Original Contribution

A depression screen and intervention for older ED patients^B

Fredric M. Hustey MD^a,*, Michael D. Smith MD^b

^aDepartment of Emergency Medicine, The Cleveland Clinic Foundation, Cleveland, OH 44195, USA

^bDepartment of Emergency Medicine, MetroHealth Medical Center, Cleveland, OH 44109, USA

Received 28 March 2006; revised 24 May 2006; accepted 28 May 2006

Abstract The objectives of this study were to determine the effect of Screening examinations for depression on the care of older emergency department (ED) patients and to assess recognition of depression by emergency physicians (EPs). This was a prospective interventional study of 267 patients 70 years or older. Patients were screened for depression using the Short-Form Geriatric Depression Scale, and the results were presented to EPs after assessing EP recognition of depression. The prevalence of depression was 16.5% (44/267; 95% confidence interval, 12.0%-20.9%). Fifteen (34.1%) of 44 patients with depression were recognized by EPs as being depressed. Screening results did not alter care in any of the 44 patients with depression. No patients were given referrals or discharge instructions specifically to address depression. Depression is highly prevalent and poorly recognized in older ED patients. Use of the Short-Form Geriatric Depression Scale did not alter care of older patients with depression.

D 2007

Introduction

Depression is one of the most common psychiatric disorders in older patients [1]. Up to 15% of community- dwelling elders may suffer from depression [2]. The prevalence is much higher among Geriatric patients seeking medical care. Approximately 1 in 4 elders seen in outpatient clinics may suffer from depression [2], whereas it may affect nearly half of those who are hospitalized [3,4]. Depression is also highly prevalent in older emergency department (ED)

This study was presented in part at the 2003 annual meeting of the American Geriatrics Society in Baltimore, Md, and the 2003 annual meeting of the Society for Academic Emergency Medicine in Boston, Mass.

^B This study was supported by The Cleveland Clinic, Cleveland, OH.

* Corresponding author. Tel.: +1 216 445 4558; fax: +1 216 444 1703.

E-mail address: [email protected] (F.M. Hustey).

patients. Nearly one third of older patients seen in the ED may suffer from depression [1,5]. Unfortunately, most of these older patients may go unrecognized by emergency physicians (EPs). They may also be unlikely to receive mental health referrals or other interventions, specifically for depressive symptoms [1,5].

Older patients with depression may be at higher risk for morbidity and mortality than the general population. Older adults have the highest suicide rate of any age group in the United States [6]. Depression has also been associated with increased mortality from other causes [7,8]. In addition, depression has been linked to more frequent use of the ED, prolonged hospital stays after admission, poor compliance with health care recommendations, and higher rates of healthcare resource use in general [9-12].

It has been recommended that social, psychologic, and functional status issues be routinely addressed as part of the

0735-6757/$ - see front matter D 2007 doi:10.1016/j.ajem.2006.05.016

older patient’s emergency encounter. Psychiatric screening has been previously shown to have some limited effect on physician behavior in the ED [12]. However, interventions aimed at older patients with depression are few. To our knowledge, there has been only 1 published study involving an intervention in this ED population without significant effect on depressive symptoms at follow-up [13]. We designed a study with the objective of determining the effect of a short depression screen on the care of older ED patients. A secondary objective was to assess recognition of depression by EPs.

Materials and methods

Study design

This was a prospective interventional study involving a convenience sample of older patients presenting to an ED between July 2000 and November 2001. Sampling periods were varied to include day, night, weekday, and weekend shifts with frequencies based on the usual ED presentation times of older patients. This study was reviewed and approved by the hospital institutional review board. The institutional review board waived the requirement for written informed consent for participating patients. Verbal informed consent was obtained for all patients participating in the study. For patients in whom cognitive impairment was suspected, attempts were made to contact proxies. Written consent was obtained from the 16 attending EPs participat- ing in the study.

Study setting and population

The study was conducted at an urban teaching hospital with approximately 55000 ED visits per year and an affiliated ED residency program. All patients aged 70 years or older presenting to the ED during the study period were eligible for enrollment. Patients were excluded if they refused to participate, were critically ill, or were unable to communicate or cooperate with data acquisition. Only the initial visit was included for patients presenting to the ED more than once.

Study protocol

Eligible and consenting patients were evaluated with a battery of Screening tools by trained research assistants.

