coaches and athletic administrators. Because ATCs spend countless h ours on injury evaluations, developing treatment plans and preparing for activities in addition to educating athletes, coaches and parents about injury time lines and rehabilitation processes they need crucial time management skills. Athletic trainers have stressful day to -day duties that over time ( spanning from mere months to many years) can lead to burnout, decreased physical vigor and decreased social support from friends and family.3 Significance Burnout has been shown to increase the incidence of disease and dictate the severity of i llness .18 A thletic t raining p rofessionals are not exempt from the chronic illness es and health problems that common ly plague society as a byproduct of burnout .2 Research indicates that AT professionals or ATCs, like other health care professionals, have decreased job satisfaction, increased job turnover, and high incidence of physical and mental health problems.36, 8, 9, 11, 1315, 19 Determining the impact of burn out and perceived wellness h as the potential to dictate policy changes by the NATA or the governing body of athletic trainers and can reduce the potential for future health related problems as well as promote job and life satisfaction in certified athleti c trainer s .8, 20 Understanding the determinants of burnout and their effects on perceived wellness among ATCs hold promise for creating effective i nterventions for ATCs and those in other professions as well.

PAGE 18

18 Early burnout discussions in athletic training focused on both physical and psychological signs and symptoms including: headaches, gastrointestinal or cardiovascular disturbances fatigue anxiety, depression, sleeplessness, and even sexual dysfunction.4, 5 Furthermore, lack of knowledge regarding burnout and the accompanying symptoms, caused many individuals to turn to drugs, tobacco and/or alcohol, and as a result compound ed their own health problems.4, 5 Certain job characteristics (e.g. >40 hour work weeks, multiple areas of focus) were suggested to predispose AT professionals to burnout. 2, 21 While this information is informative it typifies the current understanding of professional burnout in AT professionals, a s anecdotal, and lacking of evidence .2 In the most recent peer reviewed article on burnout the author also suggests role conflict and ambiguity as well as organizational demand s as factors that contribute to burnout.9 A lthough the actual in cidence of burnout among athletic trainers is unknown, researchers have shown that athletic training professionals experience burnout and are susceptible to burnout because of job demands 36 To elucidate the mechanisms of professional burnout valid and reliable instruments must be used. Previously, the Maslach Burnout Inventory (MBI)3 and more recently the Copenhagen Burnout Inventory, have examined burnout in non AT professionals and non -service oriented professionals respectively.22 The m ost recent attempt at assessing athletic training burnout used the instrument called the Athletic Training Burnout Inventory (ATBI). This instrument was based on the MBI and uses the addition of more items to make it more specific to athletic training usi ng constructs like time commitment and organizational support.9 Additionally accurate and consistent measures of the wellness domains should be used. Physical activity is known to increase overall health and quality of life, but little in formation is known regarding the fitness and activity levels of AT professionals .6 Previous research has indi cated which job

PAGE 19

19 settings allow AT professionals to be the most active (clinical setting) however, no burnout data was collected.6 Athletic trainers who work in traditional sport se ttings may find participating in physical activity difficult due to travel requirements and team coverage which tends to create a lack of co ntrol over ones work schedules 6 Furthermore, mi nimal research is available on social support and mental health in AT professionals. This lack of re search demonstrates the need to determine the overall perception of health and wellness in the athletic training profession as well as the current levels of social support and mental health Limited research may be due to the handling of the burnout syndrome by AT professionals For example, many AT professionals are able to recharge during the off season By refocus ing the approach on the profession and interact ing in positive ways to improve t he relationship s with staff and clients the athletic trai ner is able to reexamine the time allotments and daily job duties, and learn to delegate responsibility to other staff and students during the off season .4 These types of c oping mechanisms have allowed AT professionals to survive! If the AT profession is to continue to grow, AT professionals must do more th a n survive especially in a cultural environment that is placing an increased priority o n high school athlet ics and yea r round competition at all level s AT professionals will continue to face greater challenges and threats to their physical, social, and mental health unless the mechanisms of burnout can be elucidated and evidence based interventions are designed to incr ease overall wellness or well being Burnout is thought to be preventable by consciously scheduling leisure time and pleasurable activities even if the activities are physical in nature .4 Additionally, s everal studies have discussed coping methods or personality traits that may aid in the reduction of job burnout.3, 20 However, the current lack of literature on AT burnout and wellness prohibits a clear understanding of the causes and c onsequences of burnout and decreased wellness

PAGE 20

20 Statement of the P roblem Burnout increases the prevalence of disease and dictates the severity of illness in health care professionals .18 Stressful job settings, long hours, and an increased number of stress ors are mechanisms of professional burnout.25, 18 The professional requirements of athletic training inherently demand a significant time commitment and personal sacrifice. These demands may cause a decl ine in health as time commitments generate imbalances between personal and professional lives.2, 3, 10, 23, 24 Furthermore, the imbalances are believed to cause a cascade of behavioral changes which contribute to a decrease in ones physical, social and mental health domains, thereby leading to changes in physical activity, group socialization/supp ort and mental well being .3 6 Unfortunately, little is understood about the mechanisms of professional burnout with in the field of AT. Furthermore, no information exists regarding associations between the domains of wellness and burnout prevention in AT or the ir perceived levels of wellness Based on these observations, it is evident that the mechanisms preventing and leading to professional burnout and decreased levels of wellness are unclear and need to be explored. Until these mechanisms are elucidated, professional burnout and morbidity will be common place within the field of athletic training Research Q uestions 1. To determine the impact or r oles that the wellness domains (physical, mental, social) have on burnout in AT professionals The hypothes is states that wellness domain scores will differ between AT professionals with high and low burnout scores. Specifically, predict ing that burnout susceptible AT professionals (high burnout scores) will report decreased levels of social, mental, and physical health. This result would suggest that these domains of wellness are part of a multi -factorial mechanism leading to burnout.

PAGE 21

21 2. To determine th e impact or roles that wellness domains (physical, mental, social) have on perceived wellness in AT professionals The hypothesi s states that wellness domain scores will differ from the scores of perceived wellness in the AT professionals. Specifically, p redict ing a disconnect (lack of a correlation) between wellness domain scores and perceived wellness in AT professionals. This result would suggest that AT professionals can not accurately determine their health and represents a component of a multi -facto rial mechanism leading to burnout and decreased perceptions in wellness 3. To determine the impact or roles that demographic factors have on the burnout and the perception of wellness in AT professionals The hypothesi s states that hours worked, work exp erience and job setting will account for the variance in burnout and perceived wellness scores in AT professionals. Specifically predict ing that those with more hours worked and less work experience working in high level college or professional sports wi ll have higher burnout scores and lower perceived wellness scores. Operational D efinitions Perceived Wellness: Measured with the items from the six subscale (health: physical, spiritual, psychological, social, emotional, and intellectual) of the Perceived Wellness Survey instrument Burnout: Measured with the items from the three subscales (work, client and personal burnout) of the Copenhagen Burnout Inventory instrument Mental health: Measured with the Mental Health Inventory 5 Instrument Physical health: M easured with the items from the three indices (work index, sport index, leisure time index) from the Baecke Physical Activity questionnaire Social health: Measured with the subscale items (emotion, tangible, affectionate, and positive social) from the MOS Social Support Survey instrument

PAGE 22

22 Interventions Upon determining the levels of burnout, wellness, and the levels of the health domains, possible policy changes or interventions can become available to those workers (athletic trainers) that are experienci ng burnout or decreased wellness or perceived wellness. The knowledge of the relationships and ability to predict burnout and decreased wellness through either demographic variables or health domain scores will allow for entering and established professionals to improve both their general health and quality of life. Assumptions 1 All participants will accurately report their perceptions of wellness burnout and wellness domain levels. 2 The instrument used is appropriate for collecting the intended informati on 3 The instrument is appropriate to collect information to answer the proposed research questions. Delimitations 1 The participants are from only one section of the National Athletic Trainer s Association 2 The study is delimited to the specific items and su bscales of the currently used instrument. 3 The instrument is a computer accessible tool and data is gathered only through this medium. 4 The data is depend e nt on subject recall and subject interpretation of the instrument questions Methodological Limitations 1 The results of this study depend upon receiving an adequate response rate. 2 Response bias from the subjects can have effects on the survey outcomes. 3 T he use of computer based instruments may not allow access to all of the population. 4 Results will not be generalizable to all athletic trainers, only those in the southeast United States

PAGE 23

23 CHAPTER 2 LITERATURE REVIEW Burnout The condition known as burnout first appeared in the 1970s in workers who were involved in human service professions or health care workers .25, 26 These workers were thought to experience an overwhelming exhaustion from the work environment, a det achment from the job and finally a decrease in personal accomplishment.25, 26 The condition called b urnout is tho ught to be caused by working to provide services or aid to people in need and is characterized by stressors that deplete emotional health and create a decrease in motivation and commitment.25, 26 Thus the core of burnout in past research is job related and specifically the relationship between the client and the provider. Burnout has been attribut ed to many factors from role conflict, role ambiguity, 9, 25 low social support from coworkers,3, 27 guidance, organizational stress, wor k load3, 9, 28 30 perceived lack of control, satisfaction with resources,29 and even different personal ity traits.3 With so many different fac tors determined by a variety of researchers the cause of burnout still remains a huge mystery While the factors causing burnout are many and still yet to be determined, what is known about burnout is that global outcomes like decreased overall health an d job satisfaction, increased job turnover decreased quality of care, and even substance abuse problems do occur in helping professions such as physicians and athletic trainers .35, 9, 25, 26 And even more specific outcomes of burnout in these professions are physical and emotional exhaustion, fatigue anxiety, depression, sleeplessness, insomnia, depression, anxiety, irritability, alcohol and drug use and even marital prob lems and sexual dysfunction .4, 5, 13, 26 Because burnout has many potential causes and just as many harmful outcomes the ability to measure burnout is still being dissected and discussed at great lengths by many researchers.

PAGE 24

24 The prevention of burnout is just as mystifying. Many r esearchers believe burnout is work related or specifically work role related Other researchers believe that burnout is personality related and yet other researchers conclude burnout to be a combination of many factors With the different potential cause s, diverse possible ways to prevent burnout are scattered throughout the literature. The main burnout researchers Maslach and Schaufeli believe that prevention should be focused on the individual, and one s capacity to cope in the workplace.25, 31 Their other component of prevention is to make changes or interventions in the organization and the organizational environment.25 Other researchers suggest educating the individual and teaching specific coping strategies that promote good work life balance and skills like time management.23 Several authors discuss coping mechanisms and personality traits that can also be shown to prevent burnout.25, 32 One author mentions the focus on the prevention of burnout through the use of, or the promotion of personal health and professional well being in physicians. This includes all levels of health: physical, emotional, psychological and spiritual.29 As more research shows the causal links of burnout and other factors more and more potential prevention interventions can be used for all levels and types of workers. In past research one measure has b een used primarily to determine burnout and is considered by many as the gold standard of measuring burnout. This gold standard is called t he Maslach Burnout Inventory (MBI) and it measure s the three constructs defined in the burnout syndrome: emotional exhaustion, depersonalization or cyni ci sm and decreases in personal accomplishment.25, 26 The MBI instrument was designed to look at the attitudes and feelings of workers who were experiencing burnout and was originally measure d by frequency and intensity.26 The MBIs original populations included nurses, probation officers, mental health workers, counselors, teachers, police officers and social workers, but has since expanded to

PAGE 25

25 physicians, physical therapists and athletic trainers2, 4, 5, 9, 13 15, 20, 29, 3337 By using spousal and work evaluations the MBI was shown to have convergent, and discriminant validity.26 Test/ retest reliability and internal consisten cy was also determined for the MBI.26 In more recent research by Kristense n the author discusses three subdivisions of burnout : personal, work and c lient related, which is different from the three levels of burnout that Maslach describes as emotional exhaustion, cynicism or depersonalization and decreased self accomplishment or self efficacy .22, 25 This study was part of a bigger project called the Project on Burnout, Mot ivation, and Job Satisfaction (PUMA). The PUMA study was a five -year prospective study on employees working in the human service sectors, with a variety of workers including those that work in s tate psychiatric facilities, welfare offices, and home care services.22 The PUMA study looked at a large variety of bot h general and specific health outcomes and included questionnaires like the Short Form 36 for general health and other mental health questionnaires.22 Within the PUMA study the C openhagen Burnout Inventory (CBI) was designed by Kristensen et al., and was used to measure the one concept of burnout. The CBI use d three measures separate or totaled to determine burnout which differed from the MBI which had three measures each used indepe n d ently.22 The CBI and the authors of the CBI proposed that these subdivisions focus on only fatigue and exhaustion as its core and remove the personal accomplishment that the MBI mea sures.22 However, the authors dont rely only on these tw o core ideas, but also include the idea of different domains of a person s life, hence the inclusion of per sonal, work and client burnout.22 The CBI showed both validity and reliability as part of the PUMA study. The internal consistency or reliability was high at 0 .85 to 0 .87. The validity was shown with cor relations from the Short Form 36 or SF 36 instrument. The highest correlation was between the personal

PAGE 26

26 burnout subdivision and the vitality part of the SF 36 ( 0 .75). The lowest correlation was between client burnout and general health ( 0 .34).22 The CBI was also directly compared to the MBI and found that bot h are equal when determining high burnout levels 38 T he most recent article using the CBI showed that there was a relationship betwe en sickness absence days (days missed at work) and sickness absence spells per year (groups of days missed from work) .39 These findings are part of the results from the PUMA study mentioned earlier in the text. The findings include some other interesting outcomes The number of people who had moderate alcohol consumption were less burned out then those with lower levels of alcohol consumption. T hose that consumed heavy amounts of alcohol were more burned out tha n those that drank moderately or not at all. Th e result of drinking more alcohol was also found to be the same the study follow up. T hose that did not drink alcohol reported an increase in burnout at follow up.39 Another recent tool used for measuring burnout in athletic trainers is the Athletic Training Burnout Inventory (ATBI).9 This instrument is a modified version of the MBI, and included three constructs called emotional exhaustion and depersonalization, level of stress, and level of organizational support. This instrument then eliminated the personal accomplishment construct from the MBI.9 The scale for recording information was also amended from the MBI and the n changed to use a consistent scale throughout the entire instrument. The instrument was field tested and sent to five experts for face validity and was returned with few amendments.9 The one main amendment was to modify the scoring scale and avoid compar ing the ATBI to the MBI. Content validity was determined with feedback from ten ATCs currently employed in large Midwestern NCAA DI universities.9 Reliability of the ATBI was accomplished through a first mailing to 50 ATCs in the f our NCAA universities. The first construct, emotional exhaustion and depersonalization, has been

PAGE 27

27 shown to have a Cronbachs A lpha of 0 .87, the second construct, level of stress, had a Cronbachs A lpha of 0 .57 and the third construct, organizational suppor t had an alpha of 0 .7. The a priori alpha value was set at 0 7 .9 Since the second construct did not meet this criteria several items were suspect and then the construct was split into two more constructs called administration responsibili ty and time commitment, which had Cronbachs alphas of 0 .74 and 0 .60 respectively.9 Even though these separate constructs did not meet the a priori 0 .7, the items were not changed and researchers rationalized that if a bigger population w ere used this might produce better results.9 Since the split of the level of stress construct, into the administration responsibility and time commitment, a four construct instrument was created with a total of 50 items.9 This particular instrument (ATBI) is based on the MBI, but has been modified in several ways specific to the athletic training professional job roles and show s promise in the res earch areas of burnout and may produce similar results when compared to the ot her burnout instruments used in current research.9 Another group of researchers have cast speculation on the MBI and its ability to measure burnout. A conflict exists in that the MBI involves measuring the concept of burnout with three co nstructs in the MBI. The concept of depersonalization defined by Maslach may not be part of the burnout concept.22 The CBI developed by Kristensen uses three sub dimensions for personal, work related and client related burnout. This instrument focuses on fatigue and exhaustion, which might be more of an accurate measurement of burnout.22 Th is discussion or debate o ver the two main burnout inventories put forth by Kristensen and Maslach involves the functionality or appropriateness in each instruments specific usage The first discussion is the circularity of burnout and the MBI. According to Kristensen, the MBI m easured only burnout in human services professions thus only allowing those professionals

PAGE 28

28 work ing within an area related to human services as possible subjects of MBI studies .22 However according to Schaufeli a general version of the MBI has been developed to address this possible discrepancy.40 Another discussion point is the ability to combine the subscale scores from the MBI, because the first two scales are scored high and when the subscale of personal accomplishment is decreased it is difficult to combine and have a total burnout score because personal accomplishment works differently th a n the other scales in the MBI .22, 40 T he CBI has the three subdivisions that can be measured both separately and then combined for a total score.22 The MBI has questions that can be extreme and trigger hostile responses, where the CBI is straightforward and lacks any possible hostile responses.22, 40 Another difference between the two measures is the defining of burnout itself. The MBI defines the b urnout term as ones relationship with work and not with people at work, a nd therefore refers to only the job stresses or stressors as causes of burnout. These instruments are trying to determine what levels of burnout are occurring in working professionals and people in general. O ther outcomes of burnout research are to determine factors that impact or lead to burnout. Many factors have been suggested in past research and some researchers are still looking for more factors. Some of the factors that pa st research found include: long work hours, travel schedule, limited personal resources, difficult workloads, multiple job demands ,3, 20, 21, 25 and some demographic and personal factors may play a role in bu rnout, such as gender, marital status and personality characteristics like, hardiness and copying styles.3, 20, 21, 25 Another area of research for burnout that is currently under review is the coping strat egy employed by workers that survive high burnout jobs. 32, 34 Active, adaptive and problem based focus es are more often used tha n any type of pass ive or emotional focus and reveal that job

PAGE 29

29 burnout can be reduced with active interventions.34 An article on athletic trainers looked at the coping responses of certified athletic trainers in graduate schools. This article found that burnout was reported from sources of stress at work and found that two of the people interviewed intended to leave the profession.32 Aside from coping, another interesting factor that might affect burnout is whether the person is a paid professional or a voluntee r.41 In a study by Gabassi, the researcher looked at volunteers versus paid hospital workers and found that those who performed the same jobs, but were volunteers were less burned out th a n the paid professionals.41 The volunteers had a mean score lower th a n those of paid professionals leading to the conclusion that volunteers seemed to be less emotionally exhausted than paid professionals. This differen ce in scores coul d be a result of the motivation that exist s between volunteers and paid professionals.41 Volunteers have the ability to control when to leave or to appear for a job Volunteers also have no financial attachment to the job and thus fewer expectations than paid professionals. A key aspect of burnout that affects everyone is the amount of or quality of care that is given to patients or clients. In a study by Calzi, the researcher found decreased quality of care in relation to middle levels of burnout scores.42 A decrease in quality of services then would redu ce the effectiveness of the system believed to help the patients or clients. This study looked at several different health care providers including: physicians, nurses, therapists and technicians.42 The author believed that t he nature and possible causes of burnout or what was called psychological distress emerged from the work that each different provider carried out during the daily routines.42 The issue of burnout not only causes problems for those workers that experience the burnout syndrome, but also cause s problems for the clients and / or patients seen by th e health care worker. The ability to determine factors that cause burnout and managing those factors and

PAGE 30

30 preventing them from occurr ing could lead to healthier workers and potentially safer health care environments for patients and clients. Accordingly, Calzi suggests that monitoring for high levels of personal accomplishment and decreasing the emotional exhaustion scores simultaneous ly could allow for the prevention of burnout reaching high levels.42 The inclusion of multiple areas of workers shows that many aspects of burnout and exhaustion exist in different positions.42 Social S upport and Burnout Social support has been reported as a factor in job stress but is not well specified in conceptual models.43 S ocial support has a complex definition and there are a variety of different types that lead to the definition of social support. Many researchers incorporate a variety of different types of support such as : emotio nal, esteem, tangible or affectionate.43, 44 Another problem existing with social support is that social support can be defined as support outside the workplace and at the workplace (work versus non-work) .45 A study by Bradley showed that those that received support in a work environment were in better health but the evidence doesnt show that the management and coworkers were the reason for the increased health. 43 This outcome adds more to the confusion on where social support and the perception of social support actually helps a persons health in regards to levels of burnout Another type of social support, which is found in person to person support or those that exchange resources, plays a key role in maintaining health and prevention of destructive effects from environment and social stress.46 This type of perceived social support can function as a n individuals buffer from stressful life events.46 These stressful life events include job stressor s and work related incidents Eith er aspect of social support can be important and both play important roles in regards to burnout and health. However, the facet of social support found at work has shown to be more closely related to emotional exhaustion and a thought exists that burnout and social support are reciprocal in their relationship.45 Overall, a lack of social support

