Cotrimoxazole prescribing by dispensing and non-dispensing doctors: do they differ in rationality?

Dispensing doctors (DDs) have been found to prescribe significantly more drugs, more injections and more antibiotics per patient than non-dispensing doctors (NDDs). However, the rationality of prescription in relation to diagnoses and symptoms has not been studied.

To identify and assess differences in the rationality of cotrimoxazole prescription by DDs and NDDs evaluated by drug choice and dosage in relation to diagnoses and symptoms.

Cohort study of 28 private sector DDs and 25 NDDs in Zimbabwe, using retrospective data from 688 patient records. Drug choice and dose in relation to the recorded diagnoses and symptom(s) were assessed. A new method was developed and tested to measure rationality in a specific and realistic way. A 'correctness-score list' was developed based on a panel of four evaluators assessing the rationality of cotrimoxazole prescription in relation to diagnoses and symptoms. This list was then used as a reference for the assessment.

The two groups of prescribers did not differ significantly with respect to rationality of drug choice. Both DDs and NDDs prescribed cotrimoxazole to patients for diagnoses and symptoms for which cotrimoxazole could not be justified (41.4%vs. 43.5%). DDs compared with NDDs used subcurative doses more often (26.4% of the encounters compared with 11.7%; P = 0.035) and correct doses less frequent (58.0% compared with 72.6%; P = 0.047).

The prescription of cotrimoxazole by both DDs and NDDs was widely irrational. They prescribed cotrimoxazole too frequently, and DDs used subcurative doses to a wider extent. By this prescribing practice, both groups, but especially the DDs, contribute to increasing health hazards, cost and risk of developing bacterial resistance. The method for assessing rationality of drug prescribing was found feasible, and it is recommended that it be further tested and developed.