Prevalence and associated features of depression and psychological distress in polio survivors

Prevalence and associated features of depression and psychological distress in polio survivors

1056 Prevalence and Associated Features of Depression and Psychological Distress in Polio Survivors Denise G. Tate, PhD, Martin Forchheimer, MPP, Ne...

745KB Sizes 0 Downloads 1 Views


Prevalence and Associated Features of Depression and Psychological Distress in Polio Survivors Denise G. Tate, PhD, Martin Forchheimer,

MPP, Ned Kirsch, PhD, Frederick


MD, Allison Roller, MA

Tate DC, Forchheimer M, Kirsch N, Maynard F, Roller A. Prevalence and associated features of depresABSTRACT. sion and psychological distress in polio survivors. Arch Phys Med Rehabil 1993;74:1056-60. l This cross-sectional study examines the prevalence of psychological distress and depression among I16 polio survivors. It investigates demographic, medical, and coping differences between subjects with (n = 17) and without (n = 99) these symptoms. Subjects were administered the Brief Symptom Inventory (BSI), the Coping with Disability Inventory (CDI), and a questionnaire about their polio histories. The BSI provided measures of psychological distress and depression that defined the subgroups. The CD1 assessed coping behaviors. BSI scores for the overall sample were within the normal range indicating no major distress, depression, nor elevated somatic complaints. Several significant differences were found between the two subgroups. On average, depressed/distressed subjects reported an increase in pain (p < .Ol) and further deterioration of their medical status since the time of their physical best subsequent to the onset of polio (p < .Ol). They consistently rated their health as poorer than did nondepressed/nondistressed subjects (p < .OOl). They also reported less satisfaction with life and their occupational status (p < .OOl) and displayed poorer coping behaviors combined (p < .OOl). Selected variables such as life satisfaction. pain, decrease in activity, and current living situation accounted for 51% of the variance when predicting distress and depression among this group of polio survivors. lcl 1993 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and


The importance of psychological factors in successful adaptation to disability or chronic illness, such as breast cancer. diabetes, and spinal cord injury. has been increasingly recognized.‘-3 However, little is empirically known about the psychological adjustment process of people who are coping with the late effects of polio, including health behaviors and accompanying lifestyle changes. Polio survivors, like other populations who lived for many years with irrevocable losses due to illness (ie, the permanent loss of bodily functions or the functional loss of particular body parts), are of particular interest to rehabilitation professionals because of the new interest in studving aging among people with chronic disabilities4 After initial rehabilitation from acute polio and adapting to any permanent losses of muscle strength and function, many are now experiencing “secondary disability” with new symptoms, both physical (fatigue, muscle weakness, and pain) and psychological (stress, depression. anxiety).5,6 Current findings in health psychology research suggest that persons who believe that they have little control over their health experience high levels of psychological distress and depression and often exhibit a lack of appropriate coping skills when faced by stressful events.‘,* These factors can contribute to poor rehabilitation treatment outcomes. The assumption that distress or depression are necessary in coping with loss is prevalent in the literature.‘.’ ’ but Wortman and Silver” dispute this view. They refute the assumption From

the Department



Suhmitted This

of Physical



for publication

June 75.

research was supported



No commercial


1991. January


31, 1993.

the Center

party havmg a direct or indirect confer


the authors

are associated.

a benefit

for Disease Control.



upon the authors

in the SuhJeCt

matter ofthis

or upon an) organization


arc not available.

‘( 1 1993 bv the American c&Physical





has or will


and Rehabilitation.


by a grant from






of Rehabilitation

and Rehahilitaiion


Phys Med Rehabil Voi 74, October 1993


and the American

that depression and distress generally follow loss. They further contend that inaccurate assumptions or expectations held by health care professionals may unnecessarily exacerbate feelings of distress that are present among those who encounter loss. Different findings have been reported with respect to the prevalence of depression and psychological distress among polio survivors. Based on her personal and professional experience, Frick13 suggests that polio survivors may experience personal devaluation, isolation. and depression as psychological responses to the onset of postpolio symptoms. A more recent study by Conrady and colleagues’4 used the Symptom Checklist-90-Revised (SCL-90-R) to assess distress in polio survivors seeking care in a postpolio clinic. They noted exceptionally high distress especially with respect to symptoms related to somatization. depression, and psychoticism. A study by Freidenberg and colleagues’5 suggests also that disturbances of mood were common among postpoliomyelitis patients with postpolio syndrome (PPS) being seen at an orthopaedic outpatient poliomyelitis clinic. However. patients with PPS did not have greater depression, as measured by the Beck Depression Inventory (BDI). compared to postpoliomyelitis patients without PPS. Cameron,16 on the other hand. saw no evidence of dcpression in a group of 34 individuals diagnosed as experiencing PPS. I6 He also used the BDI to assess depression and compared these results to those of a group of normal control subjects. Differences were found, however, on the specific coping skills used by PPS subjects. with these subjects engaging more frequently in denial as a positive coping skill. Our purpose was twofold: first. to clarify these controversial findings by determining the prevalence of depression and psychological distress in a sample of 1 16 polio survivors who had a diagnoses of PPS and who enrolled in a voluntary state registry: and second, to identify demographic. medical and behavioral (ie. coping behav’ors) dif-



