411

ORIGINAL ARTICLE

A Longitudinal Study of Depression From 1 to 5 Years After Spinal Cord Injury
Jeanne M. Hoffman, PhD, Charles H. Bombardier, PhD, Daniel E. Graves, PhD, Claire Z. Kalpakjian, PhD, MS, James S. Krause, PhD
ABSTRACT. Hoffman JM, Bombardier CH, Graves DE, Kalpakjian CZ, Krause JS. A longitudinal study of depression from 1 to 5 years after spinal cord injury. Arch Phys Med Rehabil 2011;92:411-8. Objective: To describe rates of probable major depression and the development and improvement of depression and to test predictors of depression in a cohort of participants with spinal cord injury (SCI) assessed at 1 and 5 years after injury. Design: Longitudinal cohort study. Setting: SCI Model System. Participants: Participants (Nϭ1035) who completed 1- and 5-year postinjury follow-up interviews from 2000 to 2009. Interventions: Not applicable. Main Outcome Measure: Probable major depression, defined as Physician Health Questionnaire-9 score of 10 or higher. Results: Probable major depression was found in 21% of participants at year 1 and 18% at year 5. Similar numbers of participants had improvement (25%) or worsening (20%) of symptoms over time, with 8.7% depressed at both 1 and 5 years. Increased pain (odds ratio [OR], 1.10), worsening health status (OR, 1.39), and decreasing unsafe use of alcohol (vs no unsafe use of alcohol; OR, 2.95) are risk factors for the development of depression at 5 years. No predictors of improvement in depression were found. Conclusion: In this sample, probable major depression was found in 18% to 21% of participants 1 to 5 years after injury. To address this high prevalence, clinicians should use these risk factors and ongoing systematic screening to identify those at risk for depression. Worsening health problems and lack of effective depression treatment in participants with SCI may contribute to high rates of chronic or recurrent depression in this population. Key Words: Depression; Rehabilitation; Spinal cord injuries. © 2011 by the American Congress of Rehabilitation Medicine AJOR DEPRESSION is a prevalent and highly disabling M secondary condition associated with SCI. The point prevalence of major depression typically is estimated to be in the

From the Department of Rehabilitation Medicine, University of Washington, Seattle, WA (Hoffman, Bombardier); Department of Physical Medicine and Rehabilitation Medicine, Baylor College of Medicine, Houston, TX (Graves); Department of Physical Medicine and Rehabilitation Medicine, University of Michigan, Ann Arbor, MI (Kalpakjian); and College of Health Professions, Medical University of South Carolina, Charleston, SC (Krause). Supported by the Department of Education, National Institute on Disability and Rehabilitation Research, SCI Model Systems: University of Washington (grant no. H133N060033), Baylor College of Medicine (grant no. H133N060003), University of Michigan (grant no. H133N060032), and Shepherd Model System (grant no. H133N060009). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Correspondence to Jeanne Hoffman, PhD, Dept of Rehabilitation, University of Washington, Box 356490, Seattle, WA 98195-6490, e-mail: jeanneh@uw.edu. Reprints are not available from the author. 0003-9993/11/9203-00001$36.00/0 doi:10.1016/j.apmr.2010.10.036

range of 15% to 23%.1 This is substantially higher than the point prevalence of major depression in primary care settings (10%)2 or the 1-year prevalence of major depression in the U.S. population (6.6%).3 Depressive symptoms are linked to a host of negative outcomes, including pressure ulcers and urinary tract infections,4 lower self-appraised health,5 fewer leisure activities,6 poor community mobility, poor social integration, and fewer meaningful social pursuits.7,8 Depressive symptoms even predict mortality after controlling for other injury severity– and health-related variables.9 Although much is known about the point prevalence and correlates of major depression after SCI, far less is known about the longitudinal course of major depression in people with SCI. Most longitudinal studies of depression reported total group mean values or other descriptive data that did not show the clinical course in the subgroup that was depressed.10-14 Of the longitudinal studies that provided data for the clinical course of people who have become depressed, several emphasized a high rate of recovery from depression after SCI, but focused primarily on adjustment to injury within the first year after injury. Judd et al15 reported that 38% of people with SCI scored higher than 14 on the BDI at least once during inpatient rehabilitation, whereas 18% (47% of the depressed group) recovered before discharge from inpatient rehabilitation. Kishi et al16 reported that 21.7% of 60 inpatients with SCI met DSM-III diagnostic criteria for major depression while hospitalized, but half of these acute-onset depressive episodes had remitted by 3 months. In a study based on International Classification of Diseases–9th Revision–Clinical Modifications codes from a population-based administrative data set in Alberta, Canada, Dryden et al17 reported that depression was diagnosed by a physician in 34 of the 201 patients studied (17%) during the initial hospitalization. Of these, 50% discontinued depression treatment with the physician by discharge, implying that they had recovered. Others found more lasting rates of depression. For example, Craig18 and Hancock19 and colleagues reported that 30% of patients with SCI were significantly distressed (depressed and

List of Abbreviations AIS BDI CHART DSM MDD PHQ-9 SCI American Spinal Injury Association Impairment Scale Beck Depression Inventory Craig Handicap Assessment and Reporting Technique Diagnostic and Statistical Manual for Mental Disorders major depressive disorder Physician Health Questionnaire-9 spinal cord injury

Arch Phys Med Rehabil Vol 92, March 2011

