SSRI antidepressant selection and concomitant prescribing of anxiolytics and sedative-hypnotics: evidence from primary care in the United Kingdom

Summary

Concomitant prescribing of benzodiazepines during antidepressant therapy may increase patient exposure to undesirable side-effects, increase the total cost of therapy, and reflect impairment of health-related quality of life. This study examines concomitant benzodiazepine prescriptions after initiation of therapy on three commonly prescribed SSRIs: fluoxetine, paroxetine, and sertraline. A retrospective analysis of prescribing behaviours of physicians from general practices in the United Kingdom utilises a statistical model that addresses the non-negative values and skewed distribution found in count data measures of benzodiazepine prescriptions. An adjustment is made for potential sample selection bias that might result from factors correlated with both antidepressant choice and benzodiazepine prescribing. The distribution of patients receiving benzodiazepine prescriptions across drug cohorts is: 5.4% of the fluoxetine cohort; 6.4% of paroxetine patients; and 4.3% of sertraline patients. The main results show that initiation of antidepressant therapy on paroxetine is associated with receipt of an additional 0.19 benzodiazepine prescriptions in the following six month period relative to initiation on fluoxetine. However, for the average patient receiving a prescription for an anxiolytic or a sedative-hypnotic, initiation on paroxetine is associated with receipt of an additional 3.42 benzodiazepine prescriptions relative to treatment with fluoxetine. For the average patient, antidepressant therapy with sertraline was associated with 0.05 fewer benzodiazepine prescriptions relative to treatment with fluoxetine. Also, for the average patient who receives benzodiazepine prescriptions, initiation of therapy on sertraline is associated with 0.95 fewer benzodiazepine prescriptions relative to fluoxetine treatment.

Keywords: :

• antidepressants
• resource utilisation
• sample selection