Training included brief lectures on altered mental status and depression in the elderly, an overview of each of the surveys used, and practice screenings with each of the surveys under the supervision of the primary investigator. The testing battery was composed of a screen for depression using the previously validated self-administered Short-Form Geriatric Depression Scale (SFGDS) [14], a screen for cognitive impairment using the previously validated Orientation- Memory-Concentration test (OMC) [15], a screen for delirium using the Confusion Assessment Method, a standard CAGE questionnaire, and, as part of a preplanned subanalysis, a 2-question depression screen. Patients who requested assistance were given help by the research assistant in completing the surveys. The effect of the use of the mental status screenings (OMC and confusion assessment mode [CAM]) and data regarding the 2-question depression screen have been published elsewhere [16,17]. Standardized cutoff scores for the detection of depression, cognitive impairment, and alcohol abuse were used. Initial screens were observed by the primary investigator to ensure reliability. Patients and families were also interviewed regarding past medical history, living environment, and availability of home health support. Charts were reviewed for further history of depression or dementia.

Attending EPs, blinded to screening results, were

interviewed to assess for recognition of depression. Inter- views were conducted after ED physicians were able to verbalize patient final disposition and care plan, but before patient discharge from the ED. Planned Patient dispositions and referrals were recorded at this time. Results of the surveys were then unblinded, and EPs were informed of positive or negative scores. Emergency physicians were then reinterviewed regarding any change in care based on survey results. There were 16 attending EPs eligible to participate in the study. All were residency trained and board certified or board eligible in emergency medicine. The mean caseload per treating physician during the study period was 2.28 patients per hour (range, 1.85-2.70; median, 2.39).

Measurements

A change in care resulting from the use of the screen- ing tools was recorded if there were any changes in a patient’s disposition, diagnostic evaluation, or referrals made after presentation of the survey results to the EP.

Table 1 Recognition of depression by EPs
Population	Sensitivity	Specificity
All patients with depression	15/44 (34.1%; 95% CI, 20.5%-49.9%)	196/223 (87.9%; 95% CI, 83.6-92.2)
White	12/29 (41.4%; 95% CI, 23.5%-61.1%	95/111 (85.6%; 95% CI, 79.1-92.1)
African American	3/15 (20.0%; 95% CI, 4.3%-48.1%)	99/110 (90.0%; 95% CI, 82.8-94.9)
Male	4/18 (22.2%; 95% CI, 6.4%-47.6%)	88/98 (89.8%; 95% CI, 82.0-95.0)
Female	11/26 (42.3%; 95% CI, 23.4%-63.1%)	108/125 (86.4%; 95% CI, 80.4-92.4)

Fig. 1 Patient participation.

Physician recognition of depression was determined pro- spectively by comparing answers obtained from physician interviews with results of the SFGDS as the criterion standard. Standardized scores on the SFGDS for the detection of depression were used to determine the prevalence of depression (score of 5 or more indicative of depression).

Data analysis

Changes in care resulting from survey use and preva- lence data are reported as proportions with 95% confi- dence intervals (CIs). Sensitivity and specificity of EP recognition of depression with 95% CIs are also re- ported. To comply with the institutional review board requirement to protect physician confidentiality, data were not collected on individual EPs and are reported as aggregate data only.

Results

Three hundred twenty-seven eligible patients were screened during the study period, of whom 60 were sub- sequently excluded (Fig. 1). Of the remaining 267 patients, 140 (52%) were white and 151 (57%) were female. Age

ranged from 70 to 102 years (mean, 77.82 F 5.71 years).

The prevalence of depression was 16.5% (44/267; 95% CI, 12.0%-20.9%). The mean age of patients with depres- sion was 79.9 years. The mean age of patients without depression was 77.5 years. Only 8 (18.2%; 95% CI, 8.2%- 32.7%) of 44 patients positive on the SFGDS had a prior history of depression. Ten (22.7%) of these 44 patients (95% CI, 11.5%-37.8%) were taking antidepressants at the time of the ED visit. The prevalence of cognitive impairment based on OMC scores in the overall study population was 25.8% (69/267; 95% CI, 20.6%-31.1%). Patients with cognitive impairment were more likely to screen positive for depression (19/69, 27.5%; 95% CI, 17.5%-39.6%) than nonimpaired counterparts (25/198, 12.6%; 95% CI, 8.0%-17.3%).

Prospective recognition of depression by EPs is de- scribed in Table 1. Fifteen (34%) of the 44 patients with depression were recognized by EPs as being depressed. There were trends toward better recognition among whites and females, although these differences were not statistically significant (Table 1). There was also a trend toward a better recognition in patients with a previous history of depression documented. Physicians recognized depression in 5 (63%) of 8 of patients with a medical history of depression vs 10 (28%) of 36 of those without.

Physicians altered care in 0 (0%) of 44 (95% CI, 0%-8%) patients screening positive for depression after receiving the depression screening results. In addition, none of these 44 patients were given referrals or discharge instructions specifically to address depression.

Emergency department patient dispositions are described in Table 2. The overall Hospital admission rate was 38.2% (102/267; 95% CI, 32.4%-44.0%). Patients screening positive for depression were more likely to be hospitalized from the ED. Overall, 25 (56.8%) of 44 (95% CI, 41.0%- 71.6%) patients with depression were admitted to the hospital, compared with 77 (34.5%) of 223 (95% CI, 28.3%-40.8%) for those who did not screen positive on the SFGDS (relative risk, 1.62; 95% CI, 1.18-2.21).

