Patient factors associated with SSRI dose for depression treatment in general practice: a primary care cross sectional study

BACKGROUND Antidepressant prescribing continues to rise. Increased long-term prescribing and higher doses are contributing to current growth; however, patient factors associated with the use of higher doses remain unknown. This study's aim was to investigate patient factors associated with selective serotonin re-uptake inhibitor (SSRI) prescribed daily dose for depression treatment in general practice. METHODS A stratified sample of low to high prescribing practices were selected. Routine individual patient-level data were extracted one practice at a time: September 2009 to January 2011. Patients included were ≥18 years, and prescribed an SSRI for depression. Logistic regression analysis was undertaken to assess individual predictor variables on SSRI daily dose by standard therapeutic dose versus higher dose, as SSRIs demonstrate flat dose response curves for depression treatment. Predictor variables included: age, gender, deprivation, co-morbidity, smoking status, being prescribed the same SSRI for ≥2 years, and patients' general practice. For a subgroup of patients a second sub-group analysis included long-term benzodiazepine and/or z-hypnotic (B&Z) as a predictor variable. RESULTS Inter-practice SSRI prescribing varied significantly; practice point prevalence ranged from 2.5% (94/3697) to 11.9% (359/3007) of the practice population ≥18 years old; median 7.3% (250/3421) (χ2 = 2277.2, df = 10, p < 0.001). Overall point prevalence was 6.3% (3518/52575), with 5.8% (3066/52575) prescribed SSRIs for depression of whom 84.7% (2596/3066) had data for regression analysis. Higher SSRI doses were significantly associated with, in descending order of magnitude, individual practice attended, being prescribed the same SSRI for ≥2 years (Odds Ratio (OR) 1.80, 95% CI 1.49 to 2.17, p < 0.001) and living in a more deprived area (OR 1.55, 95% CI 1.11 to 2.16, p = 0.009). Higher SSRI doses in the B&Z subgroup were significantly associated with individual practice attended, being prescribed a long-term B&Z (OR 2.05 95% CI 1.47 to 2.86, p < 0.001) and being prescribed the same SSRI for ≥2 years (OR 1.94, 95% CI 1.53 to 2.47, p < 0.001). CONCLUSION Higher SSRI doses for depression were associated with practice attended and being prescribed the same antidepressant for ≥2 years. As long-term antidepressant use increases, the use of higher doses may further contribute to prescribing growth.