Receiving guideline-concordant pharmacotherapy for major depression: Impact on ambulatory and inpatient health service use

Abstract

Objective: This study aimed to determine the associations between guideline-concordant pharmacotherapy for depression and the use of health services in the year following diagnosis. Method: This population-based, retrospective cohort study examined Quebec drug plans between 1999 and 2002. We included beneficiaries aged 18 to 64 years who were newly diagnosed with an episode of depression by primary care physicians and psychiatrists between October 1, 2000, and March 31, 2001, and who made at least one psychotropic pharmacy claim within 31 days of diagnosis. We defined guideline concordance as the receipt of recommended medication, starting dosage, and treatment duration as defined by the Canadian Network for Mood and Anxiety Treatments guidelines. We measured outcomes on use of ambulatory (number of visits to prescribing physician, other physicians, or emergency departments) and inpatient (hospitalization) services. Results: There were 2742 patients (mean age 42 years; 64% female patients) who met the study criteria. Of the 2047 (75%) patients to whom an antidepressant was dispensed, 1958 (71%) received a recommended first-line medication, 1297 (63%) received a recommended starting dosage, and 304 (15%) received a recommended duration. According to the 3 criteria, only 8% were treated appropriately; 21% received benzodiazepines rather than antidepressants. There were 2 median visits (inferquartile range [IQR] 1 to 3) to prescribing physicians, 0 visits (IQR 0 to 1) to other physicians, and 0 visits (IQR 0 to 0) to emergency departments; 497 (18%) patients were hospitalized. In separate multivariate models for repeated measures, recommended first-line medication, dosage, and duration were associated with more prescribing physician visits. Recommended first-line medication reduced the odds of hospitalization. Conclusion: Guideline concordance was associated with more visits to prescribing physicians and lower odds of hospitalization.