Current issues of ACP Journal Club are published in Annals of Internal Medicine


Diagnosis

Review: Case-finding instruments identify depression in primary care

ACP J Club. 1995 Nov-Dec;123:70. doi:10.7326/ACPJC-1995-123-3-070


Source Citation

Mulrow CD, Williams JW Jr, Gerety MB, et al. Case-finding instruments for depression in primary care settings. Ann Intern Med. 1995 Jun 15:122:913-21.


Abstract

Objective

To evaluate case-finding instruments (CFIs) (questionnaires and checklists) for identifying patients with major depression in primary care settings.

Data sources

English-language studies were identified through a search of MEDLINE using the terms depressive disorder, depression, diagnosis, and specific names of CFIs cited in previous reviews, bibliographies of relevant papers, and contact with experts.

Study selection

Trials were selected if they studied patients attending primary care clinics, if both a CFI and a diagnostic standard were administered, and if the diagnostic standard had formal criteria for depression. Exclusion criteria were diagnosis without formal interview procedures or diagnostic criteria or patient selection based on specific conditions or demographic characteristics.

Data extraction

Data extraction included specific CFIs and their characteristics, patient demographic characteristics, diagnostic standard used, and number of persons who screened positive or negative and who met the criteria for major depression. Mean sensitivities and specificities were calculated, with weighting based on sample size.

Main results

9 CFIs were evaluated in 18 studies. Administration took 2 to 10 minutes (8 CFIs ≤ 5 min). 6 were at a grade 3 to 5 reading level and 3 were at a grade 6 to 9 level. CFIs included the Beck Depression Inventory (2 studies, 21 items, depression-specific), Center for Epidemiologic Studies Depression Screen (7 studies, 20 items, depression-specific), General Health Questionnaire (5 studies, 28 items, psychiatric illness), Hopkins Symptom Checklist (1 study, 25 items, depression category), Medical Outcomes Study Depression Screen (3 studies, 8 items, depression-specific), Popoff Index of Depression (1 study, 15 items, depression-specific), Primary Care Evaluation of Mental Disorders (1 study, 2 items, depression category), Symptom Driven Diagnostic System-Primary Care (2 studies, 5 items, depression category), and Zung Self-Assessment Depression Scale (4 studies, 20 items, depression-specific). With 1 outlying study excluded, sensitivities ranged from 67% to 99%, with an overall sensitivity of 84% (95% CI 79% to 89%). Specificities ranged from 40% to 95% with an overall specificity of 72% (CI 67% to 77%). No statistically significant differences in operating characteristics were observed among the instruments.

Conclusion

Most case-finding instruments designed to identify patients with major depression in a primary care setting have reasonable operating characteristics.

Source of funding: Agency for Health Care Policy and Research.

For article reprint: Dr. C.D. Mulrow, Audie L. Murphy Memorial Veterans Hospital (11C6), 7400 Merton Minter Boulevard, San Antonio, TX 78284, USA. FAX 210-567-4423.


Commentary

The use of CFIs for detecting depression in primary care practice is a germane topic in the evolving era of managed care. Patients with depression and other psychosocial illnesses often present with somatic symptoms, and their care is expensive. Much of the expense is incurred because the diagnosis of major depression is made as a "diagnosis of exclusion" that requires expensive testing to "rule out organic disease" rather than proactively identifying depression as causal or as contributing to the patient's distress (1). Approaches that facilitate the identification of depression and other psychosocial problems have the potential to improve both the quality and cost of care (2).

Mulrow and colleagues provide a valuable service in comparing the performance of available depression CFIs. Their finding of little difference is not surprising given the broad CIs for the operating characteristics of the CFI instruments. Thus, clinically important differences may have been missed. Their finding that if a CFI were given to 100 primary care patients who had a 5% prevalence of major depression, only 4 of 31 patients with positive responses will be "true positives" should give physicians pause before they broadly use these CFIs in practice. Positive responses, however, may serve as markers for other psychosocial problems that primary care physicians would find useful to know or at least explore.

A danger in using CFIs is the indiscriminate use of antidepressant medications in patients with a positive result before a careful history is done to determine whether depression is present. Further studies should look at how physicians respond to a positive screen and the subsequent outcomes in patient care.* Meanwhile, physicians should consider routinely using these tools only if they carefully evaluate the responses before either prescribing medication or labeling the patient with a psychiatric disorder.

David L. Bronson, MD
Kathleen S.N. Franco, MDCleveland Clinic FoundationCleveland, Ohio, USA


References

1. Franco K, Tamburino M, Campbell N, et al. The added costs of depression to medical care. Pharmacoeconomics. 1995;7:284-91.

2. Bingham RL, Plante DA, Bronson DL, Tufo HM, McKnight K. Establishing a quality improvement process for the identification of psychosocial problems in a primary care practice. J Gen Intern Med. 1990;5:342-6.


*Update note

A recent study has been done to address this area (1).

1. Williams JW Jr, Mulrow CD, Kroenke K, et al. Case-finding for depression in primary care: a randomized trial. Am J Med. 1999;106:36-43.