Table 2 Patient disposition from the ED
Population	Hospitalized at index ED visit	Admitted to ED observation unit	Discharged home from ED
Overall study	102/267 (38.2%; 95% CI,	45/267 (16.9%; 95% CI,	120/267 (44.9%; 95% CI,
population	32.4%-44.0%)	12.4%-21.3%)	39.0%-50.9%)
Patients with	25/44 (56.8%; 95% CI,	6/44 (13.6%; 95% CI,	13/44 (29.5%; 95% CI,
depression	41.0%-71.6%)	5.2%-27.3%)	16.8%-45.2%)
Patients without	77/223 (34.5%; 95% CI,	39/223 (17.5%; 95% CI,	107/223 (48.0%; 95% CI,
depression	28.3%-40.8%)	12.5%-22.5%)	41.4%-54.5%)

Discussion

Our study contributes to the limited data that are currently available regarding older ED patients with depression. Although there are several studies addressing the prevalence and/or recognition of depression in older ED patients [1,5,18], to the best of our knowledge, there has been only 1 prior study published as part of a larger-scale intervention involving this population [13].

The substantial prevalence of depression in our popula- tion is in concordance with prior studies. Three previous studies have reported rates of depression in older ED patients of 27% to 32% [1,5,19]. The prevalence of 17% in our study is somewhat lower than previously found, but may also have been affected by different screening tools used [1,5] and current state of health at the time of the ED visit. In addition, a significant number of patients excluded from our study because of inability to cooperate with data acquisition or refusals may have been cognitively impaired. Rates of depression in patients found to have cognitive impairment in our study were more than twice that of nonimpaired counterparts. Excluding these patients may have resulted in lower prevalence rates of depression detected.

The poor rate of recognition of depression by EPs identified in our study is also in agreement with the few previous studies available. Meldon et al [5] reported no evidence of recognition of depression by EPs in any patient based on chart reviews of 70 depressed older ED patients. This was followed by a study in which physicians completed questionnaires rating the likelihood of depression [1]. Recognition was still poor, with only 27% of patients identified as being depressed. Our study also suggests that physicians may be less likely to recognize depression in men and minorities. Recognition of depression by EPs in women and whites was nearly double that of African Americans and men. Although these differences did not achieve statistical significance because of small sample sizes in these subgroups, they were suggestive of trends toward poorer recognition in these populations.

Providing EPs with the results of depression screening examinations did not affect the care of any of the patients in our study. Specifically, no patients with depression were given referrals specifically to address depressive symptoms. Several factors may have influenced these findings. There may be an assumption among EPs that most of these patients have primary care providers who are already managing their depressive symptoms or with whom patients can address these issues later. However, in our study, more than 80% of patients with positive scores on the SFGDS had no prior history of depression. These potentially new cases may have benefited from referrals for more detailed assess- ments. Depression may also not be considered a priority by EPs. The rapid pace of the ED often results in the urge to narrowly focus on the chief complaint. Physicians may not feel that they have the time or the resources to explore secondary issues such as depression. Illnesses of higher

acuity are also likely to overshadow depression on a regular basis. Physicians may not view depression as important unless it is the patient’s chief complaint or it is accompanied by findings of greater urgency such as suicidal ideation. Emergency physicians may also not appreciate the role of depression in health care outcomes. Prior studies have suggested that depression may contribute to increased mortality in medically ill older adults [8], play a role in increased Health care use [12], and affect outcomes of older patients receiving emergency care for obstructive pulmo- nary disease [10]. Depression may also affect health care compliance, including in those patients discharged from the ED. In addition, our findings suggest that depression may be associated with increased rates of hospitalization from the ED in older patients. Patients with depression were more than 1 1/2 times more likely to be hospitalized at the index ED visit than nondepressed counterparts. Although comor- bidities and acuity of illness at ED presentation were not controlled for in these analyses, these data suggest that further study may be warranted in this area. Finally, the timing of presentation of the SFGDS results in our study may have been suboptimal. Presenting these results to EPs earlier in the ED course before discharge planning had been completed may have given them more of an opportunity to act on the results.

Screening for depression in the ED setting presents a special challenge. The rapid pace of the ED and the high acuity of the patient population make detailed psychiatric interviews unfeasible. The previously validated 15-item SFGDS has been shown to have both good sensitivity and good specificity for the detection of depression in older patients [14], and has been shown to perform similarly to the longer 30-item form [20]. In addition, several shorter methods including 2- and 3-question screens have been shown to have good sensitivity and specificity for the detection of depression in older ED patients [16,19]. One strategy that has been suggested is to screen all older ED patients with a brief 2-question tool followed by a more detailed assessment for those patients with possible depres- sion [16]. The brevity and reliability of this tool as an initial screen makes it feasible for use in the fast-paced environ- ment of the ED.