and reliance on administrative data. Pϭ. moderately severe.33 Each of the 9 depression items reflects the persistence of the symptom during the past 2 weeks: 0 (not at all). and 20 or higher. for which the prevalence of depression is 2. sports. and 160 people completed year 5 but not year 1. This longitudinal study follows up people over time. we predicted that worsening in each of these areas would predict the development of depression from 1 to 5 years after SCI.36 Subjective health status. If depression after SCI occurs in a larger segment of the population who move in and out of depressive episodes. Subjects were categorized into 2 broad levels of injury. Cause of injury was divided into 5 categories: motor vehicle. Craig et al20 reported that 10 people who scored higher than 14 on the BDI during acute rehabilitation and did not receive an early cognitive-behavioral intervention continued to report moderate to severe depression through 12 months postinjury.1 We based depression severity categories on the ranges proposed in the original PHQ-9 validity study: 0 to 4 indicates minimal. resulting in a total score of 0 to 27. we predicted that improvement in pain.27.7%. mental health care should focus on early identification of at-risk persons and lifelong treatment focused on that subgroup. Kemp et al21 described 15 persons with major depression who refused a treatment trial. Research on the course of depression after SCI is conflicting and incomplete.35 Additionally. the literature is limited in terms of the overall number of studies. violence. For the same reasons.29. small sample sizes. which included the PHQ-9. universal depression screening should occur as part of both acute and postacute care. All sites contribute to a uniform national data set with the goal of examining the course of recovery and outcomes of persons Arch Phys Med Rehabil Vol 92. and discharge was in the geographic catchment area of the system. depression severity scores were unchanged during a 24-week follow-up period. but less likely to be white (66%.28 community integration. were more likely to be men (81%. 34. A single item from the 36-Item Short Form Health Survey was used to assess subjective health . and there are even more sparse data for the course of depression beyond 1 year after injury.67Ϯ5. moderate. whereas those with scores less than 10 are considered nondepressed. Measures Demographic and injury-related characteristics were obtained from interviews. These people were 5 to 37 years post-SCI and an unspecified number had chronic depression. but had depression scores similar to those in the present sample. It is important to improve our understanding of the course of depression after SCI because if depression after SCI is confined to people who become depressed soon after injury and are chronically depressed. Injury characteristics. physical examinations. and data were collected during the 1. Those who completed year 5 but not year 1 were similar in age and sex. Department of Education. Because we were interested in examining change over time. Moreover. or pedestrian. and hospital records near the time of the person’s initial rehabilitation hospitalization. we used the SCI Model System longitudinal data set to examine the course of depression from 1 to 5 years after SCI. 2 (more than half the days). March 2011 after SCI. injury-related. and research protocols were approved by each model system’s local institutional review board.001) and had higher depression scores at year 5 than those in the present sample (mean Ϯ SD score.77. Hoffman anxious) during their initial hospital stay and remained distressed through 1 to 2 years after SCI. Two treatment studies also emphasized that depression after SCI tends not to improve. tetraplegia versus paraplegia. METHODS Participants Participants in the present analysis are from 16 SCI Model System centers located throughout the United States.30 and subjective health status1.73). completed inpatient rehabilitation.24-26 substance abuse.22 we hypothesized that a higher proportion of people with SCI would be depressed at both 1 and 5 years compared with the general population. We chose this cutoff because for medical patients. National Institute on Disability and Rehabilitation Research. and were less likely to be white (67%. (1) What percentage of people with probable major depression at 1 year after SCI are recovered by 5 years postinjury? (2) What percentage of those with probable major depression at 1 year have evidence of chronic or recurrent depression at 5 years? (3) What percentage of people who are nondepressed at 1 year after SCI have probable major depression at year 5? (4) What percentage of people with SCI do not have probable major depression at either time? (5) What demographic.031). On an a priori basis. Probable MDD.002).S. those with probable MDD at year 1 were considered remitted if they scored in the minimal range (0 – 4) at year 5. Our research questions were as follows. or other clinical variables predict clinically significant depression improvement or development of depression between 1 and 5 years postinjury? Based on prior research suggesting that depression in the postacute phase may be largely chronic20. severe. 5. PϽ. 10 to 14. or 3 (nearly every day). Based on widely used clinical guidelines. The AIS32 was used to classify injury severity into complete (AIS grade A) versus incomplete (AIS grades B–E). sustained a traumatic SCI. PϽ.21 and recent evidence that depression is undertreated. Those who completed year 1 but not year 5 were significantly younger (mean Ϯ SD age.14 Participants who score 10 or more on the PHQ-9 are classified as having probable MDD. were admitted to each model system within 1 year of injury. we defined clinically significant improvement as a decrease of 5 or more points on PHQ-9 score from year 1 to year 5. Because there is a paucity of data about the course of depression beyond the first year after SCI.23 We also examined potential risk and protective factors. funded by the U. mild.and 5-year follow-up interviews between 2000 and 2009. 15 to 19. The study sample consisted of 1035 persons with traumatic SCI who completed both 1. All participants provided informed consent for data collection. PϽ. Item scores can be summed. The SCI Model System program.69 – . Prior research has shown that the PHQ-9 conforms to a unidimensional scale34 or is characterized by 2 factors that are highly correlated (rϭ.and 5-year follow-up assessments.7y.2 The PHQ-9 is based on the 9 symptom criteria of DSM-IV MDD. In this untreated group.001). We used the PHQ-9 to identify cases with probable MDD and measure depression severity.1Ϯ14.35 Accordingly. this cutoff score resulted in a depression prevalence rate consistent with results derived from SCI studies using structured diagnostic interviews. has been in existence for more than 30 years. a cutoff of 10 or higher yielded the best combination of sensitivity (88%) and specificity (88%) compared with an independent DSM-IV diagnosis of major depression made by a mental health professional.412 SPINAL CORD INJURY AND DEPRESSION.31 would predict clinically significant improvement in depressive symptoms between 1 and 5 years after SCI. fall. only people who completed interviews at years 1 and 5 were included in the present analysis.17. 1 (several days). in a large SCI survey study. 5 to 9. An additional 612 people completed year 1 but not year 5. Participants were 17 years or older. Pϭ.001).