PAGE 31

31 has been eluded to as fatal for the person lacking perceived social support,46 and can possibly exacerbate burnout, depression and inability to cope with work demands.30 Current research shows the role of social support or perceived social support is still in development and needs to be further discussed in the literature. Health and Bu rnout Burnout has severe effects on a persons work life, but minimal research shows exactly the outcomes burnout has on overall health and wellness. One study by Honkonen discussed the relationship between burnout and musculoskeletal diseases among women and cardiovascular diseases in me n.18 This study used some 8000 30 plus year old subjects living in Finland to secure a total of 3470 participants who completed a health questionnaire and a clinical health exam. Th e study found that those women who experienced burnout experienced musculoskeletal disease which included: chronic low er back pain, chronic neck pain and hip and knee osteoarthritis.18 Although m en experienced burnout and cardiovascular disease as a possible outcome from job strains, this result wasnt shown in women.18 C hronic fatigue syndrome or CFS is an other possible outcome from burnout in professionals.18 With so many instances of health issues involving burnout as a leading cause there is a combined thought of burnouts effec t on health and the inverse relationship of wellness or lack of leading to burnout. Health and W ellness Several definitions of health exist among healthcare professionals within the country. As stated by the World Health Organization, the three main a spects of health are mental or emotional well being, social well being, and physical well being .47, 48 M ost people have a picture of health as being only about the physical health or just being absent of illness .48 However, in order to maintain a good quality of life or well being an individual needs to balance all the different aspect s of health. While most p eople only consider physical health, overall health

PAGE 32

32 incorporates mental or emotional health as well as social health. Each health category plays a very important role in maintaining balance and well being H ealth categor y measurements occur several ways, by using reliable and valid self reports or with concrete, objective lab testing. The latter being much more intensive and time consuming as well as difficult in getting the appropriate number of subjects in adequate amounts of time. According to Healt hy People 2010, health status can be measured not only by quality of life but also by variables such as birth rates, death rates, life expectancy, morbidity and risk factors for disease as well as the current status of things like health care and health i nsurance .49 Health y People 2010 mentions three top or leading health indicators : level of physical activity, overweight and obesity, and mental health levels 49 Although Healthy People 2010 talks about the general population, many health care providers are overworked and do not present a healthy balance between work and personal life leading them to fall into the realm of decreased activity, overweight/obesity and decreased mental health levels. The imbalance may be linked to several detrimental outcomes such as: decreased physical or social health, which in turn can possibly lead to professional burnout and other health concer ns. Allied H ealthcare P rovider s Burnout Many professions experience some type of burnout and ultimately a decrease in one of the health domains. The following allied healthcare professionals compare similarly to athletic trainers in their job description s and job setting s as well as similarities in education and job roles. A brief description of physicians, physicians assistant s and physical therapy burnout and wellness is included to give a basic background of the current literature in other professions outside of athletic training

PAGE 33

33 Physician Burnout Many health care professionals including physicians face job stresses that lead to burnout .12, 13, 35, 36, 50 The physician has a lif e filled with many factors related to burnout, which leads to decreases in mental health and related declines in physical health. The gold standard for measuring burnout, the Maslach Burnout Inventory or MBI measures burnout in physicians.12, 35, 36, 50 Physicians, like other health care professionals, are in a unique and critical situation in regards to burnout. Their decisions or reactions could have devastating consequences for those patients w ho are in contact with them for their services .13 The burnout of physicians relates to patient care and health of the physician. The decrease in mental health as a result of burnout affects physical and social health. In the studies on physician burnout the authors use several different characteristics to define specifics of physicians, such as specialization and geograph y One study by Ozyurt et al. looks at predictors of burnout in Turkish physician s.36 The instruments for this stud y included the MBI and the Minnesota Satisfaction Questionnaire (MSQ). The MBI was a reliable measure of burnout with Cr onbachs A lpha coefficients at 0 .81, 0 .70 and 0 .73 for the three subsections of the MBI .36 The three respective subsections include: emotional exhaustion, depersonalization and personal accomplishments. The Cronbachs coefficient for the MSQ showed reliability with a score of 0 .88 36 The results showed those with higher number of shifts had significantly higher scores on emotional exhaustion and depersonalization and lower scores on personal accomplishment or satisfaction .36 In the Ozyurt study the number of vacations taken per year was found to be significant as a variable for every subscale of the MBI .36 Acco rding to this study, the authors found similar findings Other studies state that insufficient personal time or vacation were predictors of burnout as well as working in the public sector and higher number of shifts .36 Shanafelt et al. used the MBI to measure burnout in internal medicine residents .12

PAGE 34

34 The author found that 76 % of the sample was found to meet the criteria for burnout. The associated stresses that lead to burnout were similar to the Ozyurt study, such as inadequate sleep, lack of leisure time, and long shifts over 24 hours .12 According to Ozyurt, a ll of these results are consistent with other studies conducted internationally. Due to the stresses physicians experience, burnout occurs in many settings and in many countries, but is measured with the same instrument, the MBI .12, 13, 36 Secondary to burnout, exercise plays a pivotal role in health and wellness. P rimary care physicians who performed aerobic exercise regularly were more likely to counsel their patients about aerobic exercise.51 Those physicians that performed resistance training program were also more likely to counsel patients about participating in resistance training .51 According to a study by Peterson et al. current medical students are already falling behind in health levels while attending medical school .52 When medical students do not participat e in exercise or activity, burnout can begin early and cause detrimental effects on physician as well as client health. P hysician A ssistant Burnout The physician assistant (PA) is a nother health care profession al that faces the same burnout syndrome as physicians and athletic trainers These health care providers work alongside doctors performing many of the same tasks and dealing with the same types of job related pressures. Minimal research exists on physician assistant burnout. The only study done by Bell et al. in 2002, used the MBI to determine the l evels of burnout in emergency room PAs (EMPA).15 Specifically, 66 % of EMPAs showed moderate or high depersonalization and 59 % showed high emotional exhaustion .15 In the study, burnout was found to be a problem among emergency room PAs, but not as significant as in ER doctors .15 Literature states the problems of burnout in physicians occur due to certain stresses. PAs experience some of the same stres s ors as do other health care professions, leading to a decrease in well being.

PAGE 35

35 Physical T herapist B urnout Physical therapists (PTs), a third health care profession that provides orthopedic care for patients can be exposed to the stresses that cause burno ut syndrome and problems in well being. PTs work in a variety of setting s including hospitals and clinics. A 1 983 study stated that 53 % of PTs reported burnout .37 In this deca de the current literature of physical the rapy burnout rates increased drastically. A study conducted by Balogun in 2002 revealed that the percent of moderate and high emotional exhaustion found when using the MBI rose to 88 % of the sample .14 Balogun stated that the emotional bu rden of heal th care can possibly lead the health care provider to become frustrated causing them to develop negative attitudes towards their clients and the work environment itself .11 Of the therapists listed in the study over 50 % felt highly emotionally overextend ed and most felt negative attitudes towards both work and clients .14 The studies by Balogun and Shuster used the gold standard of burnout the MBI as the valid instrument .14 37 As a result of understaffing in clinics and hospitals due to the shrinkin g health care dollar which allows for more patients needing to be seen by fewer PTs, burnout continues to be a problem.11, 33 According to the study done by Donahue et al. the physical therapists in their sample showed a moderate level of burnout in only four years of working in the profession .33 Donahue et al. reported that 45 % of respondents sc ored in the high range for emotional exhaustion, 20 % high on the depersonalization and then 60 % in the low of personal accomplishment .33 T hese number s reveal that a higher number of PTs feel burnout in their current lives. Like physicians and physician assistants a level of burnout creates a decrease in the quality of care and well being .14, 33 Physical therapists and athletic trainers experience sim ilar job characteristics and play similar roles that can lead to burnout. The roles of PTs and athletic trainers mimic each other in the clinical interactions of clients with orthopedic injuries and the goal of return ing them to their

PAGE 36

36 functional levels pr ior to injury. Client work, especially those that need to return to work or even participate in daily life normal activities can be extremely stressful for the healthcare professional, particularly when coaches, parents or even insur ance companies are pre ssuring for the patient to return to full function Athletic Trainers Burnout and Wellness Burnout in the A thletic Trainer Athletic trainers (ATC), also listed as allied health professionals, practice in an intensely stressful medical setting, work ing l ong hours often under stressful conditions .2 5 The exact numbe r of ATCs experiencing burnout is unknown; what is known is that ATC s do experience burnout syndrome and a few studies have discussed possible coping methods or personality traits that have allowed ATCs to continue to function in their work environments .3, 32 Athletic trainers in particular, work in very stressful, and life threatening situations sometimes under coaches and athletic departments that place unnecessary pres sure on the athletic trainer to return key athletes to play.10 Athletic trainers interact with a number of other professionals in addition to a number of injured athletes, leading to more potentially stressful interactions.3 With the amount of stresses and the number of hours worked there is little confusion as to why burnout exis ts in the profession of athletic training The original burnout/stress management articles in athletic training originate from the early 1980s and from the same author. Geick began studying the burnout syndrome in athletic trainers in 1982. Geicks fi rst article was a collection of case reports that exhibited some of the early warning signs and symptoms of the burnout syndrome.4 According to Geick, s igns and symptoms can be physiological, psychological and behavioral in nature. Physiological signs include: increased pulse rate, headaches, gastrointestinal or cardiovascular disturbances and fatigue. Psychological signs include: anxiety, depression, sleeplessness, and ev en sexual

PAGE 37

37 dysfunction .4 These signs and symptoms may result in the athletic trainer turning to drugs, alcohol or tobacco as a reaction to the burnout .4 A study by Hendrix et al. used several instruments such as the MBI, the Hardiness test and the Perceived Stress Scale to determine burnout and other personality traits that lead to or decreas e burnout .3 Hendrix et al. studied 118 certified athletic trainers working in NCAA universities that sponsored football, with approximately 50 % of the s ample population being male.3 The study results disclosed that perceived stress was a predictor of emotional exhaustion. Specifically, those with decreased hardiness scores and low s ocial support reported higher levels of perceived stress and developed poor lifestyle behaviors.3 A more recent study o n AT burnout by Clapper et al. found the Athletic Training Burnout Inventory (ATBI) to be reliable, but did not rep ort the actual burnout scores.9 The informati on that was reported stated that the burnout results were similar to past research and that ATCs experienced higher perceived stress and therefore higher emotional exhaustion and depersonalization scores.3,9 Clapper report ed that age was a significant variable for organizational support .9 However, neither marital status nor gender showed any significant differences in regards to organizational support.9 The most recent burnout article by Kania et al. discusses the personal and environmental characteristics that predict burnout in N CAA athletic trainers. The researchers contacted 600 NCAA collegiate athletic trainers working as clinical athletic trainers. Burnout was measured using the Maslach Burnout InventoryHuman Services Survey (MBI HSS), which is a modified version of the orig inal MBI.10 Subscales for this instr ument are the same as the MBI in regards to emotional exhaustion, depersonalization and personal accomplishment. Of the 600 contacted ATCs, 206 of them participated in the study. Fifty two percent of the participants were male, 95

PAGE 38

38 % were Caucasian and 47 % were married. Interestingly, 80 % of the sample held at least a masters degree.10 Information from the MBI HSS brought to light different results in comparisons to other research findings. Overall burnout scores for this population were considered low with only 66 of the 206 falling in the high, average or low burnout score ranges.10 Researchers found that personal characteristics such as highest degree attained, race, years certified and gender and environmental characteristics like injury type frequency and pressure from coaches can possibly predict burnout.10 Specifically, personal characteristics were ab le to account for 4 5 % of the variance in emotional exhaustion, 21 % in depersonalization and 25 % in personal accomplishment. The stress level was the significant predictor of burnout in athletic trainers. Environmental characteristics accounted for much less of the variance and the highest portion of the variance was 17 % for emotional exhaustion. Specifically, coach pressure, number of athletes and injury type and frequency were the environmental characteristics that were significant predictors.10 In the November 2000 Hunt article from t he NATA (National Athletic Training Association) news, the author writes about the topic of burnout. K ey points assoc iated with burnout among athletic trainers w e re discussed Those points being the number of w ork weeks spent working more tha n 40 hours per week, and focus ing on athletes and administrators, parents, coaches and colleagues. With so many roles and responsibilities, the author points out the dwindle in family and personal time for hobbies and other activities .2 In Geick s article the author discusses was to m anage the burnout syndrome including things like the ability to recharge at the end of a season, refocus ing on the humanistic approach of the job, interact ing in positive ways to ensure positive relationship s with staff and clients,

PAGE 39

39 reexamining time allotm ents and daily job duties, and learn ing to delegate responsibility to other staff and students .4 All of these things can and ha ve allowed the athletic trainer to survive t he stressful profession. Burnout syndrome is a preventable issue that requires a conscious effort. The ways that Geick suggested to handle burnout include: an active outside life, proper health habits, restructuring of behaviors, and analyzing job stre sses .4 The author states that the control of ones environment allows for control over ones job. The successful ATC has a mixture of variables that give him or her the ab ility to avoid burnout. Some of those variables fall into a spiritual or moral realm. Other variables are as simple as scheduling leisure time and pleasurable activities even if the activities are physical in nature .4 Athletic training involves many stressful situations and a need exists for stress management within the profession. M ental H ealth in A thletic Trainers Many circumstances and situations lead t o burnout in athletic training. A fter having discussed some of the causes of burnout that cross most medical professions, the focus turns to those factors that cause burnout, and ways to eliminate or reduce these contributors to burnout in athletic trainers. One of th e first articles on athletic trainer burnout discusses poor or negative relationships within the work place, be it athletes or athletic directors .5 This is only one aspect of what causes burnout and decreases mental health in th e athletic training profession. Other facto rs that contribute to burnout are the number of athletes that the athletic trainer is in contact with daily, the total number of hours worked in the training room, and the relationships that develop within the work place .3 A key concept to remember is th at different stressors occur in different settings and preparing an athletic trainer for each settings diff erent work dimensions may be instrumental in reducing th e burnout that some athletic trainers experience in the workplace.

PAGE 40

40 Physical A ctivity in the A thletic Trainer Athletic trainers are no different tha n other s amples of the population and should be monitored for obesity and decreased well being The current research on athletic trainer s fitness and activity levels is minimal. To date only one study discusses the fitness levels of athletic trainers .6 The ATC needs not only the minimal levels of act ivity for health, but their job roles and responsibilities require them to have at least some level of fitness .6 The literature states that those athletic trainers working in clinical settings ar e more likely to be physically active outside of the workplace .6 This is most likely due to the decreased number of hours clinical athletic trainers work. Athletic trainers that work in sport settings may find it difficult to have a regular exercise routine due to the travel h ours and lack of control over their work schedules .6 The study conducted by Cuppett and Latin included 636 a thletic trainers in the Midwest region, and showed that male ATCs were no higher in physical activity levels than th ose of the general population. However, female athletic trainers did show more activity th a n the general population .6 Since the athletic training student is exposed to all of the benefits of physical activity in their education curriculums as mand ated by the National Athletic Trainers Association Educational competencies it is of grave concern as to why more athletic tra iners are not physically active.53 Recentl y, a second study was conducted to determine the health and fitness habits of athletic trainers. The author contacted 1000 athletic trainers randomly throughout District 4 of the NATA. Only 275 athletic trainers responded, yielding a 27 % response rate.10 Approximately 50 % were females and the mean age was 34. Interestingly the auth ors found that BMI or Body Mass Index in this sample was high for females ; 47 % were considered overweight or obese and 74 % of males either reported being obese or overweight.10 BMI number s that high reveal that

PAGE 41

41 athletic trainers are currently not in a health y weight range and are predispo sing themselves to a variety of diseases. The amount of hours worked for athletic trainers change in regards to whether a sports team is in -season or out of -season. Those athletic trainers who are considered in -season tend to work more hours due to prac tice, travel and increased number of competitions. The authors found a small difference in health habits in regards to in -season versus out -of -season time frames.10 Forty five percent of the athletic trainers in the sample who reported being in -season reported he althier habits during their out of -season time period. Forty seven percent of those who report ed being out -of -season reported no difference between seasons.10 The BMI results show that ATCs who are not in -season are still not able to maintain a healthy weight. In the sample only 41 % met the ACSM recommended guidelines for exercise (30 minutes of exercise, 5 days per week).10 Another interesting finding from Kania et al. was the lack of nutritional health and the use of alcohol and tobacco from respondents The author found that in general, certified athletic trainers did not meet the recommended guide lines for nutrition as recommended by the USDAs Dietary Reference Intake (DRI) for individual food groups.10 Approximately, 7 % of female AT C s and 3 % of male ATCs reported consuming more than the USDA recommended amount of alcohol (0 1 drinks for females and 1 2 drinks for males).10 Eleven percent of the total sample reporte d not consuming alcohol at all, and a very small percent age (1%) reported any type of tobacco use. It seems that use of tobacco and alcohol products among athletic trainers is lower than the general population The author suggests the time commitments as possible reason s for higher BMI, and lower nutritional intakes among ATCs

PAGE 42

42 Barriers to P hysical H ealth in A thletic Trainers Health y People 2010 cites lack of time, lack of access to facilities and lack of safe areas to be active as main reasons or barri ers to physical activity .49 In Cuppetts study t he author attr ibutes the barriers to activity in athletic trainers as a lack of control over work schedule, citing specifically barriers like team travel time and practice times. Cuppett also mentions responsibilities of teaching as another possible barrier .6 In general, m ost if not all health care professionals work in some type of stressful environment and work long shift s which contribute to a lack of time for physical activity. These factors lead to obesity and morbidity, as well as decreased family and social interaction opportunities. Social S upport Levels of the A thletic Trainer The current literature on social sup port relating to athletic trainers is minimal. The literature in athletic training offers a look at social support and perceived stress.3 The conclusion of the study by Hendrix revealed that decreased social support combined with decreased hardiness showed greater levels of perceived stresses.3 Th e specific social support instrument used by Hendrix was the Social Support Questionnaire and was a section of a burnout study in athletic training To date this is the only study to look at socia l support in athletic training.3 Domains of Health and Wellness Physical H ealth Physical activity is a key component in reduc ing m any chronic illness es and conditions such as diabetes, cardiovascular problems and obesity. The total medical costs and loss of productivity as a result of obesity was an estimated 99 billion dollars in 1995.49 Physical activity plays a role in enhancing psychological well being and prevention of premature mortality.49 Activity levels of 2030 minutes per day can reduce the chanc es of these conditions and illness es In fact according to the CDC, the majority of people in the United States do not do enough

PAGE 43

43 physical activity to ward off such conditions and illnesses .54 H ealth care professionals, who promote health and well being in their professional careers are part of the this majority .6 Physical health can be measured in several ways; includ ing the use of surveys that are a form of self rated health measure or self rep orted instruments 55 Other ways include taking actual health measurements like blood pressure, height and weight and using those measurements to determine BMI, chol esterol levels or even through the use of job stress and burnout surveys. A measure like the Behavioral Risk Factor Surveillance System asks participants about physical activity levels along with health status and health related quality of life, hypertens ive awareness and cholesterol awareness in order to determine an overall health level 54 T here exist s many different self reported fitness instruments. Some commonly used valid and reliable tools for measuring fitness levels include: the International Physical Activity Questionnaire and the Baecke Q uestionnaire on Habitual P hysical A ctivity 6, 5658 The Baecke questionnaire has been proven valid and reliable in several studies (see Appendix A). 6, 5658 A concern with these survey tools stems from the specific attention it pa ys to one area of well being over another. Many instruments choose to focus on one aspect of health or even one particular aspect of activity, like work related activity versus actual leisure time activity.59 Mental H ealth Physical health is not the only part of wellness or well being. Mental health plays a vital role in wellness. According to some research mental health can have more th a n one construct with a breakdown in psychological, emotional or intellectual wellness levels .60 Mental health encompasses direct illnesses such as depression, anxiety, autism and schizophrenia. Another important contributing factor to mental health measure associated with many health care professionals is the condition known as burnout .25

PAGE 44

44 Burnout is comprised of three core dimension s and defined as a prolonged response to stressors on the job .25 Those dimensions according to Maslach include: emotional exhaustion, depersonalization, and reduced personal accomplishment .25 Burnout occurs as the result of quantitative job demands, such as the number of clients or hours worked. Another factor is the actual job itself including coworkers, client selection or role conflicts .25 A final category is the organization of the job including things such as downsizing or mergers which can c ontribute to job burnout .25 Personality characteristics contribute to burnout. Demographics such as age and educat ional levels have been found to relate to burnout. Other factors associated with the individual include personality characteristics, such as hardiness or attitudes related to work .3, 25 Another aspect that may lead to burnout and decreased mental health is inadequate levels of social support levels .25 Social H ealth Social health is one domain of overall health or wellness. Social health or wellness can be defined as the perception o f having social support .60 Social support is found in relationships w ith family and friends, not just coworkers. Social support is believed to positively impact a persons mental and emotional wellness44 and should be studied along with the physical and mental domains as well as the impact social support has on burnout and perceived wellness. Social suppo rt can be broken into two approaches: 1) the functional support or help one receives from interpersonal relationships, and 2) the structure of the interpersonal relationships.44 These approaches of social support are part of the overall concept of whole health or wellness and relate to ot her health problems. Social support and the social networks or relationships associat ed with social support can be an important fa ctor in avoiding things like burnout or decreased wellness.

