• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

A Multicenter Study of Depression Among Emergency Department Patients

Posted on: Thursday, 16 December 2004, 03:00 CST

Abstract

Objectives: The authors sought to determine the 12-month prevalence of depression among emergency department (ED) patients using a single-question screen. Methods: This cross-sectional study was conducted in four Boston-area EDs. For two 24-hour periods, consecutive patients aged 18 years or older were interviewed, excluding those who were severely ill, potential victims of sexual assault, or emotionally disturbed. During the interview, patients were asked "Have you had any of the following problems during the past 12 months?" Patients answered "yes" or "no" to a list of health problems that included depression. In a validation study, the authors found that this simple approach correlated well with results from the validated Center for Epidemiologic Studies Depression Scale. Results: Of 752 eligible patients, 539 (72%) were interviewed. Of these patients, 30% (95% confidence interval = 26% to 34%) reported depression within the past 12 months. Compared with their nondepressed counterparts, depressed patients were more likely middle-aged, female, and of lower socioeconomic status. Depressed patients were more likely to be smokers and to report a diagnosis of asthma or arthritis/rheumatism. In a multivariate analysis, factors that were independently associated with depression were lower level of education, smoking, and self-reported anxiety, chronic fatigue, and back problems. Conclusions: A 30% 12-month prevalence of depression among ED patients was found. Depressed patients had a distinct sociodemographic and health profile. In the future, awareness of risk factors for depression in the ED setting and use of simple screening instruments could aid in the recognition of depression, with subsequent referral to mental health services. Key words: depression; screening; emergency department; risk factors. ACADEMIC EMERGENCY MEDICINE 2004; 11:1284-1289.

Depressive disorders are common.1 3 The national 12-month prevalence of major depressive disorder is 6.6%, affecting approximately 13.7 million U.S. adults.1 In 2001, the World Health Organization identified depression as a leading cause of disability worldwide and emphasized the importance of effective management and treatment of depression in primary care settings.4 However, primary care and emergency physicians often fail to recognize and treat depression in their patients.5-8 This may affect the patient, with a potential for loss of productivity, functional decline, and increased mortality.9-11

To improve the accurate identification and subsequent treatment of depressed patients, the U.S. Preventive Services Task Force (USPSTF) recommends screening adults for depression in primary care settings that have systems in place to assure accurate diagnosis, effective treatment, and follow-up.12 Screening of patients whose profile suggests increased risk can be particularly effective.13 However, the prevalence and risk factor profile of depression has not been well defined in the emergency department (ED) setting. Previous studies have examined depression in ED patients with acute and chronic pain, obstructive pulmonary disease, alcohol abuse, suicidal ideation, and homelessness.14"20 Other studies focusing on geriatric ED patients have identified the prevalence of depression to be approximately 30% and associated with lower education, lower income, more medical conditions, decreased functional ability, and poorer overall health.8'21 These studies, however, focus on subgroups of the ED population. There is a need to study depression in the general ED population and assess the applicability of, and implementation strategies for, broad-based depression screening.

This multicenter study estimated the 12-month prevalence of depression in the ED using a singlequestion screen and described the demographic, socioeconomic, and health status profile of patients who reported having experienced depression during the past 12 months. We hypothesized that depression would be more common in the ED setting than national community-based estimates and that it would correlate significantly with lower income, lower education, and greater health impairment.

METHODS

Study Design. This was a multicenter, cross-sectional study. The institutional review boards of the participating hospitals approved the study, and verbal consent was obtained from all patients before enrollment.

Study Setting and Population. In each ED, consecutive patients were enrolled over two 24-hour periods (one Wednesday and one Saturday) during July 2002. Inclusion criteria were age 18 years or older and ability to give verbal informed consent. Exclusion criteria were previous enrollment in the study, patients exhibiting acute distress, potential victims of sexual assault, and cognitively disabled patients (e.g., psychosis). Patients with language barriers to participation were included whenever translation was available by study staff, ED staff, or interpreter. The study was conducted in four EDs in the Boston area: two university-affiliated urban hospitals, one community urban hospital, and one community suburban hospital. All sites are members of the Emergency Medicine Network (www.emnet-usa.org).