8 (Nϭ1004) 23. and decrease in unsafe alcohol use were risk factors for the development of depression. Table 1: Demographics Variable At Enrollment Age (y) Sex: men Race White African American Other Ethnicity: non-Hispanic Cause Motor vehicle Fall Violence Sports Pedestrian Injury severity Paraplegia incomplete Paraplegia complete Tetraplegia incomplete Tetraplegia complete Minimal deficit Education: Նhigh school Married 37.8%) met criteria for probable MDD at 5 years. In addition. correlations among variables of interest used in the logistic regression were less than . 1. Total N when less than 1035 is noted.0 0. 124 (57. Additional tests were conducted to determine whether site (nesting) had an influence or variables predicted to affect depression correlated highly (tests of intraclass correlations for both year 1 and year 5 PHQ scores were .9%) scored in the nondepressed range. and 3. RESULTS Demographic data for the 1035 participants are listed in table 1. suggesting no relationship by site). and 71 (6. unsafe alcohol use at both times. community integration. Community integration. However. Change in pain was calculated by subtracting year 1 ratings from year 5 ratings.40 Change in unsafe alcohol use was coded as follows: 0 indicates no unsafe alcohol use at either time. Most (Nϭ724) participants (70.1% at year 5.5 77. Results are listed in table 3. In addition. we predicted the development of MDD by year 5 in the group that was nondepressed at year 1.2 (Nϭ1005) 40. occupation.1Ϯ14.2 35.1%) experienced a clinically significant improvement in depression severity. The second logistic regression model to determine possible protective or risk factors for developing depression between year 1 and year 5 is listed in table 4. PϽ.SPINAL CORD INJURY AND DEPRESSION. Participants on average were aged 37 years. a participant who has decreased unsafe alcohol use is almost 3 times more likely to become depressed than someone who has never engaged in unsafe alcohol use. social integration. reducing concerns for multicollinearity.6% at year 1 and 18. community integration. and 77.8 74. Unsafe alcohol use. sex.8 (Nϭ1033) 54. Most injuries occurred in motor vehicle collisions (54.2 18. which was .95 for the subscales) and validity established through correlations with independent observers (. health status). with higher scores indicating increasing pain levels.15.5%). Rates of probable MDD did not vary significantly based on sex at year 1.7 91. at year 5.9 18. cause and level of impairment). Two logistic regression models were used to examine the relationship between the change in our a priori predictor variables (pain.5% were men. The 19-item CHART Short Form was used to measure this domain. 2. In the first regression equation. and frequency of binge drinking (Ն5 drinks/occasion) in the past month.2% were white. Predictors of Clinically Significant Improvement in Depression or Development of Depression The first logistic regression predicting clinically significant improvement in depression from year 1 to year 5 showed no significant predictors from those chosen a priori (improvement in pain. The original CHART has high test-retest correlations (.93 for the total CHART. except for the correlation between change in mobility and change in occupation. declining health status. at 5 years. Change in health status was calculated by subtracting year 1 ratings from year 5 ratings. physical integration). participants were classified as having unsafe alcohol use if they were women and consumed more than 7 drinks a week. Participants were asked about the number of occasions of drinking per week during the past month. March 2011 . unsafe alcohol use at year 1 and no unsafe alcohol use at year 5. Our measure of substance abuse was unsafe alcohol use.00. For example.29.83). Of those with probable MDD at year 1.0%) scored in the nondepressed range at both times.7 1.2 81. we predicted clinically significant improvement in depression severity (decrease Ն5 points) at year 5 in the group that had probable MDD at year 1. Hoffman 413 status. Of those not depressed at 1 year.9%) were in remission from probable MDD.8%) or falls (23.39 Change in each domain was calculated by subtracting year 1 scores from year 5 scores.05). with higher scores indicating greater participation.37 Participants were asked to indicate their current health status from excellent (1) to poor (5) on a 5-point scale. 97 (11. or reported any episodes of binge drinking during the past month. race. Table 2 lists additional detail about how depression severity changed over time by using the 5 ranges of PHQ score severity. health status. unsafe alcohol use) and outcomes after controlling for demographic variables (age. .7%) reported probable MDD on both occasions (chronically depressed). alcohol use. Participants rated their level of pain from 0 (no pain) to 10 (pain so severe you could not stand it). 74.4 (Nϭ1034) NOTE. Statistical Analyses Descriptive statistics provide information about the population.38. no unsafe alcohol use at year 1 and unsafe alcohol use at year 5. Higher scores correspond to greater participation in each domain (mobility. Results from the second model suggest that increasing pain. With the second regression equation. Prevalence and Natural History of Probable Major Depression Percentages of the sample with probable MDD were 20. such that a higher score would indicate worsening health status. Values expressed as mean Ϯ SD or % of total.8 28. whereas 90 (8. average number of drinks consumed per occasion.5 11.1 (Nϭ1012) 17.0 9. African AmerArch Phys Med Rehabil Vol 92. were men and consumed more than 14 drinks a week. being African American was a protective factor compared with being white for the development of depression.1 4. the rate of probable MDD was significantly greater in women than men (23% of women vs 16% of men. 135 (63.80 –. We used a numerical rating scale to measure current pain level. Based on the National Institute on Alcohol Abuse and Alcoholism standards for safe drinking. Pain.