PAGE 45

45 Theoretical Framework Theory of Action is a theory that pro poses actions determine all deliberate human behaviors. The Theory of Action is actually two separate theories called the Theory in Use and the Espoused Theory The Theory of Use is defined as what the person actually does, where as the Espoused Theory is what the person thinks he or she does.61, 62 These theories serve as mechanisms to link thought and action together. The use of these two theories can bring to light what one thinks when actually describ ing health and what one actually does towards health or wellness.61, 62 Many health care providers such as physicians, physical therapists and athletic train ers promote health or well being to their clients and athletes. Health care professions worry about client well being, but are proven to neglect the undertaking of proper steps for their own well being by being the target and at the forefront of many burnout and health studies.3, 4, 6, 11 15 Recently, an article discussed the habits of athletic trainers and showed that many certified athletic trainers had better health and fitness habits th a n the general population, but still did not meet recommended times and intensities by the ACSM.63 Summary In summary, all of the previously mentioned health care professions have some type of burnout present in their daily lives.35, 9, 11 15 No empirical instrument takes into account the burnout syndrome or mental health state and combines the data from the MBI and other measures of w ell being, such as social and physical well being. As an athletic trainer, each area of well being is important to measure as well as determining barriers or related problems to the various aspects of well being. A n instrument that would incorporate the multiple aspects of health while determining burnout levels would help with creating an overall measure of well being of the athletic trainer With such an instrument the potential of looking at the well being of other allied healthcare professionals is a lso increased Measurements of well being in the athletic

PAGE 46

46 training profession and understanding the potential barriers of wellness and attitudes of athletic trainers can lead to reductions in chronic health issues and lead to an athletic training professio n that presents decreased levels of burnout and increases in overall health or wellness.

PAGE 47

47 CHAPTER 3 METHODS Study Setting and P articipants There are approximately 30,000 a thletic trainers working in the United States The athletic training population s pans the entire United States as well as other countries throughout Europe and Asia. Within the United States, members of the National Athletic Trainers Association (NATA) are split into 10 districts. D ata was collected from one of the districts (Distri ct 9) There are a pproximately 3000 members in the South e as tern Athletic Training Association ( District 9) O f these 3000 members, approximately 14 00 (~ 46 % ) have opted out of being contacted about survey instruments. Members of the NATA have the option to maintain privacy from solicitors or marketers a nd t h u s only 1 600 members from District 9 were contacted in this study. ATCs in District 9 were contacted from different job settings constituting a purposeful convenience sample Students of any standing we re excluded from the sample because they d id not work full time due to their lack of awareness and inexperience with certain aspects of the profession Th is population may be studied in future projects. The selection of participants for this study encompa sses settings where certified athletic trainers work with some clientele such as student, athlete or other type of patient These settings include: professional s p orts, high schools, community colleges, universities sports and educational settings indus trial jobs and clinic al work performed within District 9 Study Design A cross sectional web based survey design was used to determine the relationships between the m easure s of perceived wellness and burnout including the three domains of health: socia l support, mental health, and physical activity levels among athletic trainers. The goal wa s to determi ne the relationship between social support, physical activity, mental health and

PAGE 48

48 perceived wellness and burnout levels as well as t he direct and indirec t effects on perceived wellness and burnout D emographic data including age, gender, years of work experience, hours worked per week, and job settings was collected and the hours worked and years experience was included in the structural equation model pat h analysis. Materials A survey instrument w as developed from existing valid and reliable instruments to measure perceived wellness, burnout physical activity, mental health and social support in athletic trainers (See Appendix A). The instrument took approximately 15 minutes to complete and was emailed to each participant with a direct link to the survey website SurveyMonkey.com (Survey Monkey Portland, OR). Instrumentation Valid and reliable survey instruments were selected from the literature to me asure the domains of wellness ( mental, physical social ), burnout, and perceived wellness in AT profession als22, 38, 44, 5660, 6469 The following instruments were combined and used in their entirety: Baecke Physical Activity Questionnaire (BPAQ), Mental Health Inventory 5 (MHI 5)*, M edical O utcome S tud y: Social Support Survey (MOS -SSS)*, Copenhagen Burnout Inventory (CBI), and Perceived Wellness Survey. The final instrument contained 95 closed ended Likert type items. The survey was organized so that each instrument formed a distinct subscale intended to measure constructs of interest. Oblique principal component cluster analysis was performed All items loaded on the appropriate subscale The measurements of the wellness domain constructs were used to determine predictive and causal relationships to th e variables of perceived wellness and burnout The instrument s scale and subscale validity and reliability have been previously established. 22, 38, 44, 56 60, 64 69 See A ppendix B for a copy of the survey and T able A 1 in Appendix A for validity and reliability scores Cronbachs A lpha w as used to

PAGE 49

49 de termine the internal consistency of each separate instrument It must be noted that establishing survey reliability and validity is critical to advancing this line of inquiry. Continued work in defining an appropriate instrument predicting burnout and wel lness among athletic trainers is essential. Demographics General demographic questions included information such as: gender, age, years of job experience job setting, primary job title, and a t the average number of hours worked per week over the last mo nth. M eans standard deviations and frequencies were reported for these variables. Physical Activity P hysical activity w as assessed with the Baecke Physical Activity Questionnaire (BPAQ). This instrument measure d the perceived physical activity over th e past 12 months .5759 The Baecke Questionnaire contains three parts or indices .59 This instrument was able to report total physical activity levels for the athletic trainer as well as subscale scores for work, sport/exercise and leisure time activity. The first index ( subscale ) is the Wor k Index and that measures the amount of physical activity that a person engages in during the work day at their place of employment. Item 1 asks the person what is the main occupation. For Item 1 the investigator ha d to determine the category of job report ed by the participant for the question. Based on the article by Baecke, the job categories are low level (clerical, driving, teaching etc.), middle levels (factory work, plumbing, carpentry etc), and high level (dock work, construction work and sports).59 The athletic trainer was determined to be part of the middle level (due to the physical nature of th e job, carting coolers, kits etc.) and those that held academic or administrative positions were predetermined to be a low level. This determination for athletic trainers was based on the examples given by

PAGE 50

50 Baecke et al for teachers at the low level and fa ctory/carpentry work for the middle level.59 The other seven items in the work index were computed by questions about body position at work (e.g. at work I sit, or at work I stand with the answers: never/seldom/often/very often ) and items about loads lifted and tiredness (e.g. at work I lift heavy loads, or at work I am tired ) Item 2 was first subtra cted from six and then items 1 through 8 were totaled and finally divided by eight to give the work index.59 This subscale was calculated according to Baeckes calculation of scores of the indices of physical activity.59 The next subscale the Sport Index, ask ed each participant, do you exercise? (If answered ye s the respondent was asked what do you do and how often? ). T his question measure d the amount of intensity, time ( H ow many hours per week) and amount of time during the year (H ow many months per year). (The investigator d etermine d the intensity of the sport based on Baeckes guidelines ; low level (billiards, bowling, golf etc.), middle level (badminton, cycling, dancing swimming etc.), and high level (basketball, rugby, rowing etc.).59 Once intensity level was complete, the amount of time per week was coded as such (<1/1 2/2 3/3 4/>4 with the corresponding scores, .5 1.5 2.5 3.5 4.5). The number of months per year wa s coded as follows (<1/1 3/3 6/7 9/>10 with the corresponding scores .04-.17 .42 .67 .92). This item wa s asked twice, calculated twice, and coded as a score between 1 and 5. The highest score was reported as the item score for this specific item. Four other items ask ed about the comparison to others and exercise as part of leisure time ( e.g. D uring leisure time I sweat, and during leisure time I play a sport, with the answers: never/seldom/often/very often).59 Items 9 through 1 2 were summed and divided by 4 according to Baeckes calculation of scores of the indices of physical activity.59 The final index or subscale i s the Leisure time Index, which asks four items on leisure time activities like television watching, and questions about leisure walks or cycling ( e.g. D uring

PAGE 51

51 leisure time I watch TV or during leisure time I walk or cycle, with the answers: never/seldom/often/very often).59 Item 13 was subtracted from 6 and t hen items 13 through 16 were summed and divided by 4 according to Baeckes calculation of scores of the indices of physical activity.59 Upon completion of the subscales, a total activity score wa s calculated by summing the item s on each index for that indexs total score. Each subscale had a max possible score of five with five being high physical acti vity. The total score was calculated as a sum of the three indices or subscales and a max score was 15. See the complete instrument in Appendix B. Mental Health The Mental Health Inventory 5 (MHI 5), a subscale of the Short Form 36 (SF 36), w as use d to measure overall quality of life and more specifically feelings of depression and the presence of self -efficacy and self -worth The instrument uses items that have a six point Likert scale (e.g. H ow much time during the past month have you been a ver y nervous person?) .64, 65 The items s core s are : (6)= All of the time, (5)= Most of the time, (4)= A good bit of the time, (3)= Some of the time, (2)= A little of the time, and (1)= None of the time For items 3 and 4 the scoring i s reverse scored. The total score is determined by totaling the five scores and dividing by 25 and then multiplying by 100. This gives a maximum score of 100 and according to the study by Hoeymans et al. a cut -off score of less th a n 72 was considered unhe althy or presenting with anxiety, depression behavioral or emotional problems and psychological well being 70, 71 See Appendix B for the instrument. Social Support The M edical O utcome S tudy: Social Support Survey ( MOS -SSS) is an instrument that was developed as part of the Medical Outcome Study (MOS) The MOS was an observational study that examined variations in patient outcomes and doctor practice styles.67 The MOS SSS

PAGE 52

52 m easure s the perceived amount of social support in the forms of companionship, assistance or support received by family, fr iends and coworkers on a five point Likert scale (e.g. H ow often is each of the following kinds of support available? Is there s omeone you can count on to listen to you when you need to talk ? ).44, 66, 67, 72 The response items are: (1)= None of the time, (2)= A little of the time, (3)=S ome of the time, (4)= Most of the time, and (5)= All of the time. There are four possible subscale totals, and one overall support to tal. The overall support total wa s calculated according to the RAND health website and the Sherbourne article as t he scores of all the items minus the minimum score divided by the maximum score minus the minimum score all multiplied by 100.44, 67, 72 The total sc ore computation allows for comparison to other research with this instrument and to the scores from the 18 item version and the 4 item version. Scores were calculated for all subscales including: emotion, tangible, affectionate, and positive social. Burno ut ( P rofessional and P ersonal) Burnout (physical and psychological exhaustion that results from work or other life situations) is measured with The Copenhagen Burnout Inventory (CBI).22 The CBI has a total number of 19 items and is used to determine the total level of burnout including the three areas of burno ut (personal, work and client burnout).22 The total score is a total average of the three subscales. The items for the first section of the CBI instrument focus on personal burnout (e. g. H ow often do you feel tired? and H ow often are you physically exhausted? ). The responses to this part of the instrument us e s a five point Likert scale with the answers of (100)= Always, (75)= Often, (50)= Sometimes, (25)= Seldom, (0)= Never/almost never .22, 38 This section can be subtotaled into an average score between 0 and 100. The second part of th e instrument focus es on work burnout items (e.g. I s your work emotionally exhausting? and D o you feel worn out at the end of the working day? ).22 These

PAGE 53

53 items are answered with a five point Likert scale with the answers of (100)=Always, (75)=Often, (50)=Sometimes, (25)=Seldom, (0)=Never/almost never or the follo wing answers: (100)= To a very high degree, (75)= To a high degree, (50)= Somewhat, (25)= To a low degree and (0)=To a very low degree.22 This section can be subtotaled into an average score between 0 and 100. The final section of the CBI focuses on the aspect of client burnout (e.g. D o you find it hard to work w ith clients? and D o you feel that you give more tha n you get back when you work with clients?). These items are answered with a five point Likert scale with the answers of (100)=Always, (75)=Often, (50)=Sometimes, (25)=Seldom, (0)=Never/almost never or the following answers: (100)=To a very high degree, (75)=To a high degree, (50)=Somewhat, (25)=To a low degree and (0)=To a very low degree.22 This section can be subtotaled into an average score between 0 and 100. Perceived W ellness Perceived wellness is multidimensional and is derived from the combination of many construct or subcategories that include but are not limited to physical indicators of health such as blood pressure and cholesterol and psychological constructs like mental well being and life satisfaction. The Perceived Wellness Survey follows the multidimensional approach and incorporates the following constructs or parts of health: physical, spiritual, psychological, social, emotional, and intellectual.60 Similar to the previously described survey instruments a six point Likert scale for the Perceived Wellness Survey (PWS) measure s the p erception of wellness in the participants. The P WS is comprised of 36 items that are split into six items per construct of health. Sample items for each section are as follows: psychological (e.g. I am always optimistic about my future ), emotional ( e.g. I sometimes think I am a w orthless individual), social ( e.g. M y friends will be there for me when I need help), physical ( e.g. I expect my physical health to get worse), spiritual

PAGE 54

54 (e.g. I believe that there is a real purpose for my life), and intellectual ( e.g. I avoid activities t hat require me to concentrate).60, 68 The items for this instrument are answered using a five point Likert scale with the following answers: (1 ) Very strongly agree to (6) Very strongly disagree. Scoring for this instrument is somewhat complicated. P lease see A ppendix B which presents the instrument and Table A 2 for specifics to scoring.60 Scoring of I nstruments Each instrument has a total index or composite score that can be used to determine overall outcomes for each instrument. Most of the instruments have specific sub scales with in the overall index or composite score and can be further broken down into components of each particular area of the instrument. The composite scoring methods can be found in Table 2 ( See T able A 2 in Appendix A for index and composite scores ).22, 44, 59, 60, 67, 68, 7072 Pilot Data Collection A pilot sample consist ing of 23 ATCs outside the NATA District 9 were contacted through email to inquire about participating in the pilot study. Fourteen certified athletic trainers from outside the NATA District 9 completed the survey instrument twice, and approximately one week apart. Nine of the original 23 ATCs either did not complete the second collection or did not complete any of the instruments. A Pearson Correlation was used to determine test retest reliability for this data and instrument. The test retest results revealed that the instruments had rep eatability. Data C ollection Each of the surveys was used unchanged and entered into Survey Monkey for W eb -based access. All Survey Monkey usage fees and sampling fees we re paid through a grant received from the Southeast Athletic Training Association (SE ATA) The NATA was contacted in order to supply the email addresses for the members of District 9. The cost per email address was $.09.

PAGE 55

55 Those members of District 9 willing to participate in survey research received a n email during the fall semester sports season on Wednesday, Sept 3, 2008. from the researcher To ensure duplicate surveys were not distributed to the same individuals, t he confidential email survey linked to the participants email address so that each participant could only participate in the survey once. No incentives were provided to the participants for taking the survey. The District 9 members w ere emailed with explicit instructions, and participants were given six weeks to complete the survey. Follow up requests w ere made to the targeted individuals one, three and five weeks after the original request.73 See appendix C for a copy of the institutional IRB and Consent forms. See appendix D for a copy of the letter sent to each athletic trainer prior to answering the survey. D ata A nalysis Pa rticipants survey responses w ere automatically entered into a Microsoft Excel database (Excel 2003 Microsoft Redmond, WA) from a SurveyMonkey .com administrator (Survey Monkey Portland, OR) The data in the excel file w ere checked for missing data Co mposi te score s and subscales were calculated and then imported into Statistical Package for the Social Sciences or SPSS (Version 16.0 SPSS INC. Chicago, IL) database. Missing data were replaced using the SPSS replace missing values function which replaces the missing value with a mean score calculated by SPSS for that item from all other participants Subject demographics were calculated including: desc riptive statistics of central tendency and frequencies were conducted on the demographic data A correlation analysis was run to determine relationships between burnout, perceived wellness and the wellness domains. Subsequent analysis based upon subject demographics (i.e. job title, age, gender, education level, etc ) w ere conducted accordingly to determine fact ors of covariance. Significance wa s set at p<.05. A Structural Equation Modeling technique or path analysis, using the SAS statistical program 9.2 (SAS

PAGE 56

56 Corporate Statistics, Cary, NC) determine d the extent to which each variable is associated with the out come measures of burnout and perceived wellness (Please see figure 3.2 for the anticipated path associations). All path analysis and g oodness -of -fit statistics w ere run with the consultation of a statistics consult ant located in the Collaborative Assessme nt and Program Evaluation Services (CAPES) of the University of Florida, located in Gainesville FL and paid for through a research grant received from SEATA. Descriptive Statistics Descriptive statistics of central tendency and frequency were determine d for all of the variables including the demographics of age gender, job title, job setting, hours worked per week, and years experience, as well as for perceived wellness (PW) burnout (BO), physical activity (PA) social support (SS) and mental health ( MH) Coefficient A lpha Cronbachs A wa s used to determine th e reliability of the items of each separate instrument in the total instrument using SPSS (Version 16.0 SPSS INC. Chicago, IL) A perfect a lpha score is 1 and the closer the score is to 1 the better the internal consisten ce. Although there is no set level that is acceptable 0 .7 or better is considered sufficient. See results for internal consistence reliability and test retest reliability. Pearson Correlations Pearson correlation s were run to determine the positive or negative relationships between the following variables: years of experience, hours worked per week, physical activity (PA), mental health (MH), social support (SS), burnout (BO), and perceived wellness (PW). Overall scores for each variable w ere used t o determine the correlations. Correlations analysis was used to determine what relationships existed and to what ext e nt these relationships occurred among the variables.

PAGE 57

57 Multiple Regression s Multiple regressions analysis was calculated to determine t he predictability of select variables in this study. The first regression model used the outcome or depend e nt variable of burnout with the explanatory or independent variables of gender, job title, job setting, age, hours worked per week, years experience, mental health, social support, physical activity and perceived wellness. The second regression model used the outcome or depend e nt variable of perceived wellness with the explanatory or independent variables of gender, job title, job setting, age, hours w orked per week, years experience, mental health, social support, burnout and physical activity. The regressions analysis was used to determine variables that could predict both burnout and perceived wellness in athletic trainers Path Analysis (SEM) Upon completion of the covariance matrix a path analysis was run using the SAS statistical program 9.2 (SAS Corporate Statistics, Cary, NC) to determine effects of the following independent variables of social support, mental health, physical activity, years of experience and hours worked per week on burnout and perceived wellness. A two step approach to structural equation modeling recommended by Kline was used .74 The first step was to test the fit of the measurement model. Once the measurement model was solidified, a structural model was tested. A model was solidified based upon appr opriate goodness -of -fit indices. The path models were run on the covariance m atrix of the variables. Based on theoretical considerations, four latent variables (perceived wellness, burnout, physical activity, social support) one observed endogenous variable (mental health) and two exogenous observed variables (years experience and hours worked per week) were considered in the path model. In the original measurement model the subscales for each latent variable served as the indicators or observed variables for each of those latent variables. Mental health did not have subscales an d therefore was

PAGE 58

58 considered an observed variable without indicators in the model. See Figure 3.2 for the proposed path model. Maximum likelihood estimation was used to estimate the model and to determine the goodness -of -fit of the model by simultaneously analyzes all the endogenous variables.74 Goodness -of -fit indices were calculated to determine the extent to which correlations from data match the outcomes from the SEM. Overall fit of the models was tested using the goodness -of -fit index (GFI), c omparative f it i ndex (CFI), adjusted goodness -of -fit index (AGFI), standardized root mean square residual (SRMR), norm fit index (NFI), parsimonious NFI (PNFI), and the minimum Chi -square fit index.74, 75 Acceptable standard s of fit are GFI 0.90, AGFI 0.90, CFI 0.90 NNFI 0 .9 0 and SRMR 0 1 0 .74, 75 Path coefficients ( structural residuals, and squared multiple correlations were calculated for each structural pathway. Significance was set at p < .05 for all statistical analysis T test/ANOVA A biv ariate analysis or independent t test w as used to determine the effect of gender on burnout perceived wellness, social support, physical activity, mental health, hours worked per week, and years of experience One -way ANOVAs were used to determine the effects of job setting and job title on th e depend e nt variables of burnout, perceived wellness, mental health, physical activity and social support. The Bonferroni post hoc test was used as the follow up tests for the significant One -way ANOVAs. All analyses were run with the SPSS statistical soft ware program (Version 16.0 SPSS INC. Chicago, IL). Permission N otes This survey was reprinted with permission from the RAND Corporation. Copyright the RAND Corporation. RAND's permission to reproduce the survey is not an endorsement of the products, serv ices, or other uses in which th e survey appears or is applied

PAGE 59

59 Methodological Figures and T ables Figure 3 1 Variables of Importance and outcome measures for Athletic Trainers Burnout. (a)Demographics, (b)W ellness domains ( c) Outcome measures (Age, gender, e xperience, hours worked, & job setting) Professional Burnout Perceived Wellness Physical Activity Mental Health Social Support A B C

PAGE 60

60 Figure 3 2 Causal Model: Factors that determine Professional Burnout and Perceived Wellness Hours worked per week Years of experience Professional Burnout Perceived Wellness Physical Activity Mental Health Social Support