Study Protocol. Trained interviewers assessed patients' sociodemographic characteristics and 12-month prevalence of a range of health problems. The 12month prevalence of depression was assessed with a single-question screen. During the interview, patients were asked "Have you had any of the following problems during the past 12 months?" Patients answered "yes" or "no" to a list of health problems, including depression, back problems, allergies, chronic fatigue, headaches, insomnia, anxiety, digestive problems, sprain/muscle strain, chronic pain, addictive problems (e.g., alcoholism, substance abuse), skin problems, and lung problems. Patients who reported that they had experienced depression in the past 12 months were considered positive for likely depression. Patients were also asked if they had "ever been diagnosed" with a variety of medical conditions, including heart attack, congestive heart failure, stroke, high blood pressure, diabetes, cancer, chronic obstructive pulmonary disease, asthma, and arthritis or rheumatism. Median family income was estimated using patients' home ZIP codes.22

To examine if depression could be accurately recognized in the ED using a single-question screen, we conducted a comparative assessment of our singlequestion screen for the 12-month prevalence of depression and the well-validated Center for Epidemiologie Studies Depression Scale (CES-D).23 Consecutive patients aged 18 years or older were enrolled using the aforementioned exclusion criteria on one weekday in July 2003 in a single urban ED. After verbal consent was obtained, an interview was administered similar to the multicenter interview, including the singlequestion screen for depression and the CES-D.

This validation study demonstrated that our singlequestion screen of depression experienced during the past 12 months correlates with the CES-D screen for depression experienced in the past week. Therefore, in this study, we refer to patients who answered "yes" to having experienced depression during the past 12 months as "depressed" and those who answered "no" as "nondepressed."

Data Analysis. All analyses were performed using Stata 7.0 (Stata Corp., College Station, TX). Data are presented as proportions, means (with standard deviation), or medians (with interquartile range). Student's t-test, Kruskal-Wallis test, chi-square test, and Fisher's exact test were used as appropriate. A p-value of less than 0.05 was considered statistically significant. Variables associated with depression at p < 0.01 in univariate analysis were evaluated for inclusion in multivariate logistic regression models. The following variables were included in the initial multivariate logistic regression model: age, female gender, household income, less than high school education, all 12 health problems, asthma, and arthritis or rheumatism. Age and gender were included in the multivariate logistic regression models because of their well- validated relation with depression. A manual stepwise process was used for variable selection in the multivariate model. Factors were entered into the model and assessed as confounders based on a 10% change in the point estimate. Factors that were not considered to be confounders were removed from the model. All odds ratios (ORs) are presented with 95% confidence intervals (CIs).

RESULTS

Of 752 eligible patients, 539 patients (72%) were interviewed and 536 patients responded to the question about experiencing depression during the past 12 months. Among these ED patients, 160 (30%; 95% CI = 26% to 34%) reported depression within the past 12 months. Among women, 35% reported depression; among men, 25% reported depression. On univariate analysis, depressed patients differed from nondepressed patients with respect to age, gender, income, and education but not with respect to race/ethnicity or having a primary care provider (Table 1). Significantly more depressed patients than nondepressed patients experienced a variety of self-reported health problems (Table 2). In addition, significantly more depressed patients were current smokers as compared with nondepressed patients (39% vs. 20%, respectively, p < 0.001). Furthermore, d\epressed patients were significantly more likely to have been diagnosed with asthma or arthritis/rheumatism as compared with nondepressed patients (Table 2).

TABLE 1. Sociodemographic Characteristics of Depressed Versus Nondepressed ED Patients

After adjusting for age and gender, factors independently associated with depression included lower level of education, current smoking, anxiety, chronic fatigue, and back problems (Table 3). Because depression and anxiety were so strongly associated (OR, 9.26), we repeated the multivariate model excluding anxiety. Omission of anxiety yielded nearly identical results for several factors: lower level of education, current smoking, and chronic fatigue (data not shown). However, additional factors were associated with depression: asthma (OR, 1.71; 95% CI = 1.00 to 2.92; p = 0.05), insomnia (OR, 2.09; 95% CI = 1.31 to 3.34; p = 0.002), headaches (OR, 1.83; 95% CI = 1.16 to 2.88; p = 0.009), and chronic pain (OR, 1.66; 95% CI = 1.02 to 2.70; p = 0.04).