fcrences between depressed and nondepressed polio survivors in this sample, thus enabling the prediction of elevated levels of psychological distress and depression. The prevalence of depression and psychological distress was expected to be similar to that in the normal population with the possible exception of distress of a somatic nature. due to the many physical symptoms associated with the late effects of polio. Depressed/distressed subjects were expected to exhibit poorer coping behaviors; to perceive their own health as poorer: to report more pain and other physical symptoms: to report less satisfaction with life: and to be engaged in fewer physical and social activities. It was also expected that they would be older, live alone and without family support, not be working. and be experiencing more severe physical limitations (ie, abnormal gait pattern) related to polio. METHOD

Subjects Initially. 874 subjects from the Michigan Department of Public Health Post-Polio Registry were mailed questionnaires asking about their willingness to participate in the study.” Approximately 60% (57 1) of the questionnaires were returned and 453 subjects agreed to participate. Twenty-eight did not meet the criteria of having polio onset between 1946 and 1955. Of the remaining 42.5. a stratifed sample of 120 was selected. Stratification was based on sex. severity ofpolio residual weakness based on locomotor limitation at the time of “physical best” after recovery from polio. and self-perceived decrease in ability to do the routine activities of daily living currently compared to time of “physical best.” Data for analysis was available on I I6 subjects. This research was approved by the University’s Committee to Review Grants for Clinical Research and Investigation Involving Human Beings.

Procedures The 1 I6 subjects were tirst examined by physicians to determine the extent of their limitations related to polio, validate their diagnosis. assess gait abnormality. number of comorhidities and deterioration of their medical status from when the! were at their best to now. AI1 subjects completed the Brief Symptom Inventory (BSI). the Coping with Disability Inventory (CDI), and a personal questionnaire that provided detailed demographic. family history. and polio-related history information. After assessing participants’ levels of distress, the overall sample was divided into depressed/distressed (II = 17) and nondepressed/nondistressed (n = 99) subgroups. according to subjects’ BSI scores. Standardized t scores of 65 or greater on both the BSI/GSI (General Severity Index) and BSI/DEP (BSI’s Depression scale) were interpreted as suggesting abnormallv elevated levels of psychological distress and depression.iR Mean standardized scores for BSI scales werl: calculated and compared to scale standard scores over 65 in order to determine the overall levels of distress and depression of this sample in relation to a normative population (able-bodied nonpatients). Mean BSI scores were also calculated separately for men and women.



Differences between psychologically distressed/depressed polio subjects and those without these psychological symptoms were assessed with respect to subjects’ demographic characteristics: type and severity of polio: intensity and frequency of pain symptoms: self-perceptions of subject’s health, satisfaction with life, and occupational status. and decrease in ability to independently perform activities of daily living: and self-reported coprng behaviors. Appropriate coping behaviors were defined as behaviors and interpersonal responses designed to reflect positive adjustment’” and were assessed using the CD].“’ Mean scores for the BSI and CD1 were statistically compared using Students I-tests for independent means. Pearson y’ (Systat, 1989)” were used to compare the two subgroups with respect to demographics. disability-related characteristics. and their rated self-perceptions. Relationships among potential predictors and BSI outcome variables were examined using Spearman and Pearson correlations. To analyze the variables’ ability to predict depression/distress scores among postpolio survivors, a multiple linear-regression model was specified. Relations among these data were inferred based on the presence or absence of significant [email protected] A tvpe I error of [j < .05 was used to infer significant relationships among the variables.