PAGE 61

61 CHAPTER 4 RESULTS Pilot D ata Twenty three ATCs outside the NATA District 9 sampling frame were asked to complete the instrument Fourteen certified athletic trainers complete d the survey instrument twice, approximately one week apart. All test retest correlations were greater than 0 .6 which is considered the minimum for an instrument to show adequate test retest validity.76 Using SPSS 16.0 ( Version 16.0 SPSS INC. Chicago, IL ), test retest correlations for each of the instrument scales were calculated as: Perceived Wellness Survey (r = 0 .763), Copenhagen Burn out Inventory (r = 0 .924), Mental Health Inventory5 (r = 0 .938), Baecke Physical Activity Questionnaire (r = 0 .924), and MOS Social Support Survey (r = 0 .924), also included in Table 4 1 The test retest correlations for each instrument a re statistically signific ant (p < 0 .0 001 ) for each scale and demonstrate that the current survey instrument has comparable test retest rel iability relative to previous investigations .22, 44, 59, 64, 69 See Appendix A for the full specifics on each instruments validity and reliability in past research Internal Consistency Internal consistency for the main data collection was calculated for each composite score using Cronbachs A lpha, see T able 4 2 The Perceived Wellness Survey instrument revealed a coefficient alph a or internal consistency of 0.937, the Copenhagen Burn out Inventory instrument alpha was 0.944, the Baecke Physical Activity Questionnaire alpha was 0.620, the Mental Health Inventory 5 alpha was 0.848 and the MOS Social Support Survey alpha was 0.950. Although there is no gold standard a score of 0.7 or gr eater is considered acceptable. All the instruments exhibited sufficient internal consistency reliability, except the Baecke Questionnaire and therefore this questionnaire might be conside red suspect in this instrument since the alpha did not

PAGE 62

62 meet the acceptable level of 0.07. Although this is suspect at 0.620, the instrument is still reliable and according to past research and current pilot data the questionnaire ha s acceptable alpha and test retest relaiablity .57 59 Data M anagement The primary mailing list of ATCs in D istrict 9 consisted of 1, 596 members who were willing to be contacted for survey research purposes Fourteen athletic trainers were removed from the list by Survey Monkey beca use they opt ed out of surveys sent through SurveyMonkey.com and 48 athletic trainer e -mail addresses were returned as undeliverable. From the 1, 434 remaining respondents, a total of 447 ( 31.3% response rate) participants responded to the survey instrument Eight percent ( n =35) of the participants failed to complete the instrument in the collection time frame of six weeks, leaving a total of 412 (28.7 % response rate) who completed the instrument items Subjects were excluded if more th a n 15 % of the tota l instrument or 15 % of any single instrument was left incomplete. M issing data below the 15 % level was replaced using the SPSS replace missing values function.77 Specific item mean scores were used to generate replacement values for missing data points. 76, 77 Several subjects reported working more than 120 hours per week, which suggests data entry errors on the part of the participants. Therefore, a Winsorize technique, as sug gested by the statistical consultant, was used to limit the extreme values for the hours worked per week variable. The W insorize technique is a technique used to limit the influence of the extreme values due to recording errors of the subjects. The W insori ze percentage was set at 3.88 % Demographics and Descriptive Analysis Frequency analyses were conducted to determine the number of participants job titles and job employment settings see T ables 4 3 and 4 4 Those participants that listed a job in the other category were placed in a separate category or placed into the category that best fit the

PAGE 63

63 job title description or job setting by the primary researcher Of the 404 participants completing the job setting item 33.9 % work in the high school setting (139), 22 % in the DI college setting (91), and 12 % in the clinic setting (51) S ee T able 4 4 for all job setting frequencies The job title data in T able 4 3 revealed that a majority of the participants were either a head (172, or 42%), assistant (88, or 21.5%), or clinical athletic trainer (66, or 16.1%). Although these are the job titles with highest percentages, a variety of other categories are present including academic areas, hospital and / or community colleges. Upon further examination of the job setting and job title variables and due to the lower number of respondents in several job setting and job title categories both variables were collapsed into fewer categories by the primary researcher to provide better statistical analysis. See the colla psed job settings and job titles in T ables 4 5 and 4 6 The central tendency and reliability measures were calculated using the SPSS 16.0 (i.e. means, standard deviations and Cronbachs Alpha for the scales and subscales). The means of the different inde pendent variables provided important information. In this sample, t he hours worked per week variable (50.64 16.93 hrs ) exceed s the traditional 40 hour work week and demonstrates large variability within each job setting in the field of athletic training. The results indicate that for the years of job experience the average years of experience was only 12.63, showing that not many of th e participants had been in the profession f or an extensive period of time, demonstrating a relatively young population. Each variable was calculated as a comprehensive or total score and as a sub -score or subscale where applicable. These results can be found in T able 4 7. The means for each instrument can be compared to the norms or set values that determine if the scores from the

PAGE 64

64 subjects are considered high or low. See A ppendix B for subscales and scoring. The means of those variables by gender can be found in T able 4 8 Perceived W ellness. The scores for t he Perceived Wellness Survey range from three to 29 and the highe r the score the greater the individuals perceived wellness.69 The overall mean of the study was 1 6.57 2.72, showing a lower level of perceived wellness among athletic trainers The subscales f or physical and psychological wellness had the lowest means (4.43 .78, 4.49 .69), wi th a maximum score being a six see T able 4 7 for complete means and st andard deviations of each subscale. See T able 4 9 for means and standard deviations for each job setting and job title. Burnout Burnout scores from the Copenhagen Burnout Inventory range from 0 to 100, with a score greater than 50 being considered a high burnout score .22 This range and scoring applies to each of the subscale s in the instrument. Participants in this project had a mean score of 41.72 17.1 for the total burnout score and had the following mean scores for the each of the three burnout subscales of personal, work, and client respectively: (44.88 17.79, 42.09 17.96, 38.18 16.89) S ee T able 4 9 for complete means and standard deviations for each setting and job title. Importantly, approximately 30 % (126) of the athletic trainers in this study were in the high burnout category (>50). Ment al Health Men tal health scores f rom the Mental Health Inventory 5 range from 0 to 100 with cut points ranging from 68 to 76. Points below this range can determine the presence of mental health problems like anxiety disorders and depression .70, 78 The sample of at hletic trainers in this study had a mean total score of 73.8 15.88. Interestingly, there w ere a wide range of scores ranging from 12 to 100. See T able 4 9 for means and standard deviations for each setting and title.

PAGE 65

65 Physical Activity. Physical activi ty scores from the Baecke Questionnaire of Physical Activity have a total score range from 1 to 15, with a score close to 15 indicating more physical activity .59 Athletic trainers in this study had a total mean physical activity score of 8.55 1.35. The subscales have a score range between 1 and 5, with 5 being very active for that subscale. The subscal e mean for work index was 3.12 0.51, for sport index 2.95 0.76, and for leisure time index 2.46 0.62. See T able 4 9 for means and standard deviations for each job setting and title. Social Support The final part of the instrument is the Medical Outc omes Study (MOS) Social Support Survey, which has four subscales associated with a total score range of 1 to 100.44 Higher scores on the total item indicate higher levels of social support. The f our subscales that totaled for the composite score include : 1) emotional, 2) tangible, 3) affectionate and 4) positive social and have score ranges from 1 5 with scores close to 5 showing more social support Athletic trainers in the study have a mean total social support score of 74.52 (23.02) The subscale mean scores are as follows: emotional 3.88 1.09, tangible 3.89 1.2, affectionate 4.09 1.16, and positive social 4.04 .97. See T able 4 9 for means and standard deviations for each job setting and title. Primary Analysis Correlations Between V ariables Pearson C orrelations coefficients were determined between the variables S ignificant relationships were present between several demographic and independent variables and the outcome or depende nt variables. Pearson Correlation s demonstrated significant negative relationships between age and burnout ( r = 0 .188 p <0.000 1), and years experience and burnout (r = 0.173 p < 0 .00 0 1). A positive relationship was found between hours worked per week and burnout ( r =0.124 p <0.01). The Pearson Correlation also revealed a negative relationship

PAGE 66

66 between bu rnout and the following variables: 1) social support ( r = 0 .265 p < 0 .0001), 2) mental health ( r = 0 .704 p < 0 .0001), 3) physical activity ( r = 0 .123 p < 0 .01) and 4) perceived wellness (r = 0 .515 p < 0 .0001). All the variables showed significant negative relatio nships leading the researcher to believe that decreases in the amounts of social support, mental health, physical activity and perceived wellness are associated with higher burnout scores. All correlation s are reported in T able 4 1 0 The perceived wellne ss variable was not correlated with the any of the demographic variables of age, hours worked per week or years experience. Perceived wellness was significantly correlated with social support, mental health and physical activity. A positive relationship ex ists between perceived wellness and the variables of 1) social support ( r = 0 .388 p < 0 .0001), 2) mental health ( r = 0 .486 p < 0 .0001) and 3) physical activity ( r = 0 .200 p < 0 .0001). Th ese results suggest athletic trainers with higher levels of social support, menta l health and physical activity may have greater perceived wellness scores. Regression A nalysis A m ultiple regression analysis w as conducted to determine which of the demographics variables i.e. job title, job setting, age, hours worked per week, years of experience, and outcome variables of mental health, social support, physical activity and perceived wellness (independent variables) would be predictors of burnout ( dependent variable) The f irst regression relating to burnout showed a significant overall multivariate effect (r = 0.760, R2=0.579, F17, 375 = 30.40, p < 0 .0001) and explain ed 57 % of the variance The independent variables included: age, gender, job title, job setting, hour worked per week, years of experience, mental health, social support, phy sical activity and perceived wellness. The significant predictors for this regression model with the dependent variable of burnout were perceived wellness, hours worked per week and mental health S ee T able 4 1 1 for Beta coefficients

PAGE 67

67 A second multiple regression analysis was conducted to determine which of the demographics variables i.e. job title, job setting, age, hours worked per week, years of experience, and outcome variables of mental health, social support, physical activity, and burnout (indepen dent variables) would be predictors of perceived wellness ( dependent variable). The second regression analysis examining perceived wellness showed a significant overall multivariate effect ( r = 0 61, R2= 0 376, F17 37 5=13 3 0, p < 0 .0001) and explained 3 7 % o f the variance. The independent variables included: age, gender, job title, job setting, hour s worked per week, years of experience, mental health, social support, physical activity and perceived wellness. The significant predictors for the dependent varia ble of perceived wellness were mental health, physical activity social support and burnout. S ee T able 4 1 2 for Beta coefficients Path Analysis for Burnout and Perceived Wellness A Structural Equation Modeling technique of path analysis was used to det ermine relationships between exogenous or independent variables and endogenous or depende nt variab l es and to answer the research questions 1 3 Research question 1 was to determine the effects of the exogenous variables of hours worked per week, years of e xperience and their effects on the e ndogenous variables which included: burnout, mental health, physical activity and social support. Research question 2 was to determine the effects of the exogenous variables of hours worked per week, years of experience and their effects on the endogenous variables which included: perceived wellness, mental health, physical activity and social support. A covariance matrix for the variables and their subscales or composite scores was used to determine each variable s casua l relationship in the proposed path model or diagram. Covariance results reported in Table 4 1 3 were used to determine direct and indirect effects for the outcome variables of burnout and perceived wellness.

PAGE 68

68 A two step approach recommended by Kline was u sed for this structural equation modeling.74 The first step was to test the fit of the measurement model based on appropriate goodness of fit indices. Over all fit of the measurement model was tested using Maximum L ikelihood E stimation to determine the goodness -of -fit of the model by simultaneously analyzing all the endogenous variables.74 T he goodness -of -fit index (GFI), c omparative f it i ndex (CFI), adjusted goodness -of -fit index (AGFI), standardized root mean square residual (SRMR), norm fit index (NFI), parsimonious NFI (PNFI), and the minimum Chi -square fit index were determined in step one or the measurement model .74, 75 Acceptable standards of fit are GFI 0.90, AGFI 0.90, CFI, 0.90 NNFI 0 .9 0 and SRMR 0 10.74, 75 The measurement model had the following Maximum Likelihood Estimation results: GFI= 0.916, AGFI=0.875, SRMR=.060, NNFI= 0.88, CFI= 0.91 and Chi Square 328.210 (DF 102, p < 0 .001) and was acceptable. The ori ginal proposed model shown in figure 3.2, did not provide adequate fit. The goodness -of -fit indices provided mixed results. The CFI value (CFI = 0.9134) indicated adequate fit, while the RMR value (RMR = 2.93), GFI (GFI=0.879), adjusted GFI (GFI= 0.833), standardized root mean square (SRMR= 0.1149) and NFI (NFI = 0.887) indicated poor fit to the data. Modifications to the model were guided by hypothetical considerations and suggestions from the statistical consultant. Upon examination o f the indicators o f physical activity, the work index subscale appeared to represent a different physical activity construct relative to the sport index and leisure index. Thus, the work index was removed as an indicator of physical activity and no longer included in the m odel. In the new model the latent variable of physical activity was conceptualized with the two subscales representing only physical activity outside of work (leisure and sport). The second modification was to specify burnout as an observed variable, in

PAGE 69

69 which the total burnout score, and not the subscale scores, were used in the model. This modification was made because the results of the original model estimated a negative eigenvalue indicating that the subscales of burnout were too highly correlated. The new path model provided adequate fit to the data. The CFI value (CFI = 0. 937) indicated adequate fit, while the RMR value (RMR = .694), GFI (GFI=0. 933), adjusted GFI (GFI= 0.8 98), standardized root mean square (SRMR= 0. 053) and NFI (NFI = 0. 906) ind icated adequate fit to the data. The final and full hypothesized model including significant and non-significant pathways and com pletely standardized estimates is presented in Figure 4.1 See F igure 4.2 for the final path model with only the significant pa ths. See Table 4 14 for Beta estimates and t values and Table 14.5 for the factor loadings. Additionally, each observed variable has an error variance reported in T able 4 1 6 Perceived w ellness The largest effect on Perceived Wellness was the level of me ntal health, which had a direct effect of 0 .41 therefore for every 1.0 increase in SD of perceived wellness mental health will increase 0.41 SD Physical Activity and social support had smaller (0 17, 0 .3 1 respectively), but significant total effects on perceived wellness and thus a 1.0 increase in SD for perceived wellness a 0.17 and 0.31 increase in SD for physical activity and social support, respectively Thus, higher levels of physical activity, social support, and mental health resulted in greater perceptions of wellness in ATCs. The exogenous variables, which were years of experience and hours worked per week, did not significantly affect perceived wellness. The variance accounted for in perceived wellness by the variables of hours worked per wee k, years of experience, physical activity, mental health and social support was substantial at 42 % The results show that research question 2 was not supported, since all three domains of health are causal of the perceived wellness.

PAGE 70

70 Burnout As hypothes ized in research question 1 b urnout was directly affected by physical activity, mental health, years of experience, and hours worked. Mental health had the largest effect on burnout, at 0 .6 4 and therefore a 1.0 increase in SD in burnout leads to a 0.6 4 decrease in SD of mental health According to the model, as ATCs mental health decreases the levels of burnout increase substantially. Physical activity had a direct effect on burnout of .013 and a 1.0 SD increase in burnout decreased 0.13 SD for p hysical activity Thus, in creased levels of physical activity lead to lower levels of burnout. The exogenous variable y ears of experience had a total effect of 0.18 with a direct effect o f 0 .09 and an indirect effect of 0.09 on burnout through mental health ( 0 .1 4 x 0 64 = 0.09). T he longer an individual stays in the athletic training field the lower the ir level of burnout. Likely, those AT Cs with high levels of burnout leave the field earlier in their career. As expected, the exogenous variable of hours worked per week had a direct effect on burnout of 0 .1 1 and therefore for every 1.0 SD increase in burnout there was a 0.11 SD increase in hours per week. Hours worked per week also had a small indirect effect (0.02) on burnout through physical ac tivity. Thus, according to the model, working more hours per week leads to higher levels of burnout. The path model with the variables of hours worked per week, years of experience, physical activity, mental health and social support accounted for 53 % of the variance in burnout. Mental h ealth, s ocial s upport, and p hysical a ctivity : The model specifies that m ental h ealth is directly affected by years of experience (0 14) therefore a 1.0 SD increase in mental health had a 0 .14 SD increase in years of exp erience. Thus, more years experience, can lead to higher levels of mental health. The path model accounted for the variables of hours worked per week, years of experience, physical activity, and social support for 2 % of the variance in m ental h ealth. P hysical Activity was significantly affected by hours worked per week with a direct

PAGE 71

71 effect of 0 .1 7 However, only a small amount of the variance in p hysical a ctivity with the variables of hours worked per week, years of experience, mental health and socia l support was accounted for at 3 % Social s upport was not significantly affected by the exogenous variables of hours worked per week, years of experience, physical activity, mental health and. Secondary Analysis Bivariate Analysis Gender An independent t test was used to determine the effect of gender on burnout, perceived wellness, social support, physical activity, mental health, hours worked per week, and years of experience. A significant effect of gender was found on burnout ( t (397) = 4.44, p <0.001) mental health, physical activity, hours worked per week, and years experience. For t tests in which Levenes test for equality of variances was significant, equal variances were not assumed. See Table 4 1 7 for results of the independent t tests. The a nalyses revealed that female ATCs compared to male ATCs reported significantly greater levels of burnout, lower mental health scores, an d greater physical activity. Additionally, male ATCs worked significantly more hours per week and had more years exper ience than female athletic trainers. See Table 4 8 for all means and standard deviations of the dependent variables by gender. Research Question 3 was addressed in the One -way ANOVAs that determined the effects of job setting and job title on the dependen t variables of burnout, perceived wellness, physical activity, social support and mental health. Job setting and job title were determined to have significant effects on all the dependent variables except social support. One way ANOVA Job S etting One -way ANOVAs were computed to determine the effects of job setting on the dependent variables of perceived wellness, total burnout, mental health, physical activity, social support, hours worked per week, and years of experience. The results indicated a

PAGE 72

72 signif icant influence of job setting on each dependent variable except for social s upport. See Table 4 1 8 for results. Bonferronis post hoc tests were conducted for all significant ANOVAs The results revealed that high school AT C s reported lower perceived w ellness scores than clinic AT C s ( p = 0 .007). Professional athletic trainer s reported lower levels of burnout compared to upper college level ( p = 0 .048) and high school ( p = 0 .015) athletic trainers Those athletic trainers who worked in a clinic reporte d less burnout than upper college level ( p = 0 .035) and high school AT C s ( p = 0 .005). In regards to mental health, professional athletic trainers reported higher mental health scores compared to those in non AT settings ( p = 0 .018) C linic ATCs reported higher mental health scores than high school ( p = 0 .017) and non-AT settings ( p = 0 .014). The post hoc tests also revealed that high school AT Cs reported significantly higher physical activity scores compared to upper college level ( p = 0 .002) and clinic A T C s ( p = 0 .048). Those who work in non -AT settings work significantly less hours per week compared to professional ( p = 0 .007), upper college level ( p = 0 .003), and the lower college level AT C s ( p = 0 .001). Additionally, high school and clinic ATCs work significantly less hours per week compared to professional ( p = 0.014; p = 0.001), upper college level ( p s < 0.001), and lower college level ATCs ( p s < 0.001). Finally, high school AT C s had fewer years experience than clinic (p = 0 .001) and academic s (p = 0 .042). See Table 4 9 for all means and standard deviations of the dependent variables by job setting. Job title One -way ANOVAs were computed to determine the effects of job title on the dependent variables of perceived wellness, total burnout, menta l health, physical activity, social support, hours worked per week, and years of experience. The results indicated a significant influence of job title on total burnout, mental health, physical activity, hours worked per week, and years of experience. See Table 4 19 for results of the One -way ANOVAs. All post hoc tests

PAGE 73

73 were conducted using Bonferronis post hoc tests Head athletic trainers reported higher levels of burnout compared to clinical athletic trainers ( p = 0 .034). Those that reported not wor king in athletic training reported lower mental health scores compared to head athletic trainers ( p = 0 .011), assistant athletic trainers ( p = 0 .021), clinical athletic trainers ( p = 0 .001), and those in academics (p = 0 .007). Head athletic trainers repo rted greater physical activity than academics (p = 0 .021). The post hoc tests also revealed that AT C s in academics worked less hours per week than head athletic trainers (p = 0 .044) and assistant athletic trainers ( p = 0 .004), and that clinical athletic t rainers worked less hours than assistant athletic trainers ( p = 0 .005). Additionally, head athletic trainers had significantly more years experience than assistant athletic trainers ( p < 0 .001) and less experience than academics ( p = 0 .002). Finally, ass istant athletic trainers had less years experience than clinical athletic trainers ( p < 0 .001) and academics job titles (program directors, professor etc) (p < 0 .001). See Table 4 9 for all means and standard deviations of the dependent variables by job t itle.

PAGE 74

74 Tables and Charts for Results Table 4 1 Pilot data t est r etest c orrelations Instrument C orrelations Perceived Wellness Survey 0.763 Copenhagen Burnout Inventory 0.924 Baecke Activity Questionnaire 0.924 Mental Health Inventory 0.938 MOS So cial Support 0.924

PAGE 75

75 Table 4 2 Internal Consistency / Cronbachs Alpha (reliability) Data reliability Cronbachs Alpha Perceived Wellness Survey 0.937 Copenhagen Burnout Inventory 0.944 Baecke Activity Questionnaire 0.62 0 Mental Health Inventory 0.84 8 MOS Social Support 0.95 0

PAGE 76

76 Table 4 3 Job t itle frequency Job Tit le Total Percent Non athletic trainer 7 1.7 Head athletic trainer 172 42. 5 Assistant athletic trainer 88 21.7 Assistant Athletic Director 4 1.0 Clinical Athletic Traine r 66 16. 3 Physician Extender 5 1.2 PT/ATC 17 4.2 Sales 6 1.5 Academics* 39 9.6 Total 404 100.0 Academics include Assistant, Associate professor, Dean, Program Director and secondary education teacher lecturer, and dual positions.

