Study objective-The aim was to audit, against agreed standards, the control of bacterial meningitis, in particular completeness of notification and appropriateness of distribution of chemoprophylaxis to contacts; and to implement appropriate changes and monitor their impact. Design-The first phase involved determination, for the years 1983 and 1984, of completeness of notification by comparison with a comprehensive case register. Information about chemoprophylaxis was obtained from case notes, questionnaires to general practitioners and other records. The second phase involved introducing a programme of clinician education in the hospital with the poorest observed notification performance and re-examining performance during 1988. Districtwide education regarding chemoprophylaxis was undertaken and the situation re-examined in 1988. Setting-The study took place in Mid Glamorgan Health Authority (population 536 000), with four acute hospitals. Population- Consisted of all the residents of Mid Glamorgan Health Authority. Main results-During the first phase of the audit only 28 out of 79 cases of bacterial meningitis were notified (35%). Performance in one hospital was significantly worse than in the other three. Chemoprophylaxis was distributed to 20 out of 26 (77%) cases of meningococcal meningitis but inappropriate drugs were used in four cases and prophylaxis was distributed more widely than is recommended in 10 cases. In the phase 2 re-examination, a significant improvement in notification was observed in the hospital where special measures were taken, with no change in a "control" hospital. Chemoprophylaxis improved throughout the District, although rifampicin continued to be distributed too widely. Conclusions-As a result of this audit, measurable improvements in both infectious disease notification and chemoprophylaxis practice were obtained by the education of clinicians. The study provides a good example of a completed audit cycle in public health medicine.