Instrumentation Brief Symptom Inventory. The 6% is an abbreviated version of the SCL-90-R. The BSI measures nine symptom dimensions (somatization, obsessive-compulsive, interpersonal sensitivity, depression, anxiety. hostility, phobic anxiety. paranoid ideation. and psychoticism) and three global indices of distress (GSI, positive symptom distress index [PSDI]. positive symptom total [PST] ). Of these global indices. only the GSI was used. Subjects are asked to rate symptoms over the past week according to the perceived amount of psychosocial distress on a scale from 0 to 3, with 0 being “not at all” and 4 being “extremely.” BSI raw scores are transformed into standardized I scores with a mean of 50 and a standard deviation of IO. Cronbach’s (Yreliability coetticients for the BSI are reported to range from .7 I to $5 and validity coefficients to range from ..30 to .72, For this study’s sample, a reliability coefficient of .95 was obtained. The Coping With Disability Inventory. The CD1 is a self-rating. Likert-type of inventory and consists of 79 items divided into two subscales: a process subscale (CDI-P) containing process behavior items and an outcome subscale (CDI-0) containing outcome behavior items. representing social competence and quality of life issues. Item responses range from I (strongly disagree) to 5 (strongly agree). Reliability for the CDI, using Cronbach’s (Yis reported to vary from .78 to .8X with higher reliability for the CDI-0.“’ A concurrent validity of .53 is also reported between the CD1 and the California Personality Inventov.” To focus on coping as outcome behaviors, only items I to 42, which represent the outcome scale of the CDI, were used with this postpolio sample. Cronbach’s (Yreliability coefficient for the CDI-0 was found to be .84. Postpolio questionnaire. .4 54-page questionnaire was mailed to all research participants to complete at home. Arch Phys Med Rehabil Vol74, October 1993



Fig I-Brief Symptom Inventory mean scores for the overall sample, for distressed/depressed, and nondepressed polio survivors. Abbreviations: SOM, somatization; OC, obsessive compulsive; IS, interpersonal sensitivity; DEP, depression; HOS, hostility; ANX, anxiety; PHOB, phobia; PAR, paranoia; PSY, psychoticism; GSI, global severity index; CDI-0, Coping With Disability Inventory outcome. 0, depressed subjects; .,, total sample; f, other subjects.

The questionnaire had two parts: part 1 covered demographics and medical history and part 2 covered physical abilities, lifestyle, exercise. and psychosocial disability adjustment. This questionnaire asked subjects about their past and current health, functional abilities, and their lifestyle adjustment in order to identify physical and psychological secondary complications associated with polio. Results Psychological Distress and Depression. Figure 1 shows the mean BSI scores for the total sample, as well as for the depressed and nondepressed subjects using nonpatient norms as a comparison. BSI scores for the overall sample were in the average range, varying from 5 1.1 for hostility to 58.8 for somatization. In contrast to Conrady and colleagues’4 who found abnormal distress (~65 scores) in the areas of somatization, depression, and psychoticism, our group of 116 polio survivors demonstrated lower levels of distress and depression. All scale scores were lower than 65. When their individual scores were compared to BSI scores reported for able-bodied persons (fig l), only 15.8% of the sample had elevated BSI scores (~65) for depression. BSI scores for depressed/ distressed subjects (GSI and DEP > 65) were also elevated across other symptom areas with exception of phobic anxiety symptoms (6 1.7). This subgroup included a relatively equal number of men (47%) and women (53%). Men versus women. Figure 2 shows BSI mean scores for polio survivors by gender. There were no significant differences between the mean scale scores for the two groups. None of the scores were over 65. Somatization scores were the highest for both men (58.5) and women (58.9). Demographic and disability characteristics. Table 1 provides demographic and disability-related data for both Arch Phys Med Rehabil Vol74, October 1993



the overall sample and the two subgroups (by depression and distress). Significant differences between the two subgroups were observed with respect to subjects’ current living situation (x’ = 4.8 1, p < .05: more depressed/distressed subjects reported living alone). Differences were also noted with respect to seeking professional help (x2 = 10.1, p < .O1: more depressed/distressed subjects sought help); physician assessment of subjects’ deterioration in medical status and gait abnormality (depressed/distressed subjects were more likely to display heightened deterioration: x2 = 2.8, p < .O1: and gait abnormality: xz = 1 1.9, p < .O1). Table 2 summarizes subjects’ self-perception data. A majority of depressed/distressed subjects rated their health as poor or very poor (x’ = 24.7, p < .OO1) and they reported less satisfaction with their occupational status (x’ = 14.9, p < .OO1) and life satisfaction (x2 = 26.6, p < .OO1) than did other subjects. They also reported greater increase in pain since their physical best (x’ = 1 1.8. p < .O1) and were more likely to rate their current pain as excruciating when compared to nondepressed/nondistressed subjects (x’ = 3.9, p < .05). However, the two subgroups were not different with respect to self-perceived decrease in their ability to perform daily living activities. This difference only approached significance (p < .07). Coping behaviors. The CDI-0 mean score for the group was within the average range (mean = 150; range, 98 to 180; standard deviation = 16.8). These findings are similar to those of other studies using this same coping inventory. For example. Kulkarni and Blom” reported a CDI-0 mean score of 154 and standard deviation of 16.5 for a sample of 30 physically disabled persons displaying average coping skills. Depressed/distressed polio survivors obtained a significantly lower (1: 3.9 1: p < .OO1) CDL0 mean score (mean = 132.8; standard deviation = 20.8) than did nondepressed/nondistressed survivors (mean = 153.3: standard deviation = 13.9). Predictors of distress and depression. To predict dis-