PAGE 77

77 Table 4 4 Job s etting frequency Job Setting Total Percent Non AT setting 12 2.9 Professional Athletic Trainer 23 5.6 D I athletics 91 22. 5 D II athletics 36 8.8 D II I athletics 11 2.7 Community College 10 2.4 High school 139 3 4 4 Clinic 51 12.6 Hospital 18 4.4 Academic 13 3.2 Total 404 100

PAGE 78

78 Table 4 5 Job t itle f requency c ollapsed Job title Frequency Percent Head athletic trainer 173 42.93 Assistant athletic trainer 89 22.08 Athletic director 4 0.99 Clinical (AT clinic, PT, PE, PA) 96 23.82 Academic (professor, PD, CC) 40 10.17 Total 402 100.00

PAGE 79

79 Table 4 6 Job s etting frequency coll apsed Job setting Frequency Percent Professional sports 23 5.7 DI, DII 128 3 1.7 DIII, CC 22 5.5 Assistant Athletic Director 4 1 .0 High School 141 34.1 Clinical 71 17.7 Academics position 14 3. 5 Total 403 100.

PAGE 80

80 Table 4 7 Means, s tandard d eviations, Chronbachs A lpha s cales and subscales Variable Subscale Mean Standard Deviation Cronbachs Age 36.52 9.56 Years of Experience 12.65 9.09 Hours worked/week 50.62 16.25 PW 16.57 2.8 0.937 Psych 4.49 0.69 Physical 4.43 0.78 Emotion 4.6 0.65 Spirit 4.77 0.7 Social 4.73 0.67 Intellect 4.5 0.6 BO 41.72 17.1 0.944 Personal 44.88 17.79 Work 42.09 17.96 Client 38.18 16.89 MH 73.85 15.87 0.848 PA 8.55 1.35 0. 62 Work Index 3.13 0.51 Sport Index 2.96 0.77 Leisure Index 2.47 0.63 SS 74.52 23.02 0.95 Emotion 3.89 1.1 Tangible 3.9 1.19 Affectionate 4.1 1.15 Positive social 4.05 0.97 Means and Standard Deviations for total scales and su bscales of Perceived Wellness (PW), Burnout (BO), Mental Health (MH), Physical Activity (P A), and Social Support (SS). Cronbachs Alpha for scales

PAGE 81

81 Table 4 8. Means and standard deviations for each gender Variable Gender N M ean SD t DF p Years Exp males 236 14.93 9.84 6.692 396.93 0.001 females 163 9.33 6.88 Hours males 236 53.23 15.90 3.646 397 0.001 females 163 47.03 17.73 PW males 236 16.71 2.74 1.150 397 0.251 females 163 16.39 2.80 BO males 236 38.60 16.82 4.441 397 0.001 females 163 46.25 17.06 MHI males 236 76.76 13.90 4.20 305.96 0.001 females 163 70.30 16.70 PA males 236 8.40 1.33 2.68 397 0.008 females 163 8.77 1.36 SS males 236 75.94 21.98 1.355 397 0.176 females 163 72.79 23.92 M eans and Standard Deviations for total scales of Perceived Wellness (PW), Burnout (BO), Mental Health (MH), Physical Activity (PA ), and Social Support (SS) spli t by gender. significant at .001

PAGE 82

82 Table 4 9 Means and s ta ndard d eviations of scales by job setting and job title Variable Setting N Mean SD Variable Title N Mean SD Wellness NON AT 13 16.02 3.30 Wellness NON AT 8 15.92 3.82 Pro AT 23 17.56 2.28 Head AT 172 16.31 2.61 Upper College 128 16.49 2.60 Assist AT 88 16.44 2.54 Lower College 21 15.92 3.12 Head AD 4 15.90 2.01 High school 139 16.10 2.56 Clinical 95 17.11 3.11 Clinic 70 17.59 3.31 Academics 40 16.95 3.01 Academia 13 16.60 2.42 Total 407 16.58 2.79 Tot al 407 16.50 2.79 Burnout NON AT 13 49.05 16.06 Burnout NON AT 8 59.34 19.93 Pro AT 23 31.39 15.51 Head AT 172 43.74 17.95 Upper College 128 43.21 16.68 Assist AT 88 44.04 15.91 Lower College 21 46.56 19.18 Head AD 4 47.66 34.56 High school 139 44.47 16.53 Clinical 95 36.97 16.73 Clinic 70 35.21 18.39 Academics 40 37.87 15.20 Academia 13 38.89 14.37 407 41.82 17.44 Total 407 41.82 17.44 Mental Health NON AT 13 63.38 23.88 Mental Health NON AT 8 54.00 27.21 Pro AT 23 81.57 9.32 Head AT 172 73.26 15.67 Upper College 128 72.70 15.73 Assist AT 88 72.60 15.14 Lower College 21 72.19 15.56 Head AD 4 67.00 34.00 High school 139 71.83 16.22 Clinical 95 77.26 14.27 Clinic 70 79.54 14.59 Academics 40 75.40 14.89 Academia 13 74.77 9.98 407 73.82 15.93 Total 407 73.82 15.93 Physical Activity NON AT 13 8.19 1.48 Physical Activity NON AT 8 8.03 0.92 Pro AT 23 9.03 1.33 Head AT 172 8.75 1.31 Upper College 128 8.29 1.22 Assist AT 88 8.53 1.21 Lower College 21 8.15 1.65 Head AD 4 8.06 1.41 High school 139 8.92 1.30 Clinical 95 8.51 1.35 Clinic 70 8.32 1.32 Academics 40 7.99 1.70 Academia 13 8.43 1.64 Total 407 8.55 1.35 Total 407 8.55 1.35

PAGE 83

83 Table 4 9. Continued Variable Setting N Mean SD Variable Title N Mean SD Social S upport NON AT 13 73.72 28.00 Social Support NON AT 8 63.77 27.73 Pro AT 23 76.61 17.62 Head AT 172 73.03 23.58 Upper College 128 72.71 24.12 Assist AT 88 74.09 23.94 Lower College 21 69.02 30.97 Head AD 4 81.77 16.61 High school 139 75.14 21.26 Clinical 95 76.93 20.13 Clinic 70 78.62 20.73 Academics 40 77.39 23.84 Academia 13 68.91 28.68 Total 407 74.50 33.95 Total 407 74.50 22.95 Hours/Week NON AT 13 39.10 21.02 Hours/W eek NON AT 8 44.54 18.96 Pro AT 23 58.91 26.50 Head AT 172 52.40 18.39 Upper College 128 56.86 15.82 Assist AT 88 55.31 15.47 Lower College 21 61.62 12.70 Head AD 4 50.50 8.42 High school 139 46.73 14.47 Clinical 95 46.43 14.30 Clinic 70 43.46 12.88 Academics 40 43.72 14.90 Academia 13 48.08 15.48 Total 407 50.61 16.85 Total 407 50.61 16.87 Years Exp NON AT 13 16.27 5.11 Years Exp NON AT 8 13.57 4.89 Pro AT 23 15.87 9.74 Head AT 172 13.20 9.03 Upper College 128 12.67 10.27 Assist AT 88 6.72 4.09 Lower College 21 10.24 6.54 Head AD 4 17.25 17.13 High school 139 10.30 7.52 Clinical 95 14.47 9.65 Clinic 70 15.55 9.55 Academics 40 18.96 8.56 Academia 13 18.31 8.06 Total 407 12.71 9.13 Total 407 12.71 9.13

PAGE 84

84 Table 4 10. Correlations between variables Variable 1 2 3 4 5 6 7 8 1. Age 1.000 0.024 *0.915 0.188 0.073 0.022 0.069 2. Hours/week 1.000 0.040 0.010 0.055 0.058 0.052 3. Years exp 1.000 0.173 0.083 0.0006 0.107 4. Burnout 1.000 0.515 0.265 0.123 0.704 5. Perceived wellness 1.000 *0.388 *0.201 *0.486 6. Social support 1.000 0.114 *0.352 7. Physical activity 1.000 8. Mental health 1.000 *Significant at p<0.0001 0 .001 0 .01 Significant at p< 0 .05

PAGE 85

85 Table 4 1 1 Regression model for b urnout Variables Beta F p value Gender (Male) 3.57 2.63 0.0001 Age 0.14 0.87 0.3849 Title (Head ATC) 3.05 1.19 0.2355 Title (Assistant ATC) 1.14 0.41 0.6790 Title (Athletic Director) 2.45 0.39 0.6945 Title (Clinical) 0.02 0.01 0 .9941 Setting ( Professional ) 5.24 1.16 0 .2451 Setting ( DI & DII ) 1.48 0.40 0 .6929 Setting ( DIII & CC ) 0.98 .220 0 .8268 Setting ( High School ) 0.409 0.11 0.9161 Setting ( Clinical ) 0.117 0.03 0.9771 Years of work experience 0.039 0.23 0.8169 Phys ical Activity 0.070 0.15 0.8815 Social Support 0.018 0.65 0.5179 Perceived wellness 1.47 33.56 0.0001 Hour worked per week 0.191 21.9 0 0.0001 Mental health 0.621 184.2 0 0.0001 Significant predictors

PAGE 86

86 Table 4 1 2 Regression model for p erceived w ellness Variables Beta F p value Gender (Male) 0.374 1.41 0.0001* Age 0.019 0.61 0 .5398 Title ( Head ATC ) 0.480 0.96 0 .3352 Title ( Assistant ATC ) 0.547 1.02 0 .3079 Title ( Athletic Director ) 0.683 0.56 0 .5725 Title ( Clinical ) 0.540 1.01 0 .3141 Setting ( Professional ) 0.476 0.55 0.5859 Setting ( DI & DII ) 0.367 0.50 0.6139 Setting ( DIII & CC ) 0.084 0.10 0.9227 Setting ( High School ) 0.089 0.12 0.9056 Setting ( Clinical ) 0.972 1.23 0.2212 Hours worked per week 0.020 2.49 0.01* Year s of experience 0.020 6.2 0 0.01 Mental Health 0.023 4.92 0.05 Physical Activity 0.262 8.34 0.005 Social Support 0.028 27.10 0.0001 Burnout 0.055 33.56 0.0001 Significant predictors

PAGE 87

87 Table 4 13. Covariance t able Scale 1= years; 2= hours; 3=psychological; 4= physical; 5= emotional; 6=spiritual; 7= social; 8= intelligent; 9= burnout; 10= mental health; 11= sport index; 12=leisure index; 13= emotional; 14=tangible; 15= affectionate; 16= positive social.

PAGE 88

88 Table 4 14. Path model Beta estimates a nd significance Independent Dependent Path Estimate t value Sig Hours worked per week Physical Activity 0.168 2.53 Hours worked per week Perceived Wellness 0 .049 1.11 Hours worked per week Professional Burnout 0 .109 3.09 Hou rs worked per week Mental Health 0 .059 1.21 Hours worked per week Social Support 0 .059 1.11 Years of experience Physical Activity .0001 0 .01 Years of experience Mental Health 0 .139 2.85 Years of experience Professional Burnout 0 .092 2.64 Years of experience Perceived Wellness 0 .020 0 .46 Years of experience Social Support 0 .008 0.15 Physical Activity (WI SI LI) Professional Burnout 0 .134 2.73 Physical Activity (WI SI LI) Perceived Wellness 0 .163 2.59 Mental Health Professional Burnout 0 .642 15.85 Mental Health Perceived Wellness 0 .408 7.77 Social Support Professional Burnout 0 .022 0 .54 Social Support Perceived Wellness 0 .312 5.64

PAGE 89

89 Table 4 15. Factor loading table Table Standardiz ed factor loading for subscales Variable Subscale Factor loading t value Physical activity Sport Index (SI) 0.56 7.1 Leisure Index (LI) 0.67 6.8 Social support Emotional 0.71 15.4 Tangible 0.73 16.1 Affectionate 0.78 17.7 Positive social 0.81 18.6 Perceived wellness Psychological 0.81 17.6 Physical 0.54 10.7 Emotional 0.74 15.7 Spiritual 0.80 17.4 Social 0.63 12.9 Intellectual 0.70 14.9

PAGE 90

90 Table 4 1 6 Error variances for path model Error Variances Variable Estimate Standard e rror t value Sport index 0.375 0.080 4.660 Leisure index 0.307 0.040 7.610 Emotional 0.606 0.051 11.850 Tangible 0.686 0.059 11.480 Affectionate 0.526 0.050 10.330 Positive social 0.333 0.035 9.460 Psychological 0.191 0.017 10.690 Physical 0.494 0 .036 13.600 Emotional 0.218 0.018 12.090 Spiritual 0.191 0.017 10.830 Social 0.341 0.025 13.130 Intellectual 0.219 0.017 12.470 Mental Health 242.580 16.860 14.390 Burnout 141.127 10.152 13.900

PAGE 91

91 Table 4 1 7 Independent t test g ender Independent Samples t test for dependent variables by gender Dependent Variable t df p Hours worked per week 3.646 397 0.000 Years of experience 6.692 396.938 0.000 Perceived Wellness 1.150 397 0.251 Burnout 4.441 397 0.000 Mental Health 4.062 305.962 0.000 Physical Activity 2.681 397 0.008 Social Support 1.355 397 0.176 Significant variables

PAGE 92

92 One -way ANOVAs for dependent variables by Job Setting Table 4 1 8 Job s etting DV df Sum of squares Mean Squares F p PW Between Groups 6 136.060 22.677 2.987 0.007 Within Groups 400 3036.640 7.592 BO Between Groups 6 8037.713 1339.619 4.64 0.000 Within Groups 400 115486.778 288.717 MH Betwe en Groups 6 5868.809 978.135 4.022 0.001 Within Groups 400 97275.615 243.189 PA Between Groups 6 42.158 7.026 4.001 0.001* Within Groups 400 702.466 1.756 SS Between Groups 6 2803.867 467.311 0.885 0.506 Within Groups 400 211162.008 527.905 HRS Between Groups 6 16605.000 2767.500 11.179 0.000 Within Groups 400 99025.000 247.563 Years Between Groups 6 2301.555 383.593 4.861 0.000 Within Groups 400 31562. 550 78.907 Significant

PAGE 93

93 One -way ANOVAs for dependent variables by Job Title Table 4 1 9 Job t itle ANOVA DV Source of Variation df Sum of squares Mean Squares F p PW Between Groups 5 51.534 10.307 1.324 0.253 Within Group s 401 3121.165 7.783 BO Between Groups 5 4636.575 927.315 3.128 0.009 Within Groups 401 118887.916 296.479 MH Between Groups 5 4739.544 947.909 3.863 0.002* Within Groups 401 98404.879 245. 399 PA Between Groups 5 22639.000 4.528 2515.000 0.029 Within Groups 401 721.985 1.800 SS Between Groups 5 2412.962 482.592 0.915 0.471 Within Groups 401 211552.913 527.563 HRS Between Groups 5 6348.272 1269.654 4.659 0.000 Within Groups 401 109281.976 272.524 Years Between Groups 5 5144.830 1028.966 14.367 0.000* Within Groups 401 28719.719 71.620 Significant

PAGE 94

94 Figur e 4 1 Full model with significant and non significant paths Figure abbreviations: LI: Leisure Index; SI: Sport Index; PS: Psychological; PH: Physical; EM: Emotional; SP: Spiritual; IN: Intellectual; TA: Tangible; AF: Affectionate PO: Positive Social Se e error variance table for errors into subscales T able 4 1 5

PAGE 95

95 Figure 4 2 Path model with significant paths only

PAGE 96

96 CHAPTER 5 DISCUSSION Many professionals who work in close proximity with clients or patients experience some level of job burnout an d thus have decreased health and wellness l evels. Current research relating to the profession of athletic training and burnout is minimal and does not include any mention of the health domains and relationship to burnout. The primary objective of this stu dy was to determine the contributions of the wellness domains (physical, mental and social) on an athletic trainer s burnout level. The secondary aim was to determine the contributions of the wellness domains on the athletic trainer s perceived wellness le vels. The tertiary aim was to determine the direct and indirect effects of hours worked per week, and years of experience on burnout and perceived wellness. J ob setting, job title and gender, along with the wellness domains, hours worked per week, and yea rs of experience significant ly determin ed and predict ed burnout and perceived wellness. Burnout and Perceived Wellness Path Model The original path analysis for the domains of health, perceived wellness and burnout revealed that the proposed model did no t fit the data. Several possible reasons exist for this lack of fit. With multiple variables and multiple subscales for the variables a more complex analysis is needed to determine effects in the original model Another possible reason for lack of fit is that the proposed paths are in need of modification to show a better goodness -of -fit. Upon adap tation the new model showed several statistically significant direct and indirect effects on burnout and perceived wellness. The new model indicated that three health domains had significant effects on the perceived wellness of athletic trainers. This effect is expected due to the fact that the three domains (physical, mental and social) are part of the perceived wellness instrument. However, contrary to the hyp othesized disconnect or lack of relationship between

PAGE 97

97 perceived and actual wellness domains of ATCs, no disconnect exists. This does not indicate that ATCs are healthy only that they perceiv ed wellness correctly according to the various instruments used in the study P hysical activity, mental health, hours worked per week and years of experience produced direct effects on burnout. Years of experience and hours worked per week also had effects that were indirect on burnout through mental health and physical activity, respectively. This finding supports the hypotheses that the health domains of physical activity and mental health along with years of experience and hours worked per week are factors that can cause burnout. Burnout A thletic Trainers As pre dicted the average total scores exposed that athletic trainers experience moderate burnout in either their job, their personal life or with clients as part of their job Athletic training burnout scores on the Copenhagen Burnout Inventory (CBI) ranged fr om 0 to 97, showing some athletic trainers scores encompass the extreme ends of the instrument. Direct comparisons to previous research must be made cautiously because of a multitude of burnout instruments being used by other researchers. T he current mea n of 41.81, indicates moderate burnout and is comparable with past research.3 Additionally, 30 % of the population of athletic trainers in the study were highly burned out (>50) which is an actually higher rate than previous reports .10 A study on the non athletic training population used both the CBI and MBI, and revealed that 1 5.5 % of the sample had high burnout according to the MBI while 16 % of the sample had high burnout according to the CBI.38 The non s ignificant difference between scoring instruments allows for some comparison be tween the two different scores, giving validity to the CBI in regards to burnout and leading to a potentially new instrument to be referenced when determining athletic training burnout.

