|Home | About | Journals | Submit | Contact Us | Français|
In light of the debate concerning political targets to reduce antidepressant prescribing in Scotland,1 we were interested to see Cameron et al's paper addressing the appropriateness of antidepressant prescribing by GPs.2 After consulting our Aberdeen colleagues, we examined if their findings were replicable in primary care settings in Edinburgh.
The study was carried out in two general practices in Edinburgh for five consecutive days in August and September this year. As in Aberdeen, the Hospital Anxiety and Depression Scale (HADS) questionnaire was given to patients waiting to see their doctor. GPs, blind to the questionnaire results, rated each participant on a scale of 0–3 for anxiety and depression.3 Their case notes were subsequently searched for any diagnosis of depression, antidepressant prescriptions, and indication for prescription. The GP rating of anxiety was an addition to the Aberdeen study, attempting to determine whether the presence of anxiety had any impact on how GPs diagnose depression.
Unfortunately, the response rate was very low, possibly attributable to the practice receptionists approaching the patients rather than ourselves. In the first practice there were 48 participants out of a possible 278 (17%). In the second practice only 12 took part from an eligible 500 (2.4%). Of that 60, 20% had probable depression detected by the HADS questionnaire, suggesting preferential participation from people with depression. No case of inappropriate prescribing of antidepressants was detected. Ten per cent of the population studied were rated as mildly depressed by their GP (95% CI = 0.04 to 0.35) but were not found to be depressed on the HADS questionnaire. All patients with depression were also anxious so we could not assess whether this had an impact on treatment.
These findings, despite the low response rate, are in keeping with those from Aberdeen and reinforce their concerns about Scottish Government targets to reduce antidepressant prescribing.