046 0.9) 12 (1. and almost 12% of those not depressed at year 1 were depressed by year 5.523 0.03–2.2) 11 (1.077 0. Epidemiologic data indicate that only 2.0) 57 (5. More than half of those depressed at year 1 were not depressed at year 5.001 Ϫ.8) 12 (1. Beginning with the most positive findings.8) 7 (0.311 0. pain after SCI is common and may influence depression onset and treatment. All change scores calculated as year 5 level Ϫ year 1 level.05 0.189 0.823 .43 comorbid traumatic brain injury.922 1.974 .44 bereavement-like reactions.6) 87 (8.6) 4 (0.520 0.660 0.003 0.15-17 Next.111 0.209 0.998 1.08–13. March 2011 .009 0.006 0.009 1.4) 28 (2.02 0.2) 23 (2.38–1.386 0.281 1.530 1.48 The lack of well-controlled efficacy studies of treatment of major depression in people with SCI also may contribute to inadequate treatment and more chronic depression.005 0. unsafe alcohol use at year 1 and no unsafe alcohol use at year 5.082 0.2) 1035 (100) NOTE.002 0. The finding that 12% developed probable MDD between 1 and 5 years suggests that ongoing screening for depression is needed in postacute and long-term care settings.22 Potential side effects (eg.646 0. † No unsafe alcohol use at either time (0) was compared with 1.32 0.069 0.808 3.268 0.186 Ϫ0.7 times less likely than whites to become depressed at either 1 or 5 years postinjury.007 0.32–2.99–1.160 Ϫ1.7%) may have either chronic or recurrent depression from 1 to 5 years postinjury. Characteristics such as medical comorbidities. Somewhat more participants became nondepressed (nϭ124) than became depressed (nϭ97) from 1 to 5 years.5) 5 (0.581 0.02 NOTE. there was considerable change in depression status over time.86 0.7% of the general adult population report chronic depressive symptoms. This was reflected in the small overall decrease in depression prevalence (21% to 18%) over time.13 0.98 0.190 0.157 0.273 0. sexual dysfunction or increased spasticity41) and poor tolerability may make physicians or patients with SCI reluctant to use antidepressants.47 and poor social support47 are associated with poor response to depression treatment and are common in people with SCI.9) 4 (0.4) 2 (0. 2. Hoffman Table 2: Depression Severity at 1 and 5 Years After SCI Year 5 PHQ-9 Score Ranges Year 1 PHQ-9 Score Ranges 0–4 5–9 10–14 15–19 20–27 Totals 0–4 5–9 10–14 15–19 20–27 Totals 462 (44.6) 115 (11.1) 200 (19. Other longitudinal studies also have emphasized that most people with SCI are characterized by resilience.414 SPINAL CORD INJURY AND DEPRESSION.748 0.582 1.38–3.3) 17 (1.2) 610 (58. icans were 2.02–9.1) 45 (4.810 0.3) 19 (1.4) 60 (5.4) 124 (12.72 1.3) 9 (0.42 Another possibility is that people with SCI may be less responsive to standard treatment.73 1.23 High rates of chronic or recurrent depression in people with SCI may be attributable to several factors. not depression.003 Ϫ0.2) 6 (0.01 0.7) 16 (1.7) 648 (62. these data show that 70% of the sample scored in the nondepressed range at both 1 and 5 years after SCI.641 0.61–1.002 1.062 0.831 0.99–1.982 0.338 Ϫ0.81–1. These longitudinal data show for the first time that a significant minority of participants with SCI (8. Studies of antidepressant efficacy and tolerability are needed in persons with SCI. no unsafe alcohol use at year 1 and unsafe alcohol use at year 5.232 0.38 0.213 1.5) 3 (0.720 0.99–1.001 0.8) 30 (2.592 Ϫ0.6) 8 (0.03 0. and 3. DISCUSSION This study provides a detailed examination of depressive symptoms at 1 and 5 years after SCI and potential risk and protective factors for the development of depression.46 low activities of daily living status.895 1.99–1.9) 211 (20.97–1.01 0.590 0.065 0.45 low socioeconomic status. Major depression appears to be both underrecognized and undertreated in people with SCI.239 0. Values expressed as N (% of total).61–4.3) 39 (3.800 1.5) 33 (3.9) 117 (11.830 0. Arch Phys Med Rehabil Vol 92.211 0.008 0.209 0.4) 47 (4.5) 9 (0.002 0. *Measured by using the CHART Short Form (higher scores indicate better integration).238 0.037 0. unsafe alcohol use at both times. as are studies of Table 3: Logistic Regression: Prediction of Clinically Significant Improvement in Depression From Year 1 to 5 Variable ␤ SE Significance Odds Ratio 95% Confidence Interval Age at injury Sex Race African American (vs white) All other races (vs white) Hispanic (yes vs no) Cause of injury Level and severity of SCI Paraplegia complete (vs paraplegia incomplete) Tetraplegia incomplete (vs paraplegia incomplete) Tetraplegia complete (vs paraplegia incomplete) Change in health status (higher scores indicate declining health) Change in pain (higher scores indicate higher pain) Change in physical independence* Change in mobility* Change in occupation* Change in social integration* Change in unsafe alcohol use† Constant 0.699 3.48 0.8) 25 (2.017 0. although prevalences of depression at years 1 and 5 were similar. Finally.