PAGE 98

98 Past burnout studies in athletic trainers employed the three subscales, emotional exhaustion, depersonalization and personal accomplishment, of the Maslach Burnout Inventory (MBI) i.e. emotional exhaustion, depersonalization, and personal accompli shment, as a way to measure burnout High emotional exhaustion scores corresponded to higher levels of burnout. High depersonalization scores also corresponded to higher burnout scores. However, when reviewing p ersonal accomplishment scores, a lower scor e for this subscale is considered to be burn ed out .79 This change of scoring make s it difficult t o determin e an overall burnout score, as compared to the CBI The CBI can be used as separate scales or can be combined into a single composite score making the current studys use of an instrument for burnout a successful step in giving a total burnout score .22, 38 The findings of this study indicate that athletic trainers have moderate to high levels of burnout when compare d to past findings in the area of burnout research. Hendrix et al. found that e motional e xhaustion score s for football (20.24 9.0 ) and non football (20.06 SD 8.71) athletic trainers are in the middle of the MBI scoring for emotional exhaustion correspond ing with the high to middle level of burnout in this study Additionally, the depersonalization s cor es reported by Hendrix et al. for f ootball (1 0.93 6.24) and non football (8.44 5.2) athletic trainers and the moderate burnout levels found in this study were higher tha n other professions like mental health workers (7.12 5.22) and medical workers (7.4 6 4.62) .3 Recent work by Kania et al. contradicts the research findings from Hendrix et al. by reporting that only 66 athletic trainers out of 206 (32 % ) could be classified as having any type of burnout ( high or low ). These results may stem from the use of only college level athletic trainer s, who according to the author, work with the same athletic teams on a daily basis which affects the overall burnout levels of college athletic trainers.10 The ability to work with the same athletes and see

PAGE 99

99 favorable outcomes in rehabilitation and progress towards return to play is rewarding and lessens the burnout score. With minimal past research reporting exact burnout level scores in direct comparison of burnout scores is difficult until more research and similar instruments are used in the literature. Clapper et al. used a modified MBI, called the ATBI or Athletic Training Burnout Inventory among ATCs and found similar levels of burnout as did previous investigations as well as the current study .9 However, because the ATBI is not exact ly comparable to the MBI the author cautions against making direct comparisons to previous research.9 This different instrument makes comparison to the current study limit ing Mazerolle et al. examined the relationship between burnout and the work family conflict, and found that a higher work family conflict lead to increased burnout scores. In the current study, c orrelations between work family conflict and burnout ( r =0.63) are similar to the negative correlation found between wellness and burnout (r = 0.515) Specifically, the negative correlation found showed that athletic trainers with higher total burnout demonstrate lower overall perceived wellness levels. These finding s are extremely imp ortant because previous literature was anecdotal in nature, discussing signs, sympt oms, and case histories as well as ways to manage stress and burnout.4, 5 Future wellness and burnout research should incorporate this work family conflict relationship with regard to burnout. Many c ommon factors of burnout mentioned in these anecdotal articles included job characteristics like: hours worked per week, job stressor s and a variety of work related characteristics .2, 4, 5, 80, 81 Many other factors that have been linked to burnout in health care workers include: negative feedback, scarcity of resources, and job characteristics (i.e. hours worked per week, number of clients).4, 21, 25, 30, 80 83 F actors like hours worked per week and job

PAGE 100

100 or role conflict and a decreased resource base have been linked to burnout in athletic trainers .4, 21, 25, 30, 8083 In this s tudy the amount of h ours worked per week had a significant positive relationship with burnout (r =0.124) 0.191). Thus the more hours an athletic trainer work s per week the higher the burnout scores are as was hypothesized in research question 1 This finding is consistent with other research on burnout and hours worked. M azerolle et al. found that DI athletic trainers who had higher work family conflict also had higher burnout scores leading the researchers t o believe that DI athletic trainers struggle to find a balance b etween work and family, possibly due to the number of hours worked.8 These findings only apply to DI athletic trainers and cannot be generalized to all athletic training settings. The number of hours worked per week reported by the participants in this study (50.6 1 6.25) is actually l ower th a n reported in past literature Lockard reported a range of 60 to 70 hours worked per week for high school teacher athletic trainers and 50 to 60 hours worked per week for college athletic trainers.82 The range of hours w orked per week in this study places our results in the same range as Lockard and Mensch suggest and what Mazerolle et al reported in their study.8, 81, 82 The original athletic training burnout articles by Geick suggested long hours are consistent with the burnout syndrome.4, 5 It seems that there is evidence for hour s worked per week as a cause of job burnout in this study, as well as in previous research. 8, 81, 82 Additionally, the findings demonstrated that hours worked per week had an indirect effect on burnout mediated by physical activity indicating that more work time leaves less time for physical activity and thus more likelihood of professional burnout. An interesting finding of the possible causes of burnout was the number of years of experience that the athletic trainer invested in the profession. In this study, t he correlation

PAGE 101

101 between years of experience and the level of burnout (r = 0 .173) showed a significant negative relationship. Those athletic trainers that had achieved more years of exper ience had lowe r burnout scores, suggesting that those who are burned out leave the profession altogether Mazer olle believed this as well, reporting that those with higher work family conflict had higher intention s to leave the profession. Another possible reason for decreased burnout is the ability to cope with job stressors grows with experience. 8 Additionally, the path model indicated that years of experience had an indirect effect on b urnout and was mediated by mental health. Athletic trainers who st ay in the profession longer have better mental health which leads to a decrease in professional burnout. Job setting and job title also played a role in which group tends to experience bur nout. Acco rding to the study by Hendrix et al. those athletic trainers who worked in a non football setting had lower emotional exhaustion and depersonalization scores, while reporting higher personal accomplishment scores.3 Among this sample a significant difference exists for burnout among t he different settings ( p < 0 .01) and the different job titles ( p < 0 .0001), suggesting that the job title as well as the job setting may be a potential determinant of burnout. A thletic trainers who m had the title of assistant athletic trainer reported the highest burnout means for job title while the lowest mean was reported by those athletic trainers that held athletic director jobs. For j ob setting the lower college (DIII and CC) level re ported the highest means for burnout In this sample, the l owest mean scores for burnout were found in the clinical setting. Kania et al found that high stress levels, coach es pressure to return athletes to play, and the number of athletes or teams an athletic trainer works with can predict burnout in the college athlet ic trainer population.10

PAGE 102

102 Burnout ha s many predictors A n even larger number of people will experience burnout if these predictors are not determined and managed successfully in todays athletic training profession. A multitude of other factors outside those mentioned in this study exist. Past research has suggested that factors outside ones job characteristics play a role in the burnout experienced by athletic trainers, such as hardiness, role strain, social support and work family conflict. 3, 7, 8 In this study, the relationships between t he three wellness domains (physical, mental and social) and burnout were investigated However determining all the variables of burnout is implausible. Future research should expand to include more of the many factors associated with burnout. Personal ch aracteristics such as stress le vel, leisure time and hardiness are important to understand since they could lead to decreases in mental health and less professional burnout.3, 10 Physicians Athletic trainers are not the only allied health care professional to experience the burnout syndrome. Health care professionals like physicians and physical therapists also experience this syndrome. Physicians that worked more hours or shifts per week and took less vacations were those that experienced more burnout. This studys results on burnout in athletic trainers are similar to the results found in research done on physic ians and residents, particularly in relation to the h our s worked per week. Both athletic trainers and residents or physicians work a high number of hours per week and this can lead to this burnout syndrome. Ozyurt et al. reported that physicians who work a higher number of shifts reported higher emotional exhaustion and depersonalization scores This finding supports these results that work hours positively correlat ed to burnout and that hours worked per week had direct and indirect effects on burnout in the path model .36 Ozyurt also reported that the number of vacations per year was a significant variable in all three subscales of the MBI for their popula tion of doctors, revealing also that

PAGE 103

103 taking time off can reduce the burnout syndrome in health care professionals.36 Another study by Spickard et al. repor ted that when working over 40 hours per week, the odds of experiencing burnout increased 12 15 % for each additional five hours worked per week.29 Thomas reported that one third of the residents felt overburdened by the overall work load and the intensity of the work load itself rating it high.84 Job settings in this study showed that some settings (lower college an d high school) lead to higher levels of burnout and decreased levels of wellness. Physical Therapist Physical therapists are another allied healthcare professional that experiences burnout. This group of professionals is particularly similar to athleti c trainers in job tasks and education, and patient or client interactions. Physical therapists work with similar client s particular those with orthopedic injuries. Donahoe et al. reported emotional exhaustion (EE) scores of 23, depersonalization scores of 7, and personal accomplishment scores of 37 for this group. Their population had 45 % fall in the high burnout score category (EE) which wa s similar to the burnout scores for athletic trainers in this study.33 In a separate study by Balogun et al. the emotional exhaustion and depersonali zation scores were much higher th a n past research contributing to a higher percent (58 and 40 %) of physical therapist s falling in to the high burnout category This finding is similar to the amount of athletic trainers in this study .14 Mean burnout scores for athletic trainers in this study fell in the moderately high burnout category, and the wide range of scores show that many athletic trainers are experiencing high levels of burnout. Physician Assistant Physician assistants (PA) work in a variety of settings, closely with doctors and pa tients and sometimes athletic trainers. PAs and athletic trainers have similar job roles and responsibilities within the work place. There is m inimal research available on PA burnout T he only article existing on PA burnout is on the emergency room ( ER ) p opulation of PAs. Using

PAGE 104

104 the Maslach Burnout Inventory, the authors found that 59 % of PAs received high or moderately high burnout scores, and 66 % received high or moderately high s cores for depersonalization.15 This finding suggests a greater percentage of ER PAs appear to experience burnout compared to those ATCs in the current study. However a comparison should be made with cauti on because different burnout instruments were used. Physical A ctivity and B urnout in Athletic Trainers L evels of physical activity reported in this study showed that overall athl etic trainers from the southeast United States reported moderate levels of physical activity on the Baecke Physical Activity Questionnaire (8.55 1.35) The closer this combined score is to 15 the more active a person is in the areas of work, sport or leisure activity. Cuppett et al. used the Baecke instrument and found that ath letic trainers in the Midwest United States were slightly more active with a score of 8.8 9.0 (SD 1.2).6 In regards to the individual subscales scores for the Midwest population, scores were higher in the sport subscale and the lowest for the leisure subscale.6 The present study found that the work subscale score was the highest score, and was slightly above three Additionally, in this study, t he leisure subscale score was the lowest. According to these results, ath letic trainers are somewhat active at work and so mewhat active in sport; but have decreased amounts of leisure activity. Physical activity score s from this study show that the athletic trainer falls in the middle to upper range on the Baecke instrument an d that athletic trainers are only somewhat active during a typical day. Another study by Groth et al found that only 41 % of their population of athletic trainers met the ACSM guidelines of 30 minutes or more per day of exercise five days a week.63 Groth et al. also discovered that 27 % of the females and 74 % of the males in the study were either overweight or obese.63 With Groth et al.s findings and those of this study there is an

PAGE 105

105 overwhelming need for future research that focus es on the barriers to physical activi ty among athletic trainers S everal researchers report that physical activity and leisure activity would be helpful in reducing the burnout syndrome.4, 5, 80, 81 Research on physical activity and the relationship physical activity has in regards to burnout among athletic trainers is either minimal or non existent. In this study three levels of p hysical activity (work, leisure and sport) were measured and compared with burnout scores. In the regression analysis, physical activity was not a significant predictor of burnout However when the work index was removed from the path model analysis, phys ical activity had a direct effect on burnout. Greater non work physical activity (leisure and sport) cause d ATCs to experience lower levels of burnout. Apparently physical activity at work should not be grouped with physical activity outside of the work place when examining relationships between physical activity and burnout in athletic trainers. Further research in this area is needed. Mental Health and B urnout in Athletic Trainers Athletic trainers are exposed to many factors in their jobs that cause b urnout and can decrease mental health levels. The scores on the Mental Health Inventory 5 reported in this study indicate that athletic trainer s levels of mental health ranged from 12 to 100 with high scores being considered a healthy score.64, 70, 78, The mean score found in this sample of athletic trainers was 73.85; however, the scores had a wide range and a 15.88 standard deviation. With a wide range and large standard deviation, many athletic trainers fell into a level of mental health around or below the mental health cut off, which can be indicators of depression or anxiety. According to Kelly et al. a significant cutoff score for mental health is between 68 and 76.78 In a study Hoeymans conducted on the general population, the mean mental health score was 80 and only 20 % of the population fell below the 72 cut -off point score.70 No MHI 5 scores o f athle tic

PAGE 106

106 trainers exist for comparison. However, athletic trainers mental health levels from this study are below the mental health scores of the general population assessed in the Hoeymans et al study A gender separation of the MHI 5 in this study exhibit ed ma les (76.6) reported higher scores tha n females (70.3), which is the same as was found for males (83) and females (78) in the general population.70 No study has looked directly at the relationships between mental health and burnout in athletic trainers. In this study m ental health was the most important cause of burnout in athletic trainers. The path model indicated that the better an ATCs mental health score the lower the level or score of burnout. The findings from this study support the notion that depleted emotiona l health leads to burnout.25, 26, 30 Social Support an d B urnout in Athletic Trainers Research on the level of social support in athletic trainers is minimal and to date only one study has reviewed social support for athletic trainers. T h e current study examined whether social support w as a predictor of burn out. The MOS Social Support Survey was able to determine social support levels in athletic trainers and possible relationship to burnout. In this s tudy, athletic trainers had mean social support scores of 74.52, which is a particular ly good score; however the standard deviation (23.03) and range was quite variable. The range of this instrument score was 95, with four being the lowest score and 100 being the highest score. With such a wide range, many athletic trainers may be experiencing some decreases in s ocial support while others perceive their social support as good Hendrix et al. used the Social Support Questionnaire to access the number of perceived social support network members and more importantly, the degree of satisfaction with the support rece ived. However, according to Hendrix et al. experiencing decreased scores in social support did lead to some perceived stress and thus higher scores in the burnout subscale of emotional exhaustion.3 The MOS scores found by Sherbourne and Stewart (70.1 24.2) were similar to the scores found in this study, however,

PAGE 107

107 comparison with Sherbourne and Stewarts results should be treated with some caution because the population in that study w as not athletic trainers, but patients with several different chronic illnesses.44 Surprisingly, social support did not predict or cause burnout in athletic trainers according to the path model. The MOS only assessed general levels of social support in ATCs, conceivably social support at work is related to burnout, but general social support is not. Future research should distinguish between lev els of social support at work versus social support at home or in an other environment. Perceived W ellness and Burnout Burnout i s related to the amount of illness and severity of disease that can place a person at higher risk for major illnesses However no research has observed the perceived wellness levels of athletic trainers and the potential effects of the burnout syndrome on the perceived welln ess of an athletic trainer. Scores on the Perceived Wellness Survey range from 1 to 29. Scores closer to 29 indicate a positive perception of wellness. The athletic trainers in this study had a mean (16.57 2.8) that was in the middle range of this score, thus demonstrating low to moderate perceived wellness levels. The correlation between burnout and percei ved wellness was strong, showing a negative relationship between perceived wellness and burnout A s expected higher burnout in athletic trainers was associated with lower perceptions of wellness. With a strong correlation and with burnout as a predictor o f perceived wellness, a possible clinical significant relationship exists, in which perceived wellness and burnout are plausible causes of one another. Multiple regressions analysis revealed that all three domains were significant predictors of perceived w ellness, as was burnout.

PAGE 108

108 Perceived W ellness and the Health D omains in Athletic Trainers Perceived wellness mean scores for athletic trainers were 16.57 2.8, on a 29 point wellness scale Adams et al. created the Perceived Wellness Survey using data f rom general populations In comparison to the means in this study the data is comparable.60 Intuitively, allied health care professional s should score fairly high on this instrument and even higher th a n general populations H owever the mid range score for athletic trainers is quite similar to the original population foun d by Adams et al .60 A c omparison should be made cautiously since popul ations are different however a concern exists because PWS scores for athletic trainers are in the mid range and not higher, which is considered healthier The current s tudy also looked at the gender differences of perceived wellness and found that females perceived their health as slightly poorer than males but difference s were not statistically significant When comparing different job setting s, the findings show that those athletic trainers working in lower college settings (DIII and CC) exhibited the lo west perceived wellness followed by high school athletic trainers. Clinical settings received the highest perceived wellness scores. A thletic trainers holding the job title as head or assistant generated the lowest perceived wellness scores and those with the title of clinical ath l etic trainer had the highest perceived wellness scores. These findings are possible due to the relative stable work hours and the decreased amount of working hours by clinical athletic trainers. According to the final path model and as hypothesized, physical activity, mental health and social support all had direct effects on perceived wellness. Mental health had the strongest effect on perceived wellness, followed by social support and physical activity scores respectively Each domain of wellness was positively related to perceived wellness. Therefore the greater physical activity, social support and mental health an athletic trainer has the greater t he perceptions of wellness should be in athletic trainers.

PAGE 109

109 Conclusion This study describes the relationships between the burnout syndrome in athletic trainers and perceived wellness and different health domains of physical activity, mental health and social support of wellness. It should also be noted that demographic variable s such as, years of work experience and hours worked per week should be considered in determining individuals levels of burnout and when they syndrome is likely to be experienced In conclusion, approximately 30 % of the athletic trainers from District 9 in this study reported high burnout scores, while many more were moderately burnt out Most i mportantly, the current study produced a path model of burnout and perceived wellness in athletic trainers. The model indicates that years of experience, hours worked per week, mental health, and physical activity are important factors causing burnout in athletic trainers. Additionally, the health domains play vital roles in determining perceived wellness and should be considered key in maintaining health and wel lness. Limitations Study limitations preclude the generalizability of these finding to other populations. Methodological limitations include: the use of email surveys as a means to collect data and in particular the use of a self report or respondent rec al l for the data. Furthermore, generalizability is limited due to the s ampling design which involved only one of the nine districts in the NATA. Along with the methodological limitations, several outcome limitations have occurred in this study. The respo nse rate of 2930 % is small and precludes generalizing to a population. Second ly one of the tools used, the Baecke Physical Activity Questionnaire was found to have a lower coefficient alpha tha n the other instruments and was slightly below the traditionally accepted level of 0.7. The findings of total physical activity in this study are considered normal or average; however the score should be interpreted cautiously since many athletic trainers

PAGE 110

110 partake in a fairly large amount of work activity This la rge amount of physical activity at work could in turn, lead to higher burnout or exhaustion and lesser amounts of leisure activity, also leading to higher levels of burnout. Since all instruments had previous validity and reliability metrics in the litera ture, a survey of experts was not used but would be a future step to ensure validity for this combination of instruments. Finally, the study is looking at a population whose resp ondents are possibl y too burn ed out or not willing to take the time to answer items on a survey Therefore, the very ATCs experiencing the highest levels of burnout and of most interest for this study are not likely to participate. These limitations should be continued to be addressed in future research by possibly contact ing respo ndents in more direct and qualitative processes. Implications The knowledge of the potential factors contributing to professional burnout, either in the workplace or in the personal life, can reduce overall health decrements due to the excessive stressor s of jobs in healthcare All previous athletic training research on burnout uses the MBI or some modification of the MBI. This first use of the CBI in athletic training allows for a new and more direct perspective on measuring burnout in all three areas of burnout (personal, work and client). The development of plans or interventions to reduce burnout may decrease the previously mentioned work family conflict, job dissatisfaction and intention to leave the profession. The same knowledge of potential burno ut factors can decrease the number of illness es and disease present among athletic trainers. Focusing on the enhancement of the domains of health, physical activity and mental health may reduce the severity of burnout and thus the severity of illnesses. Many health care professionals can be armed with the knowledge and tools to reduce their personal or job burnout, and also educate both current and future professionals about the causes of burnout

PAGE 111

111 Practical A pplications With so many known potential f actors for burnout and decreases in health and wellness, the ability to reduce certain job characteristics or responsibilities leading to burnout and poor wellness can be acted on by administrators and those who control hiring and job descriptions of potential employees. Educating administrators about the causes of job burnout can reduce their costs of hiring and replacing of employees like athletic trainers who leave the profession after five to ten years of service due to burnout Another potential applic ation is the adding of a burnout and health related competency to the National Athletic Training Competencies of Education book. Educators can start to manage and prevent job burnout, dissatisfaction and intention to leave the profession. Future R esearch Future research in the areas of athletic trainer s health and wellness should continue to investigate the factors that contribute to burnout. Many factors like personality characteristics, environmental characteristics, coping strategies, job engagement and other qualitative measurements in specific job settings need to be incorporated in path models as well as relationship studies. Another area that requires attention is the separation of the different health domains and their scores in relation to diffe rent job settings. Continued work on the survey instrument will allow for items, subscales or instruments to be reduced, thereby reducing the time required to complete the survey. This reduction may cause more people to participate in future studies. De termining the amounts of social support mental health a nd physical activity present in athletic trainers could increase the likelihood of avoiding physical illnesses and decreases in professional burnout. Future research needs to specifically investigate the relationship between burnout and physical activity at work.

PAGE 112

112 Future research should address different populations of athletic trainers like : graduate stude nts, undergraduate students or focusing on particular job titles, like head athletic trainer or clinical athletic trainer. Knowledge of those susceptible to burnout can lead to educational and administrative changes in both undergraduate curriculums and graduate student assistantships at all levels of athletic training. The current study indicates that those working at lower level colleges (DIII and CC) may be particularly important to focus on in regards to burnout. Direct contact of athletic trainers suspected of experiencing athletic trainer burnout can have an even bigger impact on prevention and more importantly, handling of the burnout syndrome. Focus groups can be used to obtain more information from those athletic trainers that might be experiencing burnout or decreased perceived wellness but are too busy or unaware of their symptoms to pa rticipate in traditional quantitative research methods With the use of qualitative methods like direct contact of potentially burnt out athletic trainers there is a greater chance that more effective prevention and treatment of decreased wellness, prof essional burnout and physical illnesses can be developed.

