Publicatie

Patient and GP characteristics associated with antidepressant treatment in depressed patients: a multilevel analysis.

Dijk, L. van, Volkers, A., Bakker, D. de. Patient and GP characteristics associated with antidepressant treatment in depressed patients: a multilevel analysis. Pharmacoepidemiology and Drug Safety: 2007, 16(suppl. 2)
Background: Antidepressants are frequently prescribed drugs in Dutch general practice, but patients differ in the treatment they receive. Both patient and GP characteristics may explain this difference. Objectives: To identify patient and GP characteristics associated with antidepressant treatment in depressed patients. Methods: Routinely collected general practice registration data (including a morbidity registration ) and a patient census both derived from the 2nd Dutch National Survey of General Practice (2001) covering 195 GPs and their patients. Setting: Included were 116 GPs and 6255 patients diagnosed with depression or depressed feelings. Main outcome measures: 1) patient receives at least one antidepressant prescription (yes/no); 2) in case of antidepressant prescription: TCA or SSRI? Statistical analysis: Logistic multilevel model with predictors at GP and patient level. Results: Main results: At the patient level burden of depression - depression or depressed feelings - was the strongest predictor for receiving antidepressant treatment (OR: 6.8; CI:5.6–8.3), followed by having anxiety as well (OR¼2.1; CI: 1.6–2.7). Depressed women were more likely to receive antidepressant treatment than men (OR:1.3; CI:1.01–1.7). Socioeconomic status was not associated
with antidepressant treatment. GPs in single-handed practices (OR:1.4; CI:1.01–1.9) wrote more prescriptions than other GPs. GP’s overall number of prescriptions per patient was associated with higher prescription rates for antidepressants (OR:1.2; CI:1.02–1.3), while following guidelines was associated with fewer antidepressant prescriptions (OR: 0.8; CI 0.7–0.99). The choice between SSRI versus TCA was mainly affected by patient characteristics. Patients who were younger, of Dutch origin and those who were employed were more likely to receive SSRIs. Conclusions: 1) determinants for antidepressant treatment (yes/no) and choice for a certain type of antidepressant differ; 2) Patients from higher socioeconomic backgrounds more often use SSRIs. This may be caused by their ability to express their wishes concerning medication more profoundly. 3) GPs’ evaluation of the burden of disease is the most important factor in the choice for antidepressant treatment. Therefore, good diagnostics skills in general practice are recommended. (aut. ref.)