125 0. However.534 0. will have a positive impact on depressive symptoms in the postacute setting.426 0.288 .456 0.076 1.50 0. *Significant relationships.01–1.006 1. psychotherapeutic interventions and combined interventions.369* 2. Surprisingly.996 0. The PHQ-9 has good sensitivity and specificity compared with diagnostic interviews in primary care patients. In contrast.28–6. alcohol abuse may represent a means of coping with the stresses of SCI.14 0.98–1. March 2011 .946* 1.43) of probable major depression compared with white men.006 0. These results are consistent with other research that documented the presence of later developing depression after SCI17 and the association between pain and depression after SCI.66–3.21 who showed that without specific evidence-based depression treatment.385 .99–1.01 0. These results highlight the need for depression treatment and coping skills training as elements of any plan to treat alcohol abuse in the aftermath of SCI.97 1. such as pain. Cessation of unhealthy alcohol use also may unmask the presence of a depressive disorder.386* 1.469 0.004 Ϫ0.044 0.157 1.00 0.042 0.327 0. In contrast.82–2. 1.995 0.010 0. Giving up alcohol abuse as a habitual coping strategy does not mean that the person has learned alternative strategies to cope with the stresses of loss or disability.226 .011* .04–2.54 0.20* 0.007 1.369 Ϫ1.001 Ϫ0.150 . decreasing health status.031* .040* .007 0.77* 1. cessation of alcohol use may be associated with switching to more harmful drug abuse or loss of social support (“drinking buddies” or social outlets being alcohol establishments).49 Multimodal treatment may be more effective in cases characterized by chronic depression.433 1.004 0. a large cross-sectional study of people almost 10 years post-SCI showed that African American men were at significantly greater risk (odds ratio.597 0. All change scores calculated as year 5 level – year 1 level.005 Ϫ0.769 0. We do not have data for treatment use and therefore are unable to determine the extent to which treatment received in the interim period contributed to depression improvement. Alternatively. we have no data to verify these potential explanations.675 .032* . these data suggest that probable MDD is a more independent condition.99–1.260 Ϫ2. Study Limitations Important study limitations should be mentioned as caveats to the interpretation of these data and guides for future research.488 0. A cutoff score of 10 or higher was used to indicate probable MDD because this cutoff has the most validity data to support it.080 0. This interpretation is consistent with the findings of Kemp et al. Probable MDD was assessed by using a depression screening measure.98–1.457 .499 . and poor community integration.000 1.35 However. Hoffman Table 4: Logistic Regression: Prediction of Becoming Depressed at Year 5 if Not Depressed at Year 1 Variable 415 ␤ SE Significance Odds Ratio 95% Confidence Interval Age at injury Sex Race* African American (vs white)* All other races (vs white)* Hispanic (yes vs no) Cause of injury Level and severity of injury Change in health status (higher score indicates worsening health)* Change in pain (higher scores indicate higher pain)* Change in physical independence† Change in mobility† Change in occupation† Change in social integration† Change in unsafe alcohol use* Reducing unsafe use (vs no unsafe use)* Beginning unsafe use (vs no unsafe use)* Continued unsafe use (vs no unsafe use)* Constant 0. people with SCI and MDD may remain depressed over an extended period.288 0.485 0.427 .18 † NOTE.09–1. we found that increasing pain.79* 0.624 0.009* . especially given that the study sample had lower rates of minority representation than the group that did not complete 1 of the 2 follow-up interviews.004 0.62–3. although the relationship was incompletely mediated by lower education and income in the minority sample.03 0.50 Contrary to our hypotheses. and decreasing unsafe alcohol use were risk factors for the development of depression at year 5 in those not depressed at year 1.523 .292 .35 and depression prevalence estimates derived from this cutoff more closely approximate the estimated prevalence of MDD from diagnostic studies. the PHQ-9 lacks rigorous validity testing in people with SCI and a variety of alternative scoring methods have been suggested.99–1. one could speculate that a decrease in depression symptoms over time may reflect resolution of grief.000 0.51 More research is needed about the potential relationship of race or ethnicity to depression incidence and remission in longitudinal studies. Research is needed to define grief processes in people with SCI and disentangle grief from depression.994 2. we were unable to predict who would have clinically significant improvement in depression over time. Measured by using the CHART Short Form (higher scores indicate better integration).76–6.96* 0. This may mean that probable MDD requires specific treatments that target depression directly. Although this may seem paradoxical. In addition.446 0.284 0.259 .SPINAL CORD INJURY AND DEPRESSION.006 0.099* 1.01 0.033* .000 Ϫ0. Although we have tended to think that treating associated conditions. a decrease in unsafe alcohol use also may represent a risk for worsening depression. We found that African American race was a protective factor for the development of depression at 5 years post-SCI. poor health.00 1.200 .947 .14–0.48 One clinical implication is that health care providers should be attentive to the possibility that worsening pain or health may be associated with the development of probable MDD in the postacute setting.094 0. because depression and grief symptoms also may overlap.395 Ϫ1.1 More research is needed to validate depression screening measures in people with SCI. not a structured diagnostic interview.52 If Arch Phys Med Rehabil Vol 92.