PAGE 113

113 APPENDIX APPENDIX A PSYCHOMETRIC DATA FO R INSTRUMENTS Table A 1 Table of p sychometric data Instrument Author Validity Reliability N MOS SSS Medical Outcomes Social Support Survey Sherb ourne & Stewart (1991) Correlations (health status measures) .72 .87 Internal Consistency .97 overall 2987 Wasserman, Stewart, & Delucchi (2001) Correlations (4 subscales) .72 .82 Internal Consistency .97 overall 128 12 and 4 item MOS Gjesfjeld, Green o & Kim (2007) Correlations (SF36) .98 &.96 Internal Consistency .94 &.83 330 Baecke Questionnaire on Physical Activity Pols et al. (1995) Correlations (Diary M and F) .66 and .42 Correlations (energy expenditure M and F) .56 and .44 Test retest correl ations .65 .89 range 126 Florindo et al. Pearson Correlation (Locomotion activities, leisure exercise, & percent heart rate decrease) .52, .47 &.47 Interclass correlations .69, .80, & .77 21 Baecke, Burema & Frijters (1982) Test retest for each ind ex .88, .81, &.74 309 Mental Health Inventory Veit & Ware (1983) Correlations (5 subscales) .34 .75 Internal consistency .83 .91 Stability .56 .64 5089 McCabe, Thomas et al (abstract) (1996) Correlations (GHQ) .73 Internal Consistency .84 3000 Copenhag en Burnout Inventory Kristensen Borritz et al. (2005) Correlations (SF 36 ) .34 to .75 Internal Consistency .85 .87 1914 Winwood & Winefield (2004) Correlations (MBI subscales) .38 to .45, .38 to 52 and .75 to .82 Alpha Reliability .73 .93 312

PAGE 114

114 Table A 1 continued Instrument Author Validity Reliability N Perceived Wellness Survey Adams, Bezner Steinhardt (1997) Correlations (other valid scales) .3 .7 Internal Consistency .88 .93 295,98, 53,&11 2 Bezner, Adams and Whistler (1999) Internal Consistency .92 237

PAGE 115

115 Table A 2 Scoring of instruments table Instrument Score index/ composite Possible subscales Possible max score MOS SSS Medical Outcomes Social Support Survey Total score = 100 X{ (observed score min possible score) /(ma x possible score min possible score)} Emotional support Tangible support Affectionate support Positive social interaction 100 Baecke Questionnaire on Physical Activity Total score = WI+SI+LTI WI= work index, SI= sport index, & LTI= leisure time index 15 Mental Health Inventory Total score= ((raw score of 5 items 5)/25) X 100 100 Copenhagen Burnout Inventory Total score = total average of three subscales (PB+WRB+CRB) Personal Burnout (BP) Work Related Burnout (WRB) Client Related Burnout (CRB) 100 Per ceived Wellness Survey Sum all subscales Wellness Magnitude=(P+E+S+P+S+I)/6 which now =xbar For each subscale (mean subscale xbar)2=subscale deviation Sum all subscale deviations and divide by 5 this = variance To find Wellness Balance (square root of var iance + 1.25 Total score or Wellness Composite = (Wellness Magnitude / Wellness Balance) Psychological Emotional Social Physical Spiritual Intellectual 0 29

PAGE 116

116 APPENDIX B INSTRUMENT Demographics

PAGE 117

117 Perceived Wellness Survey

PAGE 118

118

PAGE 119

119

PAGE 120

120

PAGE 121

121 Copenhagen Burnout I nventory (First e dition. November 1999)

PAGE 122

122

PAGE 123

123 Mental Health Inventory -5

PAGE 124

124 Baecke Physical Activity Questionnaire

PAGE 125

125

PAGE 126

126 MOS Social Support Survey

PAGE 127

127 A PPENDIX C PERMISSION TO USE INSTRUMENTS CBI Dear Keith Naugle, Thank you very much for your mail and for your interest in the CBI. I enclose some material for your information. If you choose to use the instrument in your study, please let me know (and send reports or papers to me when they are available, please). Best regards, Tage S. Kristensen Prof essor Fra: Naugle,Keith E [mailto:knaugle@hhp.ufl.edu] Sendt: 30. marts 2007 19:36 Til: Tage S. Kristensen (TSK) Emne: Dear Tage Kristenson I am a doctoral student at the University of Florida. I am conducting study for my dissertation about the well being of Athletic Trainers ATCs. I would like to ask your permission to use and/or modify the CBI that you developed and discussed in the 2005 article in Work and Stress. My goal is to develop a survey instrument that can be used to assess the well being of ATCs along the physical, social, and emotional dimensions. Your permission would be appreciated greatly. Thank you Keith Naugle MS, ATC MOSSS and Mental Health Instrument Dear Mr. Naugle, You will find permissions information on our website http://www.rand.org/health/surveys_tools.html including the following: All of the surveys and tools from RAND Health are public documents, available wit hout charge (for non -commercial purposes). Please provide an appropriate citation when using these products. In some cases, th e materials themselves include specific instructions for citation. Some materials listed are not available from RAND Health. Those links will take you to other websites, where you will find instructions for use. Thank you for your interest in RAND Health

PAGE 128

128 Dana Torres RAND Health From: Naugle,Keith E [mailto:knaugle@hhp.ufl.edu] Sent: Monday, March 19, 2007 5:33 PM To: RAND_Health@rand.org Subject: Dear Sir or Madame I am a doctoral student at the University of Florida. I am conducting study for my dissertation about the well being of Athletic Trainers. I would like to ask your permission to use and/or modify the MOS survey instrument that you developed and discussed in the 1991 article in soc sci and med. My goal is to develop a survey instrument that can be used to assess the well being of ATCs along the physical, social, and emotional dimensions. Your permission would be appreciated greatly. Thank you Keith Naugle Baecke Dear Keith Naugle, the questionnaire may be used freely and does not requi re for permission to be requested. Anyone is allowed to use the questionnaire on his/her own responsibility. (This was about the 20th request for permission that was sent to me in the last 15 years. Although the principal author did not become Professor the original paper has been referenced more than 600 times, thus becoming one of the top five most frequently referenced scientific papers of Dutch origin in 1982.) Best regards, Jan Burema MSc (biostatistician) From: Naugle,Keith E [mailto:knaugle@ hhp.ufl.edu] Sent: 29 March 2007 02:55 To: Burema, Jan Subject: Permission to use the Baecke Questionnaire Dear Jan Burema I am a doctoral student at the University of Florida. I am conducting study for my dissertation about the well being of Athletic Tr ainers (ATCs). I would like to ask your permission to use the Baecke Physical Activity Questionnaire. My goal is to develop a survey instrument that can be used to assess the well being of ATCs along the physical, social, and emotional dimensions. Your pe rmission would be appreciated greatly. Thank you Keith Naugle

PAGE 129

129 Keith Naugle MS, ATC PWS Keith Well done. You should get credit for finding me! : ) You have my permission to use the PWS and my best wishes as well. I refer you to my website perceivedwell ness.com for information about the scale etc. Typically, after you have collected your data, you will want to email me again with questions about the scoring methods. Regards Troy Adams On 4/2/07 3:14 PM, "Naugle,Keith E" wrote: Dr Adams I am a doctoral student looking into the wellness and health levels of athletic trainers. I am emailing for permission to use the Perceived Wellness Survey from the journal article Construct Validation of the Perceived Wellness Survey, publis hed in the American Journal of Health Studies for my dissertation. I had to search for Dr Troy Adams on the Google search engine so if you are not the Dr Adams that designed this I am sorry to bother you and could you please let me know. Thanks Keith Naugle MS, ATC

PAGE 130

130 APPENDIX D INSTITUTIONAL REVIEW BOARD UFIRB 02 Social & Behavioral Research Protocol Submission Title of Protocol: Perceptions of wellness and burnout among athletic trainers: Contributions of the wellness domains Principal Investig ator: Keith Naugle UFID #:56404790 Degree / Title: MS Clinical Coordinator UF ATEP Department: APK Mailing Address: P.O. Box 118205 100 Florida Gym Gainesville Fl 32611 Email Address & Telephone Number: knaugle@hhp.ufl.edu 3523920584 ext 1325 Co Investigator(s): UFID#: Supervisor: UFID#: Degree / Title: Department: Mailing Address: Email Address & Telephone Number: Date of Proposed Research: April 2008 to April 2009 Source of Funding (A copy of the grant proposal must be submitted with this protocol if funding is involved): Southeastern Athletic Trainers Association (SEATA) Research Grant and Mentorship Opportunity Grant pending Scientific Purpose of the Study: To measure the perceived wellness and burnout in the profession of athletic training and determine what factors lead to increased professional burnout and decreased perceived wellness. Describe the Research Methodology in Non Technical Language: ( Explain what will be done w ith or to the research participant. ) A questionnaire will be distributed to athletic trainers who are employed (full time) in the Southeast region of the National Athletic Trainers Association via email. The questionnaire is a combination of research ins truments which include: The Copenhagen Burnout Inventory1, The Perceived Wellness Survey 2,3, MOS Social Support Survey4, Mental Health Inventory5 and The Baecke Physical Activity Questionnaire6. In addition, demographic data will also be requested in the survey and will gather information regarding (gender, job setting, job title, age, education level, etc).

PAGE 131

131 Describe Potential Benefits and Anticipated Risks: (If risk of physical, psychological or economic harm may be involved, describe the steps t aken to protect participant.) There are no direct/immediate benefits to participating in this research. The hope is that burnout and perceived wellness in the individuals being surveyed can be decreased and improved respectively in the future. There are also no anticipated risks greater than those of daily activities that take place on a computer. Describe How Participant(s) Will Be Recruited, the Number and AGE of the Participants, and Proposed Compensation: Participants are recruited through the Nat ional Athletic Trainers Association (NATA) database and mailing list serve for research participants A total of 3000 subjects will be recruited for the administration of the instrument in or der to receive the anticipated 9 00 respondents (30% response rat e). The age range is from 2285 There is no compensation for the taking the survey. Describe the Informed Consent Process. Include a Copy of the Informed Consent Document: All subjects will be receiving (via email) a short introduction to this investig ation, specifically with regards to the purpose, methodology, and time commitment. Subjects will be informed that given the electronic format of the investigation, anyone who completes the survey is providing their informed consent to the primary investig ator. Principal Investigator(s) Signature: Supervisor Signature: Department Chair/Center Director Signature: Date:

PAGE 132

132 References for survey instruments: 1. Kristensen TS. Borritz M, Villadsen E, & Christensen. The Copenhagen Burnout Inve ntory: A new tool for the assessment of burnout. Work & Stress. 2005; 19(3): 192207. 2. Adams T, Benzer J, & Steinhardt M. The conceptualization and measurement of perceived wellness: integrating balance across and within dimensions. A m J H eal th Prom ot. 1997;11(3): 208218. 3. Adams TB. Bezner JR. & Whistler LS. The relationship between physical activity and indicators of perceived wellness. Am J Health Studies 1999; 15(3): 130138 4. Sherbourne CD. & Stewart AL. The MOS Social Support Survey. Soc Sc i Med 1991; 32(6): 705714 5. Veit CT. & Ware JE. The structure of psychological distress and well -being in general populations. J Consulting and Clin Psych. 1983; 51(5): 730742. 6. Florindo AA, & do Rosario Dias de Oliveira Latorre. Validation and rel iability of the Baecke questionnaire for the evaluation of habitual physical activity in adult men. Rev Bras Med Esporte 2003 9(3) :129135.

PAGE 133

133 APPENDIX E LETTER TO PARTICIPAN T Dear Fellow Certified Athletic Trainer: I am a doctoral student at the Univ ersity of Florida. I am conducting a survey of certified athletic trainers, the purpose of which is to learn about the levels of burnout and the perceived wellness of athletic trainers. I am asking you to participate in this interview because you have been identified as a certified athletic trainer. Participants will be asked to participate in a survey approximately 3 0 minutes. Your survey will be conducted by email via a website link. Please follow the link at the end of this letter to an online survey ti tled : Perceptions of wellness and burnout among athletic trainers: Contributions of the wellness domains The questionnaire consists of demographic questions, and questions relating to wellness, burnout, physical activity levels, social support and mental health. This student survey is not approved or endorsed by NATA. It is being sent to you because of NATA's commitment to athletic training education and research. If you have any questions about this research protocol, please contact me at 352 3920584 e xt 1325. Questions or concerns about your rights as a research participant rights may be directed to the UFIRB office, University of Florida, Box 112250, Gainesville, FL 32611; ph (352) 3920433. This is a completely anonymous questionnaire and upon subm ission, neither your name nor email address will be attached to your answers. Your information will be kept strictly confidential. As a fellow certified athletic trainer, your knowledge and opinions regarding this topic makes your input invaluable. Plea se take a few minutes to fill out the anonymous questionnaire you will find by clicking on this link and submit it within two weeks. By participating in the survey you are giving informed consent to use your data in the research project. Thank you for your time and consideration. Keith Naugle MS ATC LAT

PAGE 134

134 APPENDIX F CONTACT LIST REQUEST FORM

PAGE 135

135

PAGE 136

136 A PPENDIX G VARIABLE TABLES Perceived W ellness Survey Table G 1 Perceived W ellness Survey Perceived Wellness Survey # percent (responses) Question Responses Very Strongly Disagree Strongly Disagree Disagree Agree Strongly Agree Very Strongly Agree Response Count 1. I am always optimistic about my future. 0.9% (4) 3.1% (13) 15.8% (67) 33.9% (144) 35.3% (1 50) 11.1% (47) 425 2. There have been times when I felt inferior to most of the people I knew. 11.9% (51) 26.5% (113) 33.0% (141) 24.4% (104) 3.5% (15) 0.7% (3) 427 3. Members of my family come to me for support. 1.6% (7) 1.2% (5) 5.2% (22) 42.5% (181) 32.4% (138) 17.1% (73) 426 4. My physical health has restricted me in the past. 37.2% (159) 24.1% (103) 23.2% (99) 11.2% (48) 3.5% (15) 0.7% (3) 427 5. I believe there is a real purpose for my life 0.5% (2) 0.5% (2) 1. 4% (6) 27.6% (117) 30.0% (127) 40.1% (170) 424 6. I will always seek out activities that challenge me to think and reason. 0.5% (2) 0.2% (1) 5.9% (25) 39.3% (166) 32.0% (135) 22.0% (93) 422 7. I rarely count on good things happening to me. 18.3% (78) 33.6% (143) 30.5% (130) 12.7% (54) 3.3% (14) 1.6% (7) 426 8. In general, I feel confident about my abilities. 0.0% (0) 1.2% (5) 0.7% (3) 26.3% (112) 48.6% (207) 23.2% (99) 426 9. Sometimes I wonder if my family will re ally be there for me when I need them. 40.2% (171) 29.9% (127) 17.2% (73) 8.7% (37) 2.1% (9) 1.9% (8) 425

PAGE 137

137 Perceived Wellness Survey # percent #responses Question Responses Very Strongly Disagree Strongly Disagree Disagree Agree Strongly Agree Very Strongly Agree Response Count 10. My body seems to resist physical illness very well. 1.2% (5) 1.6% (7) 8.9% (38) 39.4% (168) 32.2% (137) 16.7% (71) 426 11. Life does not hold muc h future promise for me. 45.0% (192) 32.8% (140) 20.1% (86) 1.4% (6) 0.5% (2) 0.2% (1) 427 12. I avoid activities which require me to concentrate. 33.3% (141) 37.3% (158) 25.0% (106) 3.5% (15) 0.5% (2) 0.5% (2) 424 13. I alway s look on the bright side of things. 0.5% (2) 1.2% (5) 16.3% (69) 39.6% (168) 29.7% (126) 12.7% (54) 424 14. I sometimes think I am worthless individual. 56.2% (239) 19.1% (81) 18.4% (78) 5.6% (24) 0.5% (2) 0.2% (1) 425 15. My friend s know they can always confide in me to ask me for advice. 0.2% (1) 0.5% (2) 0.9% (4) 25.6% (109) 41.2% (175) 31.5% (134) 425 16. My physical health is excellent. 0.5% (2) 3.8% (16) 20.9% (88) 36.3% (153) 28.7% (121) 10.0% (42) 422 1 7. Sometimes I don't understand what life is all about. 19.3% (82) 26.7% (113) 27.4% (116) 23.6% (100) 2.6% (11) 0.5% (2) 424 18. Generally, I feel pleased with the amount of intellectual stimulation I receive in my daily life. 0.5% (2) 2.8% (12) 13.2% (56) 46.2% (196) 29.2% (124) 8.0% (34) 424 19. In the past, I have expected the best. 0.0% (0) 0.5% (2) 7.6% (32) 37.5% (157) 37.7% (158) 16.7% (70) 419 20. I am uncertain about my ability to do things well in the future. 17. 5% (74) 36.2% (153) 28.8% (122) 11.1% (47) 3.8% (16) 2.6% (11) 423 21. My family has been available to support me in the past 0.2% (1) 0.5% (2) 2.8% (12) 29.0% (123) 32.1% (136) 35.4% (150) 424

PAGE 138

138 Perceived Wellness Survey # percent #re sponses Question Responses Very Strongly Disagree Strongly Disagree Disagree Agree Strongly Agree Very Strongly Agree Response Count 22. Compared to people I know, my past physical health has been excellent. 0.2% ( 1) 2.3% (10) 10.1% (43) 35.9% (153) 32.9% (140) 18.5% (79) 426 23. I feel a sense of mission about my future. 0.2% (1) 1.2% (5) 9.7% (41) 35.9% (151) 34.7% (146) 18.3% (77) 421 24The amount of information that I process in a typical day is just about right for me (i.e. not to much and not too little. 2.1% (9) 7.3% (31) 19.2% (82) 51.2% (218) 16.7% (71) 3.5% (15) 426 25. In the past, I hardly ever expected things to go my way. 14.4% (61) 32.7% (139) 38.1% (162) 12.5% (53) 1.9% (8) 0.5% (2) 425 26. I will always be secure with who I am. 0.2% (1) 1.2% (5) 11.1% (47) 38.4% (163) 31.5% (134) 17.6% (75) 425 27. In the past, I have not always had friends with whom I could share my joys and sorrows. 20.2% (86) 21. 2% (90) 29.4% (125) 22.4% (95) 5.6% (24) 1.2% (5) 425 28. I expect to always be physically healthy. 0.5% (2) 3.3% (14) 19.1% (81) 39.9% (169) 25.9% (110) 11.3% (48) 424 29. I have felt in the past that my life was meaningless. 36.4% (154) 27.4% (116) 25.3% (107) 9.0% (38) 1.4% (6) 0.5% (2) 423 30. In the past, I have generally found intellectual challenges to be vital to my overall well being. 0.0% (0) 0.9% (4) 6.1% (26) 42.8% ( 181) 35.5% (150) 14.7% (62) 423 31. Things will not work out the way I want them to in the future. 24.4% (103) 34.4% (145) 32.2% (136) 7.6% (32) 1.4% (6) 0.0% (0) 422

PAGE 139

139 32. In the past, I have felt sure of myself among s trangers 0.7% (3) 2.6% (11) 15.8% (67) 43.5% (185) 28.0% (119) 9.4% (40) 425 33. My friends will be there for me when I need help. 0.0% (0) 1.0% (4) 4.0% (17) 42.4% (178) 31.7% (133) 21.0% (88) 420 34. I expect my p hysical health to get worse. 16.0% (67) 26.0% (109) 37.0% (155) 18.4% (77) 2.4% (10) 0.2% (1) 419 35. It seems that my life has always had purpose. 0.2% (1) 0.5% (2) 9.7% (41) 40.0% (169) 31.4% (133) 18.2% ( 77) 423 36. My life has often seemed void of positive mental stimulus. 21.5% (91) 32.8% (139) 38.4% (163) 5.4% (23) 1.7% (7) 0.2% (1) 424

PAGE 140

140 Copenhagen Burnout Inventory Table G 2 Copenhagen Burnout Inventory Cop enhagen Burnout Inventory Question Responses Personal work burnout Always Often Sometimes Seldom Never/al most never Response count 1. How often do you feel tired? 7.8% (33) 45.4% (191) 38.2% (161) 7.6% (32) 1.0% (4) 422 2. How ofte n are you physically exhausted? 1.7% (7) 20.6% (87) 44.1% (186) 27.7% (117) 5.9% (25) 422 3. How often are you emotionally exhausted? 3.3% (14) 27.8% (117) 38.5% (162) 24.5% (103) 5.9% (25) 421 4. How often do you think: "I can't take it anymore"? 1.7% (7) 11.4% (48) 24.2% ( 102) 34.7% (146) 28.0% (118) 421 5. How often do you feel worn out? 3.6% (15) 32.1% (134) 38.5% (161) 21.5% (90) 4.3% (18) 418 6. How often do you feel weak and susceptible to illness? 0.7% (3) 4 .5% (19) 23.6% (99) 46.2% (194) 25.0% (105) 420

PAGE 141

141 Table G 2 Copenhagen Burnout Inventory Question Responses Work burnout subscale To a very high degree To a high degree Some what To a low degree To a very low degree Response count 1. Is yo ur work emotionally exhausting? 7.1% (30) 19.7% (83) 39.1% (165) 24.4% (103) 9.7% (41) 422 2. Do you feel burnt out because of work? 6.7% (28) 14.7% (62) 29.9% (126) 28.3% (119) 20.4% (86) 421 3. Does your work frustrate you? 7.1% (30) 14.0% (59) 37.1% (156) 26.1% (110) 15.7% (66) 421 Always Often Some times Seldom Never/almost never 4. Do you feel worn out at the end of the working day? 6.1% (2 6) 33.8% (143) 37.4% (158) 17.0% (72) 5.7% (24) 423 5. Are you exhausted in the morning at the thought of another day at work? 3.1% (13) 10.4% (44) 28.1% (119) 33.3% (141) 25.1% (106) 421 6. Do you feel that e very working hour is tiring for you? 2.4% (10) 4.3% (18) 17.9% (75) 39.4% (165) 36.0% (151) 419 7. Do you have enough energy for family and friends during leisure time? 20.6% (87) 41. 1% (174) 31.9% (135) 5.7% (24) 0.7% (3) 423

PAGE 142

142 Table G 2 Copenhagen Burnout Inventory Question Responses Response count Client burnout scale To a very high degree To a high degree Somewhat To a low degree To a very low degree 1. Do you find it hard to work with athletes/clients? 0.7% (3) 2.1% (9) 14.2% (60) 41.8% (177) 41.1% (174) 423 2. Do you find it frustrating to work with athletes/clients? 1.4% (6) 3.8% (16) 19.9% (84) 40.4% (171) 34.5% (146) 423 3. Does it drain your energy to work with athletes/clients? 1.2% (5) 2.6% (11) 18.1% (76) 39.0% (164) 39.0% (164) 420 4. Do you feel that you give more then you get back when you work with athletes/clients? 9.0% (38) 2 2.5% (95) 29.6% (125) 23.7% (100) 15.2% (64) 422 Always Often Sometimes Seldom Never/almost never 5. Are you tired of working with athletes/clients? 1.7% (7) 4.5% (19) 20.8% (88) 30.3% (128) 42.8% (181) 423 6. Do you sometimes wonder how long you will be able to continue working with athletes/clients? 4.3% (18) 14.3% (60) 26.8% (113) 27.1% (114) 27.6% (116) 421

