We wish to report the follow-up results of a previously described randomized controlled trial1 to evaluate an educational strategy to improve family physicians' use of clinical practice guidelines for the detection and management of depression. We measured depression using the Centre for Epidemiologic Studies Depression (CES-D) scale.2 The primary outcome was the "gain" score (the difference between the first and last CES-D scores).
At 6 months, the mean gains for patients in the intervention and control groups were 17.9 and 16.5 respectively (p = 0.04) (Table 1). One year later, 18 months after the intervention, the corresponding gain scores were 17.9 and 13.4 (p = 0.09) (Table 1). There was an apparent, but not significant, deterioration of CES-D scores in the control group over the 12-month interval; the scores of the intervention group remained stable.
Table 1. Self-reported depressive symptoms at 6- and 18-month follow-up assessments
The numbers of patients available for follow-up dropped from 85 to 65 between 6 and 18 months; despite a greater difference in mean gain score at 18 months, the result is not statistically significant.
We also examined whether patients who saw a physician of their own gender did better than those who saw a doctor of the other gender. Interestingly, gender-matched physician-patient dyads showed higher mean gain scores (21.26 [SD 14.90]) than gender-unmatched dyads (16.40 [SD 13.91]) but, again, the sample was too small and the variance was too great for this difference to approach statistical significance (p = 0.18).
Although the loss of patients to follow-up in our study means that the results should be cautiously interpreted, and despite the various factors affecting retention, it is encouraging that the modest benefits that we detected at 6 months in our study appeared to be maintained at 18 months. The long-term effects of this and other medical education strategies require further investigation.