In the validation study of our single-question screen for the 12- month prevalence of depression and the CES-D, 94 successive patients were enrolled. Twentythree patients (25%) were clinically depressed as defined by the CES-D cutoff of 16. The absolute agreement between the single question and the CESD was 87% (K statistic = 0.65, p < 0.001). The sensitivity of the single question was 74%, specificity was 92%, positive predictive value was 74%, and negative predictive value was 92%. Of the 23 patients determined to be depressed according to the CES-D criteria, six responded "no" to the single- question screen, yielding a false-negative rate of 26%. Of the 71 patients not depressed by the CES-D criteria, six responded "yes" to the single-question screen, yielding a false-positive rate of 8%.

TABLE 2. Self-reported Health Problems and Medical Conditions of Depressed Versus Nondepressed ED Patients

DISCUSSION

The data show that a substantial proportion of patients in the ED screen positive for depression. The 12-month prevalence of depression as reported by ED patients was 30%, about five times greater than one national community-based estimate of 6.6% for major depressive disorder.1 We have not found similar studies measuring the overall prevalence of depression in the ED, but our results are in accordance with studies of geriatric depression in the ED that have reported prevalence rates of 27% and 33%.8'21 Our results not only demonstrate that the 12-month prevalence of depression is elevated among ED patients but also identify several factors associated with depression. The ability to identify a risk factor profile for depressed ED patients and to use a singlequestion screen for depression suggest that depression screening may be efficiently administered in the ED.

Our study, consistent with prior epidemiologic studies, indicates that depression is associated with female gender and lower socioeconomic status.24'25 The higher prevalence of depression among women compared with men is a widely documented finding in psychiatric epidemiology.24 In community-based studies throughout the world, the prevalence of major depression among women has typically been between 1.5 and 3 times higher than among men.24 Similarly, our study found the female/male risk ratio to be 3:2. Although the association is complex, lower income and education also have been found to be associated with depression.25 Our results are consistent with the study by Raccio-Robak et al. that found lower income and education level to be significantly associated with depressed geriatric ED patients.21

TABLE 3. Multivariate Logistic Regression Model of Depression among ED Patients

Depression has been linked to many medical conditions, including cardiac disease, stroke, hypertension, diabetes, cancer, chronic obstructive pulmonary disease, asthma, and arthritis or rheumatism. 14'26~29 In our study, however, depression was only significantly associated with diagnosed asthma and arthritis/rheumatism. Studies have shown that the relationship between asthma and depression is complex. Some studies show that depression contributes to increased asthmatic attacks and exacerbated respiratory symptoms but that asthmatic patients as a group are not more depressed.30'31 By contrast, our study suggests that ED patients with asthma, compared with those without asthma, are more likely to have depression and that anxiety may in part mediate this association. The asthma/ depression association may indicate that the portion of the total asthmatic population presenting to the ED is more likely to experience severe or poorly managed asthma, which has been correlated with depression.32 In addition, other studies have shown that depression is more common in patients with rheumatoid arthritis than in healthy patients.33 The reason why our data did not demonstrate that depression was significantly correlated with other medical conditions may have to do with onset times of depression. A person who experienced their first stroke or heart attack may not feel depressed immediately or have felt depressed in the previous 12 months. A study by Robinson et al. found that stroke patients can develop depression during hospitalization or many months after discharge.34

Depressed patients reported more than twice as many health problems than their nondepressed peers. Many studies, like the Medical Outcomes Study, have shown that patients with depressive symptoms or diagnosed depression report low perceived health, increased pain, and more disability in terms of days in bed.10'15'21 Our multivariate analysis indicates that many health problems, often related to perceived pain, are associated with depression, including anxiety, chronic fatigue, back problems, insomnia, headaches, and chronic pain. Our study, congruent with previous studies, also demonstrates the comorbidity of anxiety and depression.1 A patient who reports anxiety within the past 12 months is almost ten times as likely to report depression during that time period.