PAGE 143

143 Mental Health Inventory 5 Table G 3 Mental Health Inventory 5 Table G 3 Mental Health Inventory 5 Que stion Responses Response Count All of the time Most of the time A good bit of the time Some of the time A little of the time None of the time 1. During the past month, how much of the time were you a happy person? 6.6% (28) 56.9% (240) 20.9% (88) 12.3% (52) 3.1% (13) 0.2% (1) 422 2. How much of the time, during the past month, have you felt calm and peaceful? 3.6% (15) 33.6% (142) 22.5% (95) 24.9% (105) 14.2% (60) 1.2% (5) 422 3. How much of the time during the past month, have you been a very nervous person? 0.0% (0) 4.5% (19) 6.9% (29) 19.7% (83) 46.1% (194) 22.8% (96) 421 4. How much of the time, during the past month, have you felt downhearted and blue? 0.7% (3) 2.1% (9) 8.1% (34) 17.1% (72) 46.3% (195) 25.7% (108) 421 5. How much of the time, during the past month, have you felt so down in the dumps that nothing could cheer you up? 0.2% (1) 1.2% (5) 2.4% (10) 8.3% (35) 14.8% (62) 73.1% (307) 420

PAGE 144

144 Baecke Physical Activity Questionnaire Table G 4 Baecke Physical Activity Baecke Physical Activity Questionnaire Question Responses 1. Job See Table Work Index Never Seldom Sometimes Often Always Response cou nt 2. At work I sit: 2.6% (11) 25.4% (107) 45.6% (192) 25.4% (107) 1.0% (4) 421 3. At work I stand: 0.0% (0) 2.9% (12) 28.8% (121) 58.6% (246) 9.8% (41) 420 4. At work I walk: 0.0% (0) 5.5% (23) 26.1% (109) 60.0% (251) 8.4% (35) 418 5. At work I lift heavy loads: 7.9% (33) 36.5% (153) 35.3% (148) 17.9% (75) 2.4% (10) 419 6. At work I am tired: 4.8% 20) 16.9% (71) 47.7% (201) 28.3% (119) 2.4 % (10) 421 7. At work I sweat: 13.1% (55) 22.9% (96) 35.8% (150) 22.7% (95) 5.5% (23) 419 Much heavier Heavier As heavy Lighter Much lighter 8. In comparison with others of my own age, I think my work is physically: 5.8% (24) 31.9% (133) 35.3% (147) 23.0% (96) 4.1% (17) Sport Index 9. Do you do exercise activity? Yes No 77.80% 22.20% 418 9 A If yes what exercise activity do you do most frequently <1 1 to 2 2 to 3 3 to 4 >4 9 B1. How many hours per week? 11.8% (43) 17.6% (64) 22.3% (81) 21.5% (78) 26.7% (97) 363 <1 1 to 3 4 to 6 7 to 9 >9 9 C1. How many months a year? 7.4% (27) 3.3% (12) 9.9% ( 36) 10.2% (37) 69.2% (252) 364 9 A2. If you do a second exercise activity, what exercise activity do you do? <1 1 to 2 2 to 3 3 to 4 >4 9 B2. How many hours per week? 16.5% (40) 36.8% (89) 23.6% (57) 13 .2% (32) 9.9% (24) 242

PAGE 145

145 Table G 4 continued Baecke Physical Activity Questionnaire Question Respo nses <1 1 to 3 4 to 6 7 to 9 >9 9 C2. How many months a year? 5.9% (14) 8.0% (19) 15.1% (36) 16.0% ( 38) 55.0% (131) 238 Much more More The same Less Much Less 10. In comparison with others of my own age I think my physical activity during leisure time is: 10.2% (40) 27.9% (110) 32.2% (127) 23.4% (92) 6.3% (25) 394 11. During leisur e time I sweat: 12.4% (52) 25.9% (109) 40.4% (170) 18.8% (79) 2.6% (11) 421 Never Seldom Sometimes Often Always 12. During leisure time I play a sport: 21.9% (92) 30.2% (127) 36.8% (155) 10.5% (44) 0.7% (3) 421 L eisure Index Never Seldom Sometimes Often Always 13. During leisure time I watch television: 1.4% (6) 7.9% (33) 41.9% (175) 46.2% (193) 2.6% (11) 418 14. During leisure time I walk: 4.3% (18) 24.6% (103) 44.5% (186) 24.2% (101) 2.4% (10) 418 15. During leisure time I cycle: 45.8% (191) 27.1% (113) 17.5% (73) 8.6% (36) 1.0% (4) 417 <5 5 to 15 15 to 30 30 to 45 >45 16. How many minutes do you walk and or cycl e per day to and from work, school and shopping? 37.6% (157) 17.9% (75) 21.1% (88) 12.0% (50) 11.5% (48) 418

PAGE 146

146 MOS Social Support Table G 5 MOS Social Support Survey Table G 5 Social Support Q uestion Responses None of the time A little of the time Some of the time Most of the time All of the time Response Count 1. Someone to confide in or talk to about yourself or your problems. 2.4% (10) 10.0% (42) 17.4% (73) 31.0% (130) 39.1% (164) 419 2. Someone to share your most private worries and fears with. 6.2% (26) 11.7% (49) 16.7% (70) 26.0% (109) 39.4% (165) 419 3. Someone to turn to for suggestions about how to deal with a personal problem. 1.2% (5) 13.7% (57) 16.5% (69) 30.5% (127) 38.1% (159) 417 4. Someone to help you if you were confined to bed. 5.5% (23) 9.8% (41) 13.9% (58) 28 .7% (120) 42.1% (176) 418 5. Someone to prepare your meals if you were unable to do it yourself. 6.0% (25) 10.3% (43) 12.2% (51) 28.1% (117) 43.4% (181) 417 6. Someone to help with daily chores if you were sick. 7. 4% (31) 11.0% (46) 12.9% (54) 27.1% (113) 41.5% (173) 417 7. Someone who shows you love and affection. 4.3% (18) 8.8% (37) 12.2% (51) 20.3% (85) 54.4% (228) 419 8. Someone to love and make you feel wanted 4.8% (20) 8.1% (34) 12.9% (54) 20.5% (86) 53.7% (225) 419 9. Someone who hugs you. 4.5% (19) 9.6% (40) 12.4% (52) 21.3% (89) 52.2% (218) 418 10. Someone to have a good time with. 0.7% (3) 6.9% (29) 17.9% (75) 32.7% (137) 41.8% (175) 419 11. Someone to get together with for relaxation. 2.9% (12) 6.5% (27) 20.4% (85) 29.0% (121) 41.2% (172) 417 12. Someone to do something enjoyable with. 1.4% (6) 5.8% (24) 19.7% (82) 30.9% (129) 42.2% (176) 417

PAGE 147

147 LIST OF REFERENCES 1. Prentice WE. Arnheim's Principles of Athletic Training: A Competency -Based Approach. 13th Edition Ed New York: McGraw Hill; 2009. 2. Hunt V Dangerous dedication: burnout a risk in the Profession. NATA News. Vol November; 2000:810. 3. Hendrix AE, Acevedo EO, Hebert E. An examination of stress and burnout in certified athletic trainers at d ivision I -A universities. J Athl Train. 2000;35:139144. 4. Geick J, Brown RS, Shank RH. The burnout syndrome among athletic trainers. J Athl Train. 1982;17:3642. 5. Geick J. Stress management and the athletic trainer. J Athl Train. 1984;19:115119. 6. Cu ppett M, Latin RW. A survey of physical activity levels of certified athletic trainers. J Athl Train. 2002;37:281285. 7. Mazerolle SM BJ, Casa DJ. Work Family Conflict, Part I: Antecedents of Work Family Conflict in National Collegiate Athletic Associatio n Division I -A Certified Athletic Trainers. J Athl Train. 2008;43 505512. 8. Mazerolle SM BJ, Casa DJ. Work Family Conflict, Part II: Job and Life Satisfaction in National Collegiate Athletic Association Division I -A Certified Athletic Trainers. J Athl Tr ain. 2008;43:513522. 9. Clapper DC, Harris LL. Reliability and validity of an instrument to describe burnout among collegiate athletic trainers. J Athl Train. 2008;43:6269. 10. Kania ML, Meyer BB, Ebersole KT. Personal and environmental characteristics p redicting burnout among certified athletic trainers at National Collegiate Athletic Association institutions J Athl Train. 2009;44:5866. 11. Waldrop S. Maintaining enthusiasm in a challenging environment. PT Magazine. Vol June; 2003:3845. 12. Shanafelt T D, Bradley KA, Eipf JE, Back AL. Burnout and self reported patient care in an internal medicine residency program. Annals of Internal Medicine. 2002;136:358367. 13. Gunderson L. Physician burnout. Annals of Internal Med. 2001;135:145148. 14. Balogun JA, Titiloye V, Balogun A, Oyeyemi A, Katz J. Prevalence and determinants of burnout among physical and occupational therapists. J of Allied Health 2002;31:131139. 15. Bell RB, Davison M, Sefcik D. A first survey measuring burnout in emergency medicine physic ian assistants. JAAPA. 2002;15:4055.

PAGE 148

148 16. Allied Health Professions Association's website. Available at: http://www.ama.assn.org/ama/pub/category/3584.html Accessed January 27 2007. 17. The FACTS about certified Athletic Trainers and the National Athletic Trainers' Association. Available at: http://nata,org/consumer/docs/factsaboutathletictrainers.pdf Accessed January 27, 2007. 18. Honkonen T, Ahola K, Pertovaara M, et al. The association between burnout and physical illness in the general population Results from the Finnish Health 2000 Study. J of Psychosom Rrch 2006;61:5966. 19. Shaufeli WB, Greenglass ER. Introduction to special issue on burnout and health. Psychology and Health. 2003;16:501510. 20. Capel SA. Psychological and organizational factors related to burnout in athletic trainers. Athl Train. 1986;21:322327. 21. Mclain AJ. An o verview of b urno ut in a thletic t rainers. Athl Ther Today. 2005;10:1113. 22. Kristensen TS, Borritz M, Villadsen E, Christensen KB. The Copenhagen Burnout Inventory: A new tool for the assessment of burnout Work & Stress. 2005;19:192207. 23. Kania ML, Meyer BB, Ebersole KT. Burnout among ATCs at NCAA institutions is related to stress and pressure from coaches. J Athl Train. 2006;41:S 72. 24. Mazerolle SM, Bruening JE, Casa DJ, Burton L, Van Heest J. The impact of work -family conflict on job satisfaction and life satisfact ion in Division I -A athletic trainers. J Athl Train. 2006;41:S 73. 25. Maslach C, Wilmar SB, Leiter MP. Job burnout. Annu Rev Psychol. 2001;52:397422. 26. Maslach C, Jackson SE. The measurement of experienced burnout. J of Occupational Behavior. 1981;2: 99113. 27. Glasberg AL, Eriksson S, Norberg A. Burnout and 'stress of conscience' among healthcare personal. J of Advanced Nursing. 2006;57:392403. 28. Akroyd D, Caison A, Adams RD. Burnout in radiation therapist: The predictive value of selected stresso rs. International Journal Radiation Oncology Biology and Physics. 2002;52:816821. 29. Spickard A, Gabbe SG, Christensen JF. Midcareer burnout in generalist and specialist physicians. JAMA. 2002;288:14471450. 30. Maslach C, Goldberg J. Prevention of burn out: New perspectives. Applied & Preventive Psychol 1998;7:6374.

PAGE 149

149 31. Schaufeli WB, Greenglass ER. Introduction to special issue on burnout and health. Psychol and Health. 2003;16:501510. 32. Reed S, Giacobbi PR, Jr. The stress and coping responses of cer tified graduate athletic training students. J Athl Train. 2004;39:193200. 33. Donohoe E, Nawawi A, Wilker L, Schindler T, Jette DU. Factors associated with burnout of physical therapists in Massachusetts rehabilitation hospitals. Phys Ther. 1993;73:750761. 34. Jenaro C, Flores N, Arias B. Burnout and coping in human service providers. Professional Psychology: Research and Practice. 2007;38:8087. 35. Linzer M, Visser MR, Oort FJ, Smets EM, McMurray JE, de Haes HC. Predicting and preventing physician burnout: results form the United States and the Netherlands. Am J Med. 2001;111:170175. 36. Ozyurt A, Hayran O, Sur H. Predictors of burnout and job satisfaction among Turkish physicians. Q J Med. 2006;99:161169. 37. Schuster N, Nelson D, Quisling C. Burnout among physical therapists. Phys Ther. 1984;64:299303. 38. Winwoood PC, Winefield AH. Comparing two measures of burnout among dentists in Australia. Int J Stress Manage. 2004;11:282289. 39. Borritz M, Rugulies R, Christensen KB, Villadsen E, Kristensen TS Burnout as a predictor of self reported sickness absence among human service workers: prospective findings from a three year follow up of the PUMA study. Occupational Environment Med. 2006;63:98106. 40. Schaufeli WB, Taris TW. The conceptualization and measurement of burnout: Common ground and worlds apart. Work & Stress. 2005;19:256262. 41. Gabassi PG, Cervai S, Rozbowsky P, Semeraro A, Gregori D. Burnout syndrome in the helping professions. Psychological Reports. 2002;90:309314. 42. Calzi SL, Farinel li M, Ercolani M, Aliante M, Manigrasso V, Taroni AM. Physical rehabilitation and burnout: different aspects of the syndrome and comparison between healthcare professional involved. Eur Med Phys. 2006;42:2736. 43. Bradley JR, Cartwright S. Social support, job stress, health, and job satisfaction among nurses in the United Kingdom. Int J of Stress Management. 2002;9:163182. 44. Sherbourne CD, Stewart AL. The MOS social support survey. Soc Sci Med. 1991;32:705714.

PAGE 150

150 45. Halbesleben JRB. Sources of social sup port and burnout: A meta analytic test of the conservation of resources model. J App Psychol. 2006;91:11341145. 46. Callaghan P, Morrissey J. Social support and health: A review. J of Advanced Nursing. 1993;18:203210. 47. Julliard K, Klimenko E, Jacob MS Definitions of health among health care professionals. Nursing Science Quarterly. 2006;19:265. 48. Organization WH. Preamble to the constitution of the WHO. Official Records of the World Health Organization. 1948:2. 49. Services. USDoHaH. Healthy P eople 2010. Washington D.C. 2000. 50. McManus IC, Winder BC, Gordon D. The causal links between stress and burnout in a longitudinal study of UK doctors. Lancet. 2002;359:289290. 51. Abramson S, Stein J, Schaufeli M, Frates E, Rogan S. Personal exercise habit s and counseling practices and primary care physicians: A national survey. Clinical J Sport Med. 2000;10:4048. 52. Peterson DF, Degenhardt BF, Smith CM. Correlation between prior exercise and present health and fitness status of entering medical students. JAOA. 2003;103:361366. 53. Association NAT. Athletic Training Educational Competencies. Dallas TX 2006. 54. CDC. Health risks in the United States, behavioral risk factor surveillance system. 2006. 55. Haddock CK, Alker SCP, Sara AP, et al. The validit y of self rated health as a measure of health status among young military personal: evidence from a cross -sectional survey. Health and Quality of Life Outcomes. 2006;4:57. 56. Craig CA, Marshall AL, Sjostrom M, et al. International Physical Activity Questi onnaire: 12 country reliability and validity. Med Sci Sports Exerc. 2003;35:13811395. 57. Florindo AA, do Rosario Dias de Oliveira Latorre M. Validation and reliability of the Baecke questionnaire for the evaluation of habitual physical activity in adult men. Rev Bras Med Esporte 2003;9:129135. 58. Pols MA, Peeters PHM, Bueno de -Mesquita HB, et al. Validity and reliability of a modified Baecke Questionnaire of physical activity. Int J Epidimiol. 1995;24:381387. 59. Baecke JAH, Burema J, Frijters JER. A s hort questionnaire for the measurement of habitual physical activity in epidemiological studies. Am J Clin Nutr. 1982;36:936942. 60. Adams T, Benzer J, Steinhardt M. The conceptualization and measurement of perceived wellness: integrating balance across a nd within dimensions Am J Health Promot. 1997;11:208218.

PAGE 151

151 61. Anderson L. Argyris and Schons theory on congruence and learning. Research Papers in Action Research. Available at: Available at http://www.scu.edu.au/schools/gcm/ar/arp/argyris2.html#a_as_notes Accessed February 21, 2007. 62. Dick B, Dalmau T. Argyris and Schon: some elements of their models. Research Papers in Action Research [Available at: Available at http://www.scu.edu.au/schools/gcm/ar/arp/argyris2.html Accessed February 21, 2007. 63. Groth JJ, Ayers SF, Miller MG, Arbogast WD. Self reported health and fitness habits of certified athl etic trainers. J Athl Train. 2008;43:617623. 64. Viet CT, Ware JE. The structure of psychological distress and well -being in general populations. J Counsel Clinic Psychol 1983;51:730742. 65. McCabe CJ, Thomas KJ, Brazier JE, Coleman P. Measuring the ment al health status of a population: a comparison of the GHO 12 and the SF 36(MHI 5). Brit J Psychol. 1996;169:516521. 66. Wasserman DA, Stewart AL, Delucchi KL. Social support and abstinence form opiates and cocaine during opiod maintenance treatment. Drug Alcohol Dependence. 2001;65:6575. 67. Gjesfeld CD, Greeno CG, Kim KH. A confirmatory factor analysis of an abbreviated social support instrument: the MOS SSS. Res Soc Work Pract 2007;Sage publications Online first:1 7. 68. Bezner JR, Adams TB, Whistler LS The relationships between physical activity and indicators of perceived wellness Am J Health Stud. 1999;15:130138. 69. Adams T, Bezner JR, Garner L, Woodruff S. Construct validation of the Perceived Wellness Survey. Am J Health Studies. 1998;14:212220. 70. Hoeymans N, Garssen AA, P. WG, FM. VP. Measuring mental health of the Dutch populations: a comparison of the GHQ 12 and the MHI 5. Health and Quality of Life Outcomes. 2004;2:16. 71. Subcommittee TCiMSCHSR. Multiple Sclerosis Quality of Life Inventor y: A User's Manual New York, NY: National Multiple Sclerosis Society; 1997. 72. Health R. Medical Outcomes Study: Social Support Survey. Available at: http://www .rand.org/health/surveys_tools/mos/mos_socialsupport_scoring.html Accessed March 27th 2008. 73. Dillman DA. Mail and Internet Surveys, the Tailored Design Method. 2nd ed. Hoboken, NJ: John Wiley & Sons, Inc; 2007.

PAGE 152

152 74. Kline RB. Principles and Practice of Structural Equation Modeling. New York The Guilford Press; 1998. 75. Trost SG, Saunders R, Ward DS. Determinants of physical activity in middle school children. Am j Health Behav. 2002;26:95102. 76. Creswell JW. Educational Research Planning, Conducting, and Evaluating Quantitative and Qualitative Research. 2nd ed. Upper Saddle River, NJ: Pearson Education Inc. ; 2005. 77. George D, Mallory P. SPSS for Windows Step by Step: A Simple Guide and Reference 15.0 Update 8th ed. Boston, MA: Pearson Education In c; 2008. 78. Kelly MJ, Dunstan FD, Lloyd K, Fone DL. Evaluating cutpoints for the MHI 5 and MCS using the GHQ 12: a comparison of five different methods. BMC Psychiatry. 2008;8:10. 79. Gorter RC, Albrecht G, Hoogstraten J, Eijkman MAJ. Professional burnout among Dutch dentist. Community Dent Oral Epidemiol. 1999;27:109116. 80. McChesney JC, Peterson M. There is no off -season anymore. Athl Ther Today. 2005;10:610. 81. Mensch J. It's a quality of life issue. Athl Ther Today. 2005;10:3435. 82. Lockard BC. A thletic trainers providing healthcare for athletes of all kinds. Occup Outlook Quart. Vol Spring 2005; 2005:3841. 83. Scriber KC, Alderman MH. The challenge of balancing our professional and personal lives. Athl Ther Today. 2005;10:1417. 84. Thomas NK. R esident burnout. JAMA. 2004;292:28802889.

PAGE 153

153 BIOGRAPHICAL SKETCH Keith Naugle was born in Johnstown, PA on 1975, to Martin and Deborah Naugle. Growing up, Keith and his siblings were pushed to achieve as much as possible in academic endeavors Keiths pa ssion for learning and knowledge helped him to push through the many years of underg raduate and graduate studies. Keith went on to earn a BS from Roanoke College (Salem, VA) in sports medicine/athletic training. After graduating in 200 0 Keith enrolled a t the University of Illinois in Urbana -Champaign, where he earned a MS in Kinesiology in 2002. As a graduate assistant He worked with a variety of college and Paralympics athlet es with the University of Illinois wheelchair basketball and track teams Thes e experiences set the path in motion for him to pursue real life experience in the field of athletic training. Keith worked for two years as a high school athletic trainer and middle school health and PE teacher. Those two years spent as a teacher truly op ened up his eyes to a calling in higher education. Following his role as a high school athletic trainer, Keith attended the Uni versity of Florida to complete his PhD in 2009. While pursuing his doctorate, Keith was hired as the Clinical Coordinator of the undergraduate athletic training education program and was blessed with an abundant of great students and faculty who supported this dream of becoming a Doctor of Philosophy.