5' t













Fig 2-Comparison of Brief Symptom Inventory mean score by gender. Abbreviations: SOM, somatization; OC, obsessive compulsive; IS, interpersonal sensitivity; DEP, depression; HOS, hostility; ANX, anxiety; PHOB, phobia; PAR, paranoia; PSY, psychoticism; GSI, global severity index; CDI-0, Coping With Disability Inventory outcome. Cl, males; f, females.



Table 1: Demographics and Medical Characteristics of Polio Survivors ______ Depressed/ Distressed Subjects (n ox 17)

Other Subjects (n = 99)


47. I 57.Y 46.X

41.J 5X.h 49.4

64.7 35.3

77.h 22.4

37.1 52.4

58.6 ‘Il.4

11.x 11.x 76.4

9.2 ‘7 7 _-. 6X.1

29.4 7O.h

12.‘) x7.1

5.Y 37. I 41.7 5.X

0.7 5X.X 2X.6 5.Y

Y. I 23.7 II.:!

15.1 76.5

7.7 23.2 21.5 17.1 17.4 17.2 31 .A

more Sought profcsslonal help Deterioration ,>I’medical status from best tl, non (“;))+ (ialt ahnormallt) (’ 1’ Normal tquivocal hlild or modcratt. ,ihnormalit> Sccercl> ahnormal



70.0 JO.0 26.7 11.3

2X.9 Il..? 50.6 h.2

IIiHerenccs hetwcen depressed/distre%cd ,115): ‘l/I .’ ,OI 1.

Il.2 13.1



and other subjects: *(,I

tress/depression among polio survivors. researchers developed a conceptual model incorporating these demographic, medical/health status variables and other psychosocial characteristics expected to be strongly associated with BSI scores of psychological distress and depression. The strongest correlations observed were between: the BSI/GSI and life satisfaction (.SO); between the BSl/GSI and CDI-0 (~ .45): between the BSI/DEP and the CDI-0 (~F.43); and between the BSI/GSI and increase in pain since time of subjects’ physical best (-.42). All these correlations were significant at 11 <: .OOI. The multivariate linear model, (table 3). includes the following variables among its predictors of distress and depression: life satisfaction (I, < .02); coping behaviors (p c: .13); increase in pain (p < .002): decrease in activity (p < .04): seeking professional help (p x .l5): living situation (p < .04): physical change (p < .l9): and number of co-morbidities (1, < .13). Combined, these variables accounted for 5 1% of’the variance in distress and depression among this group of polio survivors.

of Polio Survivors

Depressed/ Distressed Subjects

t kher Subjects (n 99)

Ovet23ll Sample (n 116)

49.0 5I.O

45.0 55.0

5X.X -11.1

\J I .o X. I

X7.1 11.CJ

25.0 25.0 12.5 37.5 I).()

5X.3 : I .7 I .7 .?.i 5.0

54.4 70.u ’ Y -.

f,J.? 35 3


X2.4 17.1


Sample (n = 116)


(I I 7 : 4 5 or


Table 2: Self-Assessments


SC\ /“I ) l-cm& 4ge (mean in !c:irs) hlarltal status (‘, I Married Not married t mployment wtu\ (“, ) M orking Uot worhing kducation (’ : 1 c High school High school -. High school I I\lng situation (II )* 24lone M ith others I yv of polio I’, 1 Rulhar Spinal Both hut sure Number of c’o-morhidit~v<