Our study demonstrates that a single-question screen, although not a perfectly accurate replacement for the CES-D, can serve as an efficient initial screen for depression in ED patients where high rates of depression prevalence help justify screening efforts. The discrepancy between our single-question screen and the CES-D appears to be a result of the difference in the time scale for the measures of depression. While our single question screens for depression in the past year, the CES-D screens for depression within the past week. The false-positive rate (8%) found when comparing our single- question screen with the CES-D can be explained by the fact that patients who reported "depression within the past 12 months" had subsequently been treated with antidepressants and no longer presented depressive symptoms according to the CES-D (of the six patients with false positives, five had prescriptions for antidepressants and three had taken an antidepressant within the past week). The false-negative rate (26%) may be a result of the fact that ED patients felt situational depression within the past week because of their immediate sickness. For example, the CES-D, although designed to be relatively free of physical or vegetative signs of depression, nevertheless retains several items such as loss of appetite and restless sleep that may more accurately reflect recent illness-related symptoms than depression per se.

In its 2002 recommendations for depression screening in primary care settings, the USPSTF acknowledged that many formal screening tools are available, such as the CES-D. They conclude, however, that asking two simple questions about mood and anhedonia (e.g., "Over the past two weeks, have you felt down, depressed, or hopeless?" and "Over the past two weeks, have you felt little interest or pleasure in doing things?") may be as effective as using longer instruments.35 Considering the time-limited environment of the ED, the use of only one or two simple questions would greatly increase the feasibility of depression screening in the ED.

LIMITATIONS

There are potential limitations to this study. First, the presence of self-reported depression in patients was not confirmed using a clinical DSM-IV diagnosis. Although our single-question screen was correlated with the CES-D, a well-validated screening tool for depressive symptoms, neither screen can be equated with a psychiatric evaluation for major depressive disorder. The CES-D was originally developed as an instrument for psychiatric epidemiology research, not as a clinical diagnostic tool.23 Self-rated depression scales themselves have important limitations. The use of a single dichotomous cutoff may not reliably distinguish between depressive symptoms alone and major depressive disorder. However, depressive symptoms alone, even in the absence of major depressive disorder, are a significant clinical and public health problem,36 and a positive finding suggests the need for further medical and psychiatric evaluation.37 Also, these scales may not be generalizable to all populations. Particularly in the ED setting, where many patients have undergone traumatic and acute illness experiences, distinguishing between complicated grief, traumatic grief, depression, and anxiety can be challenging.38'39 In addition, not all patients who were eligible for the study were enrolled (72% enrollment). Patients exhibiting acute distress, potential victims of sexual assault, and cognitively disabled patients (e.g., psychosis) also were not included in the current study. The exclusion of these patients would likely lead to an underestimation of the prevalence of depression in an ED population.

Furthermore, our single-question screen was validated at one of the four ED sites, and it cannot be assumed that the ? obtained at one site is generalizable to all four sites. Finally, we note that patients are from one metro\politan area (Boston) and may not be representative of ED patients throughout the United States. Additional EMNet studies are planned to examine this issue. Despite these potential limitations, our study shows clear differences in sociodemographic factors and health status between depressed and nondepressed ED patients.

What is the current opinion about depression screening in the ED? An article published in 2000 on preventive care in the ED reviewed past USPSTF recommendations and concluded that existing research was not sufficient to make recommendations about depression screening in the ED.40 However, revised USPSTF guidelines now encourage depression screening in primary care settings that have systems in place to assure accurate diagnosis, effective treatment, and follow- up,12 and our data suggest that the integration of depression screening into routine emergency care merits serious consideration, especially if such screening can be linked to psychiatric treatment. In England, accident and emergency liaison mental health nurses effectively diagnose mental illness and ensure that patients with psychiatrie referrals in the ED are connected with appropriate mental health community services.41 In addition to increasing ED physician awareness about the high prevalence of depression in the ED, the use of psychiatric liaison nurses may prove advantageous in the ED setting. Because depression continues to be a leading cause of untreated disability in our country, further studies to evaluate the potential benefits of ED-based depression screening should become a public health priority.

CONCLUSIONS

Using a single-question screen, we found a concerning prevalence (30%) of depression among ED patients. Patients had a distinct socioeconomic profile. In the future, awareness of risk factors for depression in the ED setting and the use of simple screening instruments could aid in the recognition of depression, with subsequent referral to mental health services.