53 Including these symptoms may spuriously inflate rates of depression. there are people who develop major depression by 5 years despite being nondepressed at 1 year. Krause et al53. Long-term adjustment to physical disability: the role of social support. Bombardier CH. particularly as we attempt to generalize beyond the SCI Model System. amphetamines) in people with SCI.55. or treatment-resistant depression after SCI and evaluate methods for treating this most concerning subgroup. opioids) and drugs of abuse (cocaine. Cameron MG. Krause J. 8.282:1737-44. In addition. Kessler R. 3. CONCLUSIONS Prior cross-sectional depression prevalence studies have not been able to elucidate the course of depression in people with SCI.16 The prevalence of late developing depression makes it incumbent on health care providers to continue screening for major depression as part of routine follow-up for at least the first 5 years after SCI and probably beyond.289:3095-105. The design also prevents us from distinguishing recurrent from chronic depression in those depressed at both times. depression treatment preferences. 4. Crow F. We chose to include somatic symptoms because this inclusive approach is considered a reliable and valid means of diagnosing depression in people with a wide variety of comorbid physical illnesses. 2.416 SPINAL CORD INJURY AND DEPRESSION. Arch Phys Med Rehabil 2004. Hart KA. and ways to overcome those barriers.74: 255-60. analysis of the predictive validity of specific symptoms was beyond the scope of this report. JAMA 2003. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. All possible participants were not included because some people did not complete follow-up questionnaires and there were differences between those who completed the 2 follow-up interviews and those who completed only 1. efforts to replicate this study should use valid diagnostic interviews for MDD. . This study shows that the 18% of the sample with probable MDD at 5 years postinjury consists of half with chronic or recurrent depression and half with depression not present at 1 year post-SCI. Patient Health Questionnaire.56 However. References 1. insomnia. Symptoms of major depression in people with spinal cord injury: implications for screening. Tulsky D. such as fatigue. Shewchuck R. A variety of treatment-related research is needed to address the high rate of chronic or recurrent depression. et al. Health services research on the prevalence and efficacy of mental health treatment in persons with SCI is lacking. Future research should include information about mental health treatment use. 5. A prospective study of health and risk of mortality after spinal cord injury. Berglund P. Rehabil Psychol 1994. Future researchers should consider reassessing depression more frequently and using valid retrospective reporting methods.17 However. information for preinjury history of depression may have considerable prognostic value. although measurement of somatic items may be the most problematic during inpatient rehabilitation. There are limitations associated with the fact that assessments of probable MDD occurred approximately 4 years apart. JAMA 1999. Primary Care Evaluation of Mental Disorders. Tate DG. Richards JS. such information is not included in the SCI Model System database. More information about depression treatment under usual-care conditions can aid in the planning of more effective mental health services for people with SCI. treatment accessibility.55. Future research also should examine the possible depressive effects of commonly used medications (antispasm. Herrick S.68:339-43. Certain symptoms of depression might predict future depression. barriers to mental health treatment. This study does not address the validity of DSM-IV symptoms of MDD in people with SCI. Arch Phys Med Rehabil 2008. as well as new-onset depression affecting the lives of people with SCI. Fuhrer MJ. Social support and leisure activities following severe physical disability: testing the mediating effects of depression.54 Our approach and ability to predict depression were limited. Spitzer RL. 7. Therefore. worsening pain. Rintala DH. Carter R. Arch Phys Med Rehabil 1987. Therefore. Basic Appl Soc Psychol 1995. Major depressive episodes could have developed and fully remitted during the interim period. Characteristics of late developing depression also need to be elucidated. J Pers Soc Psychol 1985. and self-blame. future research should examine additional risk factors for chronic. The SCI Model Systems are not population based and therefore cannot be assumed to be representative of the full SCI population. may be attributable to physical conditions associated with SCI or to MDD. Young ME. The presence of both new-onset and chronic or recurrent depression suggests that ongoing systematic screening for depression is needed beyond the acute treatment phase. Finally. recurrent. Social support and the prediction of health complications among persons with spinal cord injuries.39:231-50. perceived control. 9. This study collected no data about the use of antidepressants or psychotherapy. interventions designed to maintain subjective health and prevent increases in pain may have the effect of preventing the onset of depressive symptoms. Pickelsimer E. Depressive symptomatology in persons with spinal cord injury who reside in the community. Williams JB. Kroenke K. Schulz R. Clearman R. Declining health. these data may underrepresent the total burden of probable MDD during the 4-year period. 6. March 2011 antiseizure. Although those with a prior history of major depression are at risk for later depression. Decker S.57 More research is needed on subgroups that develop depression soon after SCI versus in later years because depression in these different subgroups may have distinct causes and may vary in response to treatment. Demler O.89:1482-91.48:1162-72. Hoffman feasible. Elliott TR. Wilson D. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). MacDonald MR. Nielson WR.16:471-587. whereas excluding these symptoms may lead to underdiagnosis and undertreatment. and cessation of unhealthy alcohol use should alert clinicians to an increased risk for probable MDD. They found a distinctive somatic factor during inpatient rehabilitation53 that was not predictive of either somatic or nonsomatic symptoms at 17 or 29 months postinjury. all absolute estimates of the prevalence of depressive disorders must be interpreted with caution. Elliott TR. Arch Phys Med Rehabil 1993. Arch Phys Med Rehabil Vol 92. and poor appetite.85: 1749-56. Depression and activity patterns of spinal cord injured persons living in the community. Krause JS.56 have examined differences between somatic and nonsomatic symptoms on the PHQ-9 completed during inpatient rehabilitation and at approximately 17 and 29 months after injury. For example. Somatic symptoms. The impact of health maintenance and pain management on depression onset may be a fruitful area of research.

