The Prevention of Suicide in Primary Care Elderly: Collaborative Trial (PROSPECT) evaluated the impact of primary care interventions on reducing major risk factors for suicide in elderly persons. Depression is one such risk factor. Depression in later life frequently remains improperly diagnosed in primary care and often is treated inadequately. Pharmacotherapy commonly is used in insufficient dosages, and both patients and physicians may stop therapy prematurely. Bruce and colleagues report on results from the PROSPECT study.

Patients were randomized to depression treatment intervention versus usual care and followed at regular intervals for suicidal ideation, depressive symptoms, and remission rates in depressive symptoms. Practice-based depression care managers helped physicians identify depressed patients and provided patients with information on the treatment and monitoring of depression.

Depressed patients in the intervention group were given citalopram or offered psychotherapy if they declined medication. Both interventions were paid for with research funds. In the usual-care group, a minimum enhancement was provided: physicians were notified when patients met the criteria for depression and received information on treatment guidelines.

The investigators randomized patients by practice, recruiting in an age-stratified method participants age 60 to 74 and 75 years and older. Patients completed a Centers for Epidemiologic Studies Depression scale (CES-D), and patients with a CES-D score higher than 20 were eligible to participate. An additional 5 percent sample was included for comparison, as were some patients scoring 20 and below who might have had a false-negative screening result and who were identified as possibly depressed by supplemental questioning. Patients were assessed by telephone at four and eight months and had an in-person interview at 12 months. A total of 1,238 eligible patients agreed to a baseline interview; 598 patients who met the criteria for major or minor depression were selected to participate, as well as 47 patients who were selected randomly and 109 who were diagnosed with depression by supplemental questioning.

Patients in the intervention group were significantly more likely than those in the usual-care group to receive treatment at each follow-up visit. Patients in the intervention group were more likely to report suicidal ideation at baseline. At four months, both groups had similar rates of suicidal ideation, which reflected a greater decline in suicidal ideation in the intervention group. This finding applied overall and to patients with major depression, but not to the subgroup with minor depression. Depression severity did not differ significantly between the groups at baseline but decreased more in the intervention group than in the usual-care group. This finding also applied to the group as a whole and to patients with major depression, but not significantly to patients with minor depression.

Similarly, more intervention patients experienced a 50 percent or more decrease in depression scores from baseline compared with the usual-care group at four months, again with insignificant impact on those with only minor depression. In terms of remission, rates were significantly higher at four months in the intervention group compared with the usual-care group, but rates narrowed to nonsignificance at eight and 12 months. The difference in suicidal ideation was most pronounced and statistically significant at eight months, with no difference between groups at 12 months.

Suicidal ideation resolved more quickly in intervention patients than in usual-care patients. Intervention patients also had a more favorable course of depression, in terms of both severity and remission measurements, although these findings applied to patients with major rather than minor depression. Actual rates of suicide in primary practice occur too infrequently to measure the effect of an intervention such as the one studied in PROSPECT.

The strength of this trial lies in its practice-based design and application of formal depression screening and diagnosis. The feasibility of using case managers in practices needs to be assessed in real-life practice, and providing therapeutic intervention at no cost to the patient is a potential limitation of the study’s real-world application. Finally, the authors were unable to explain the higher baseline prevalence of suicidal ideation in the intervention group, an occurrence that may limit the generalizability of the study’s findings.