Our study has several limitations. Our population was a convenience sampling, and sampling bias cannot be excluded. The SFGDS was used as the criterion standard for the detection of depression instead of formal psychi- atric interview focusing on Diagnostic and Statistical Manual of Mental disorders, Fourth Edition, criteria for the diagnosis of depression. Physicians were not educated regarding the reliability of the screening tool and might not have deemed the SFGDS a reliable method for detecting depression. In addition, the results of the SFGDS may not have been as reliable in the 25% of patients in our study with cognitive impairment. Finally, small sample sizes in some subgroups contributed to limited statistical signifi- cance in these analyses.

In summary, depression is highly prevalent and poorly recognized in older ED patients. This intervention using results of SFGDS screening did not alter care of older patients with depression. Further education of EPs may help to improve care in this area.

References

1. Meldon SW, Emerman CL, Schubert DS. Recognition of depression in geriatric ED patients by emergency physicians. Ann Emerg Med 1997;30(4):442 - 7.
2. Macdonald AJ. ABC of mental health. Mental health in old age. BMJ 1997;315(7105):413 - 7 [see comment].
3. Koenig HG, Meador KG, Cohen HJ, et al. Self-rated depression scales and screening for major depression in the older hospitalized patient with medical illness. J Am Geriatr Soc 1988;36(8):699 - 706.
4. Clement JP, Fray E, Paycin S, et al. Detection of depression in elderly hospitalized patients in emergency wards in France using the CES-D and the mini-GDS: preliminary experiences. Int J Geriatr Psychiatry 1999;14(5):373 - 8.
5. Meldon SW, Emerman CL, Schubert DS, et al. Depression in geriatric ED patients: prevalence and recognition. Ann Emerg Med 1997; 30(2):141 - 5.
6. Adamek ME, Kaplan MS. Caring for depressed and suicidal older patients: a survey of physicians and nurse practitioners. Int J Psychiatry Med 2000;30(2):111 - 25.
7. Ganguli M, Dodge HH, Mulsant BH. Rates and predictors of mortality in an aging, rural, community-based cohort: the role of depression. Arch Gen Psychiatry 2002;59(11):1046 - 52.
8. Ganzini L, Smith DM, Fenn DS, et al. Depression and mortality in medically ill older adults. J Am Geriatr Soc 1997;45(3):307 - 12.
9. Ciechanowski PS, Katon WJ, Russo JE. Depression and diabetes: impact of depressive symptoms on adherence, function, and costs. Arch Intern Med 2000;160(21):3278 - 85.
10. Dahlen I, Janson C. Anxiety and depression are related to the outcome of emergency treatment in patients with obstructive pulmonary disease. Chest 2002;122(5):1633 - 7.
11. Kales HC, Blow FC, Copeland LA, et al. health care utilization by older patients with coexisting dementia and depression. [see com- ment]Am J Psychiatry 1999;156(4):550 - 6.
12. Fischer LR, Wei F, Rolnick SJ, et al. Geriatric depression, antidepressant treatment, and healthcare utilization in a health maintenance organization. J Am Geriatr Soc 2002;50(2):307 - 12.
13. McCusker J, Verdon J, Tousignant P, et al. Rapid emergency department intervention for older people reduces risk of functional decline: results of a multicenter randomized trial [see comment]. J Am Geriatr Soc 2001;49(10):1272 - 81.
14. Lesher EL, Berryhill JS. Validation of the Geriatric Depression Scale- Short Form among inpatients. J Clin Psychol 1994;50(2):256 - 60.
15. Katzman R, Brown T, Fuld P, et al. Validation of a short Orientation- Memory-Concentration Test of cognitive impairment. Am J Psychi- atry 1983;140(6):734 - 9.
16. Hustey FM. The use of a brief depression screen in older emergency department patients. Acad Emerg Med 2005;12(9):905 - 8.
17. Hustey FM, Meldon SW, Smith MD, et al. The effect of mental status screening on the care of elderly emergency department patients. Ann Emerg Med 2003;41(5):678 - 84.
18. Raccio-Robak N, McErlean MA, Fabacher DA, et al. Socioeconomic and health status differences between depressed and nondepressed ED elders. Am J Emerg Med 2002;20(2):71 - 3.
19. Fabacher DA, Raccio-Robak N, McErlean MA, et al. Validation of a brief screening tool to detect depression in elderly ED patients. Am J Emerg Med 2002;20(2):99 - 102.
20. Burke WJ, Roccaforte WH, Wengel SP. The short form of the Geriatric Depression Scale: a comparison with the 30-item form. J Geriatr Psychiatry Neurol 1991;4(3):173 - 8.