Dickson H. 38. Hancock K.46:57-67.31:349-57. Depression following spinal cord injury. Psychiatr Clin North Am 2007. 32.SPINAL CORD INJURY AND DEPRESSION. The Craig Handicap Assessment and Reporting Technique-Short Form 2000. Anxiety and depression after spinal cord injury: a longitudinal analysis. Dryden D. fatigue and depression in individ- 31. Chang E. Atlanta: ASIA. 47. Hoffman 417 10. Sareen J. Adkins R. 29. Pasternak R. Spinal cord injury rehabilitation evidence. Miller M. A prospective in-patient study. Pain and depression in acute traumatic spinal cord injury: origins of chronic problematic pain? Arch Phys Med Rehabil 1996. FAQs 2009. Smith B. Findley P. Br J Psychiatry 1989. 39.scireproject.80:339-42. Kennedy P. et al.154:668-71. editors.gov/FAQs/General-English/ default.85:292-4. Kemp B. 14. Reynolds CR. 25. Factor structure of the PHQ-9 screen for depression across time since injury among persons with spinal cord injury. 42. Szanto K. Saunders L.32:675-9. Williams J. coping strategies and post-traumatic psychological growth following spinal cord injury: a 10-year review. J Gen Intern Med 2001. Pollard C. 28. Prevalence of depression diagnoses and use of antidepressant medications by veterans with spinal cord injury. Predicting depression and psychological distress in persons with spinal cord injury based on indicators of handicap. Craig A. 49.nih. Accessed October 30. Mehta S. 4th ed. Charlifue S. Whiteneck G. Chang E. Physical activity is related to lower levels of pain. In: Eng J. Dijkers M.30:77-90. Accessed July 7. 22.76:525-32. Iosifescu D. 21.27:22-8. Rehabil Psychol 2009.33(Pt 2):221-30. In: Diagnostic and Statistical Manual of Mental Disorders. Arch Phys Med Rehabil 1986. Richards J. Changing psychosocial morbidity in people aging with spinal cord injury. Gilman S. Top Spinal Cord Injury Rehabil 1998. The Craig Handicap Assessment and Reporting Technique (CHART): metric properties and scoring. Rossi C. Williams R. Webber J. International Standard for Neurological Classification of Spinal Cord Injury. 16. Major depressive episode. Siddall P. Burrows G. Houck P.htm. J Rehabil Res Dev 2009. 34. A longitudinal analysis of emotional impact. Cairns D. Martin J. Co-occurring medical and mental illness and substance use disorders among veteran clinic users with spinal cord injury patients with complexities. p 320-7. Nicholson Perry K. Br J Health Psychol 2007. Kroenke K. 2008. Nicholas M.78:33-8. 13. Enns M.90:301-6. Spinal Cord 2009. Ullrich P.47:108-14. Accessed October 23. MacArthur initiative on depression in primary care 2010. Med Care 1993. Sambamoorthi U. Tawashy A. Tran Y. Adkins R. Fann J. Hart K. J Neuropsychiatry Clin Neurosci 1994. Spinal cord injury: a search for determinants of depression two years after the event. Mobayed M.47:789-95. Hancock K. Kraemer H. Tate D. Bosworth H. Krause J. The PHQ-9: validity of a brief depression severity measure. Richards J. Gerhart K. Craig A. 19. Mellick D.53:243-9. Rowe B.86:662-71. Long-term psychological outcomes in spinal cord injured persons: results of a controlled trial using cognitive behavior therapy. Am J Phys Med Rehabil 2007. Paraplegia 1993. 23. 19:15-23. 44. uals with spinal-cord injury: a correlational study. 15. Spitzer R. Anxiety and depression over the first year of spinal cord injury: a longitudinal study. Psychological characteristics of people with spinal cord injury-related persisting pain referred to a tertiary pain management center. George L. Smith B. Slivinski J. Social inequalities in response to antidepressant treatment in older adults. 2000. Forchheimer M.4:16-30. Bode R. Arch Phys Med Rehabil 1997. Treatment resistance of depression after head injury: a preliminary study of amitriptyline response. Wilson C. Rogers B.org/clinicians/ toolkits/materials/forms/phq9/treatment_response/. 1994. J Spinal Cord Med 2004. Kishi Y. Hays J. 30. Tang P.16:606-13. Tate D. Antidepressant exacerbation of spasticity. Psychologic adjustment to spinal cord injury during first postdischarge year. et al. A longitudinal investigation into anxiety and depression in the first 2 years following a spinal cord injury. 