Self-assessed current health status* Very good Good/Fair Poor/very poolHealth compared to 5 to 10 years ago Much better/same Much or somewhat worse Current satisfaction with life+ Very satistied to fair Mildly to vrr! unsatisfied Occupa6onsl status s;ltisfktion$ Like it ver! much L.ihc it fairI> well Neither lihe or dislike Dislike somewhat Dislike It wry much Increase in pain since time of physical best’ I .css Great deal Pain rated at Its wvst* Mild/discomti~rtiny Distressing tkxkkiting Decreased abilit! to carr! out normal ADLs None-not much Fair-great deal


7.3 -I.-l

13.1 ” i --.. -11.2

I U.Y .?x h 11.5

Ah.5 i7.j

IT.7 x2 i

x1.2 0’J.X


All responses are percentages. Ditkences between depressed/distressed subjects and other subjects: *([I c .05): ‘( 11i .O I ): ? ,I ~ .Nl I ).





Contrary to other studies that suggest that most polio survivors tend to have abnormally elevated levels of depression and distress, this study found the prevalence of depression and distress to be 15.85? among polio survivors. Epidemiologic studies indicate that 15% to 30% of adults experience a clinically significant depressive episode sometime in their lives.‘j As a self-report measure of symptoms of distress and depression. all BSI mean scores, including Table 3: Predictors of Psychological Distress/Depression Polio Survivors

Variable Life satlsfactlon Coping behaciors Increase in pain .4ctivity Irvcl Living situation Professional help Change in medical status Co-morhiditics

Co&i&m 2.lh

I 3.41 -_I


‘.‘> 3.0(1 5.d’J .‘,1

R’: .51: standard error: 7.85; I, (11 I)

Standard b:rror



0 .:



,124 .002 ,030 ,032 .I51 .lUS I if1

I .o;

I .6(1 7.hh 1.07 A.31

.6I .OUII 1: t/l

7X ohser\atwns

Arch Phys Med Rehabil Vol74, October 1993




those for somatization, were within the average range. In general. these findings support the study’s main hypothesis, with the exception of somatization scores, which were not abnormally elevated, as previously hypothesized. The findings suggest that most physical symptoms experienced by polio survivors (ie, joint pain fatigue) are perhaps different from somatization symptoms measured by the BSI tie, nausea, dizziness. heart or chest pain). Abnormally elevated levels of physical symptoms appeared to be associated with higher levels of depression. In our study, depressed/distressed subjects differed from the remainder of the sample in several ways. From a perspective of physical health. they experienced more accentuated deterioration and greater increase in pain. They had a higher number of somatization complaints. They also exhibited poorer coping behaviors, were less satisfied with their lives and with their occupational status, and were more likely to be living alone. These findings were supported our hypotheses about demographic. behavioral, and disability characteristics ofdepressed/distressed polio survivors. Contrary to our assumption regarding gender. there were no differences in the prevalence of depression and distress among men and women. Similar to other health psychology studies,24-“r1these tindings suggest a significant relationship between poor health. pain and depression experienced by persons who report self-perceived loss of control over their health and lack appropriate coping behaviors. Depressed/distressed polio survivors reported poorer health and coping behaviors than did their nondepressed/nondistressed counterparts. Self-reported increase in pain was the most significant predictor of psychological distress and depression. according to the results from the multiple regression model. In this context, other less subjective medical variables (ie, changes in medical status, number of co-morbidities as reported by physicians) were less significant. Consistent with results from other studies.“9-30 subjective self-appraisals of poor health and of increased physical symptoms. tie, pain) among this group of polio survivors appear to be associated with the degree to which distress and depression were experienced. This perspective is further supported by the fact that depressed/distressed survivors also reported seeking more medical help for their polio-related problems. Though these findings support the association between depression/distress and subjects’ self-appraisal of poor health and increased physical symptoms, further clarification is warranted. Statements regarding attributions and causality are limited by the cross-sectional design nature of this study: we recommend further studies emphasizing a cohort design and larger sample sizes. Finally. additional research is needed to identify specific coping behaviors being successfully adopted by healthy nondepressed/nondistressed polio survivors. As a treatment mode. teaching polio survivors more effective coping behaviors may prevent abnormally elevated levels of depression and distress and thus promote healthier life styles. References RB. Spinal cord injuries: psychological, tional adjustment. New \‘ork: Pergamon. 1(9X0:35.

I. Trieschmann

Arch Phys Med Rehabil Vol74, October 1993

-. Frank RG. Llmlauf RL. Wonderlich




IO. I I. 12. 13. I-l.





I Y. 20.


15. 26.