References

1. Kessler RC, Berglund P, Demier O, et al. The epidemiology of major depressive disorder: results from the National Comorbidity Survey Replication (NCS-R). JAMA. 2003; 289:3095-105.

2. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12- month prevalence of DSM-III-R psychiatric disorders in the United States. Results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994; 51:8-19.

3. Robins LN, Regier DA, eds. Psychiatric Disorders in America: The Epidemiologie Catchment Area Study. New York: The Free Press, 1991.

4. WHO. The World Health Report 2001: Mental Health: New Understanding, New Hope. Geneva: World Health Organization, 2001.

5. Gilbody S, Whitty P, Grimshaw J, Thomas R. Educational and organizational interventions to improve the management of depression in primary care: a systematic review. JAMA. 2003; 289:3145-51.

6. Nielsen AC III, Williams TA. Prevalence by self-report questionnaire and recognition by nonpsychiatric physicians. Arch Gen Psychiatry. 1980; 37:999-1004.

7. Tse SK, Wong TW, Lau CC, Yeung WS, Tang WN. How good are accident and emergency doctors in the evaluation of psychiatric patients? Eur J Emerg Med. 1999; 6:297-300.

8. Meldon SW, Emerman CL, Schubert DS, Moffa DA, Etheart RG. Depression in geriatric ED patients: prevalence and recognition. Ann Emerg Med. 1997; 30:141-5.

9. Stewart WF, Ricci JA, Chee E, Hahn SR, Morganstein D. Cost of lost productive work time among US workers with depression. JAMA. 2003; 289:3135-44.

10. Wells KB, Stewart A, Hays RD, et al. The functioning and well- being of depressed patients. Results from the Medical Outcomes Study. JAMA. 1989; 262:914-9.

11. Murray CJ, Lopez AD. Alternative projections of mortality and disability by cause 1990-2020: Global Burden of Disease Study. Lancet. 1997; 349:1498-504.

12. Screening for depression: recommendations and rationale. Ann Intern Med. 2002; 136:760-4.

13. Whooley MA, Simon GE. Managing depression in medical outpatients. N Engl J Med. 2000; 343:1942-50.

14. Dahlen I, Janson C. Anxiety and depression are related to the outcome of emergency treatment in patients with obstructive pulmonary disease. Chest. 2002; 122:1633-7.

15. Yingling KW, Wulsin LR, Arnold LM, Rouan GW. Estimated prevalences of panic disorder and depression among consecutive patients seen in an emergency department with acute chest pain. J Gen Intern Med. 1993; 8:231-5.

16. Bacaner N, Kinney TA, Biros M, Bochert S, Casuto N. The relationship among depressive and alcoholic symptoms and aggressive behavior in adult male emergency department patients. Acad Emerg Med. 2002; 9:120-9.

17. Huntley JS, Blain C, Hood S, Touquet R. Improving detection of alcohol misuse in patients presenting to an accident and emergency department. Emerg Med J. 2001; 18:99-104.

18. Rund DA, Summers WK, Levin M. Alcohol use and psychiatric illness in emergency patients. JAMA. 1981; 245:1240-1.

19. Harwitz D, Ravizza L. Suicide and depression. Emerg Med Clin North Am. 2000; 18:263-71, ix.

20. D'Amore J, Hung O, Chiang W, Goldfrank L. The epidemiology of the homeless population and its impact on an urban emergency department. Acad Emerg Med. 2001; 8:1051-5.

21. Raccio-Robak N, McErlean MA, Fabacher DA, Milano PM, Verdile VP. Socioeconomic and health status differences between depressed and nondepressed ED elders. Am J Emerg Med. 2002; 20:71-3.

22. The Sourcebook of Zip Code Demographics, ed 7. Fairfax, VA: CACI, 2002.

23. Radloff L. The CES-D scale: a self-report depression scale for research in the general population. Anonymous Applied Psychological Measurement. New York: West Publishing Co., 1977.

24. Kessler RC. Epidemiology of women and depression. J Affect Disord. 2003; 74:5-13.

25. Lorant V, Deliege D, Eaton W, Robert A, Philippot P, Ansseau M. Socioeconomic inequalities in depression: a meta-analysis. Am J Epidemiol. 2003; 157:98-112.