48. Dew M.12(Pt 3):347-62. Arch Phys Med Rehabil 2000. Reynolds CR. Psychosocial and clinical predictors of unipolar depression outcome in older adults.tbims. Weaver F. Prospective longitudinal study of depression following spinal cord injury. National Institute on Alcohol Abuse and Alcoholism. Miller W. Paraplegia 1994. Dickson H. Tate D. McHorney C. Middleton J. Richardson E.54: 198-203. com/rehabilitation-evidence. 35. Cox B. Arch Phys Med Rehabil 1995. Wainberg M. 24. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Dickson H. Manber R. Heinemann A. Washington (DC): American Psychiatric Association. 2010. Lin K. 41. 33. Kennedy P. Ware JJ. 27. 45. Acta Psychiatr Scand 1992. March 2011 . Temkin N. Rehabil Psychol 2004. Orenczuk S. Middleton J. Elliott T. Am J Psychiatry 1999. Craig A. Spinal Cord 2009. Scott M. 17. Available at: www. 2010. Cohen A. Depression following spinal cord injury.54:389-98. Wilson C.73:175-83. Treatment of major depression in individuals with spinal cord injury. 50. Hung C. Rintala D. mental health service use and adequacy of treatment. Eng J.niaaa.49: 134-9. 46. Satyanarayana S.6:237-44.156:202-8. Findley T. Stone J.31:247-63. Young M.17:238-46. Can J Psychiatry 2009. Marino RJ. Nava G. et al. Available at: http://www. Dinan T. American Psychiatric Association. Prevalence and correlates of chronic depression in the Canadian Community Health Survey: mental health and well-being. Raczek A. 18. Alcohol and marijuana use in a community-based sample of persons with spinal cord injury. Arch Phys Med Rehabil 1999. Rehabil Psychol 2008. Psychological morbidity and spinal cord injury: a systematic review. 2009. Arch Phys Med Rehabil 2011. Banerjea R. 12. Richards J. Judd F. Faster remission of chronic depression with combined psychotherapy and medicaArch Phys Med Rehabil Vol 92. Craig A. Neurorehabilitation 2004. Arnow B.depression-primarycare. 20.63:50-6. Improving measurement properties of the Patient Health Questionnaire-9 with rating scale analysis. Accessed October 14. Hancock K. 36. Craig A. 40. Hall K. et al. Depression following traumatic spinal cord injury. Am J Phys Med Rehabil 1994. Spinal Cord 2009. 43. et al. Br J Clin Psychol 1994. Bombardier C. Fann J. Kahan J.org/ combi/chartsf.67:362-5. Teasell R.77:329-35. Treatment of bereavement-related major depressive episodes in later life: a controlled study of acute and continuation treatment with nortriptyline and interpersonal psychotherapy. Stolp-Smith K. 37.81: 932-7. Teasell R. Available at: http://www. Robinson R. Arch Gen Psychiatry 2006.92:352-60. Available at: http://www. Treatment of depression following spinal cord injury: an evidence-based review. 26. 11. Dijkers M. Dickson H. Int J Geriatr Psychiatry 2002. Maynard F. Depression after spinal cord injury: comorbidities. Brown D. Fuhrer M. Hancock K. Treating depression in the medically ill. The MOS 36-Item Short-Form Health Survey (SF-36): II. Steffens D. Kennedy P. Neuroepidemiology 2005. Forrester A.25:55-61.

Coker J. Prediction of somatic and nonsomatic depressive symptoms between inpatient rehabilitation and follow-up. Arch Phys Med Rehabil 2000. March 2011 . Carter R. Pignone M. and socioeconomic indicators. Bombardier C. Krause J. Eaton W. Arch Gen Psychiatry 2008. Shao H. Nestadt G. Hoffman tion than with each therapy alone. Bombardier C. Cordes J. Factor structure and predictive validity of somatic and non-somatic symptoms from the PHQ-9: a longitudinal study after spinal cord injury.32:6-24. Spinal Cord 2010.418 SPINAL CORD INJURY AND DEPRESSION. Reed K. J Consult Clin Psychol 2008. McArdle J. Schomer K. Arch Phys Med Rehabil Vol 92. Kalpakjian C. aging. 55. Depression after spinal cord injury: relation to gender. Is this patient clinically depressed? JAMA 2002. Ramirez G. 57. Krause J. Assessment of depressive symptoms during inpatient rehabilitation for spinal cord injury: is there an underlying somatic factor when using the PHQ? Rehabil Psychol 2008:53:513-20.91:1218-24.76:239-44. McArdle J. Noël P. Arch Phys Med Rehabil 2010. Johnson K.287:1160-70. ethnicity. Krause J. Krause J. 56. 54. Williams JJ. Reed K. Brown P. 53. 52. J Spinal Cord Med 2009. Population-based study of first onset and chronicity in major depressive disorder. Kemp B.76:459-67.65:513-20.81:1099-109. Measuring depression in persons with spinal cord injury: a systematic review. 51. et al.