27. 28. 19. 30.

social and voca-


SA. Askanazi GS. Bucklew S. Elliott T. Dilherences in coping styles among persons with spinal cord injury: a cluster analytic approach. J Consult Clin Psycho1 1987;55:7’7-3 I ‘Taylor SE. Health psychology: the science and the held. Am Psycho1 I YYO;45:4&5U. Tricschmann RB. .Aging with a dtsability. New York: Demos. 19x7:59. Frich NM. Bruno RI.. Post-polio sequelae: psychological and psychological overview. Rchabil Lit I YX6;46: 106-I I, Agre JC. Rodrigrcz AA. Tafel JA. Late effects of polio: critical review of the literature of neuromuscular function. Arch Phys Med Rehabil IYY 1:7’:92!3-31. Seligman M. Learned optimism: how to change your mind and your life. New York: Pocket Books. lYY0:72-5. Kaplan RM. Behavior as the central outcome in health care. Am Psychol lYYO:45:121 I-20. Bracken MB. Shepard MJ. Coping and adaptation following acute spinal cord injury: a theoretical analysis. Paraplegia 1980:18:74-85. Deegan MJ. Depression and physical rehabilitation. J Social Sot Welfare I Y77:3:945-53. Gunther MS. Psychiatric consultation in a rchahilitation hospital: a regression hypothesis. Compr Psychiatry I97 I : 12:572-X2. Wortman CB. Cohn Silver R. The mythsofcopingwith loss. J Consult Clin Psycho1 lYXY:57:349-57. Frick NM. Post-polio sequelae and the psychology of second disability. Orthopedics lYX5:8:85 l-3. Conrady LJ. Wish JR. 4gree JC. Rodrigrez .AA. Sperling tiB. Psychologic characteristics of polio survivors: a preliminary report. Arch Phvs Med Rehabil lYXY:70:458-63. Freidcnberg DL. Freeman D. Huber SJ. et al. Postpoliomyelitis syndrome: assessment of behavioral features. Neuropsych Neuropsychol Bchav Neural 19X92272-8 I. Cameron PMG. An investigation ofappraisal processes. coping strategics and depression in post-poliomyelitis patients. Proceedings from a conference on the Emotional and Psychological Impact of a Second Disability. Ontario March of Dimes. 1989: 135-43. Maynard FM. Julius M. Kirsch N. et al. The late elfects of polio: a model for identilication and assessment ofpreventable secondary disabilities. Final Project Report. Ann Arbor. Ml: Ltniversity of Michigan Medical Center, I99 I :4-E Derogatis LR. Spencer PM. BriefSymptom Inventory: (BSI) administration and procedures manual. Baltimore: John Hopkins University School of Medicine. 1982:10-X. Lazarus RS. Folkman S. Stress appraisal and coping. New York: Springer. lYX4: I 17-3’). tiulkarni MR. Coping with disability inventory: a study ofthe reliability and vsliditv ofan instrument designed to measure coping behavior ofphvsicnllv disabled persons in LJnited States and India. E. Lansing. ML: Michigan State Ilniversity. 19X5:69. Systat Manual (Version 3.0 I 1. 2nd cd. Evanston IL: Systat C’orporation. 19XY:6X?. Kulkarni MR. Blom G. Coping with disability inventory developed. The Interconnector lYX5;X:l l-5. Wells KB. Depression as a tracer condition for the national study of medical care outcomes. Santa Monica, CA: Rand Publications Series. July 1985. Jensen MP. Turner JA. Roman0 JM. Self-efficacy and outcome expectancies: relationship to chronic pain. coping strategies and adjustment. Pain I99 I :44263-9. Ferington FE. Personal control and coping effectiveness in spinal cord in.jury persons. Res Nurs Health 1986:9:257-65. Judd FK, Franzcp DJB. Burrows CD. Depression disease and disability: application to patients with traumatic spinal cord injury. Paraplegia 199 I29:Y I-6. Umlauf RL. Frank RG. Cluster analysis, depression. and ADL status. Rehabil Psycho1 1987:32:3X-44. Buckelew SP. Frank RG. Elliott TR. C’haney J. Hewett J. Adjustment to spinal cord iniurv: stage theorv revisited. Paraplegia 199 I : 125-30. Roman0 JM. Turner JATChronic pain and depressyon: does the evidence support a relationship? Psvchol Bull I985;97: I S-34. Beutler LF. Engle D. Ore’-Beutler ME. Daldrup R. Meredith K. Inability to express intense aff‘ect: a common link between depression and pain. J Consult Clin Psycho1 lY86;54:752-9.