Improving cost-effectiveness of hypertension management at a community health centre

Objectives. To describe the pattern of prescribing for hypertension at a community health centre (CHC) and to evaluate the impact of introducing treatment guidelines and restricting availability of less cost-effective antihypertensive drugs on prescribing patterns, costs of drug treatment and blood pressure (BP) control.Design. Before/after intervention study.Setting. Medium-sized CHC in the Cape Flats area of Cape Town.Subjects. 1 084 hypertensive patients attending the CHC, who had at least two prescriptions for antihypertensive drugs during a 1-year period starting on 1 January 1992. Interventions. 1. Implementation of stepped-care guidelines for hypertension, specifying treatment with more cost-effective drugs and minimising drug treatment. 2. Reducing availability for routine prescribing by CHC doctors of 10 less cost-effective antihypertensive drugs or drug combinations.Outcome measures. 1. Mean number of drugs prescribed per patient. 2. Proportion of prescriptions for: each major class of antihypertensive drug; restricted availability and freely prescribable drugs; and more and less cost-effective drugs. 3. Mean monthly cost of drugs prescribed per patient. 4. Mean blood pressure and proportion of BP readings controlled (<160/95 mmHg) or uncontrolled (≥160/95 mmHg).Results. A mean of 1.7 active drugs was prescribed per patient per visit. The most frequently prescribed drugs were thiazide-like diuretics (44.8%), centrally acting agents (28.4%) and b-blockers (13.2%). Mean monthly drug costs per patient decreased significantly by R1.99 (24.2%) from R8.24 to R6.25 between the first and last prescription for each patient (exclusive of any reduction due to withdrawal of treatment). This was attributable to reduced prescribing of more expensive drugs withdrawn from routine use and a 51.1% increase in prescribing of the most cost-effective drugs. The overall annual cost-saving of the changes in prescribing for this CHC are estimated at R75 150. Blood pressure control did not change significantly.Conclusion. The pattern of changes in prescribing and drug costs was consistent with a causal effect of the interventions. The study demonstrates the potential forimproving cost-effectiveness of hypertension care in primary care in South Africa and the potential for research in this setting