26. Barefoot JC, Schroll M. Symptoms of depression, acute myocardial infarction, and total mortality in a community sample. Circulation. 1996; 93:1976-80.

27. Robinson RG. Poststroke depression: prevalence, diagnosis, treatment, and disease progression. Biol Psychiatry. 2003; 54:376- 87.

28. Musselman DL, Betan E, Larsen H, Phillips LS. Relationship of depression to diabetes types 1 and 2: epidemiology, biology, and treatment. Biol Psychiatry. 2003; 54:317-29.

29. Pirl WF, Roth AJ. Diagnosis and treatment of depression in cancer patients. Oncology (Huntingt). 1999; 13:1293-301;. discussion 1301-2, 1305-6.

30. Janson C, Bjornsson E, Hetta J, Boman G. Anxiety and depression in relation to respiratory symptoms and asthma. Am J Respir Crit Care Med. 1994; 149:930-1.

31. Rimington LD, Davies DH, Lowe D, Pearson MG. Relationship between anxiety, depression, and morbidity in adult asthma patients. Thorax. 2001; 56:266-71.

32. Rubin NJ. Severe asthma and depression. Arch Fam Med. 1993; 2:433-40.

33. Dickens C, McGowan L, Clark-Carter D, Creed F. Depression in rheumatoid arthritis: a systematic review of the literature with meta-analysis. Psychosom Med. 2002; 64:52-60.

34. Robinson RG, Lipsey JR, Rao K, Price TR. Two-year longitudinal study of post-stroke mood disorders: comparison of acute-onset with delayed-onset depression. Am J Psychiatry. 1986; 143:1238-44.

35. Whooley MA, Avins AL, Miranda J, Browner WS. case-finding instruments for depression. Two questions are as good as many. J Gen Intern Med. 1997; 12:439-45.

36. Johnson J, Weissman MM, Klerman GL. Service utilization and social morbidity associated with depressive symptoms in the community. JAMA. 1992; 267:1478-83.

37. Koenig HG, Cohen HJ, Blazer DG, Meador KG, Westlund R. A brief depression scale for use in the medically ill. lnt J Psychiatry Med. 1992; 22:183-95.

38. Austin LS, Godleski LS. Therapeutic approaches for survivors of disaster. Psychiatr Clin North Am. 1999; 22:897-910.

39. Ogrodniczuk JS, Piper WE, Joyce AS, et al. Differentiating symptoms of complicated grief and depression among psychiatric outpatients. Can J Psychiatry. 2003; 48:87-93.

40. Babcock Irvin C, Wyer PC, Gerson LW. Preventive care in the emergency department, part II: Clinical preventive services-an emergency medicine evidence-based review. Society for Academic Emergency Medicine Public Health and Education Task Force Preventive Services Work Group. Acad Emerg Med. 2000; 7:1042-54.

41. Callaghan P, Eales S, Leigh L, Smith A, Nichols J. Characteristics of an accident and emergency liaison mental health service in East London. J Adv Nurs. 2001; 35:812-8.

Anita Kumar, Sunday Clark, MPH, Edwin D. Boudreaux, PhD, Carlos A. Camargo Jr., MD, DrPH

From the Department of Emergency Medicine, Massachusetts General Hospital (AK, SC, CAC), and Channing Laboratory, Department of Medicine, Brigham and Women's Hospital (CAC), Harvard Medical School, Boston, MA; and Department of Emergency Medicine, UMDNJ- Robert Wood Johnson Medical School and Cooper Hospital, Camden, NJ (EDB).

Received February 20, 2004; revision received July 7, 2004; accepted August 10, 2004.

The Emergency Medicine Network is supported in part by an Emergency Medicine Foundation Center of Excellence Award (Dallas, TX). This study was supported by an unrestricted educational grant from GlaxoSmithKline (Research Triangle Park, NC). Address for correspondence and reprints: Carlos A. Camargo Jr., MD, DrPH, EMNet Coordinating Center, Massachusetts General Hospital, 55 Fruit Street, Clinics Building 397, Boston, MA 02114. Fax: 617-724-4050; e- mail: ccamargo@partners.org.

doi:10.1197/j.aem.2004.08.053

Copyright Hanley & Belfus, Inc. Dec 2004

Source: Academic Emergency Medicine

More News in this Category