Abstract
AIMS: To compare differences in microbiology testing activity between general practices within and between five hospitals in two National Health Service (NHS) regions in England. METHODS: Retrospective capture of standardised microbiology testing activity from the laboratory computer databases. Six equivalent tests were identified and compared. Data were obtained for 174 general practices in eight primary care groups, served by two NHS hospital trusts and three public health laboratories. The total catchment population was 1,180,000 people. Comparative test activities were displayed graphically and differences in median test activity and the hospital activity distributions were examined by the Wilcoxon signed rank test. RESULTS: Median testing activity differed by 200% (urine) to 800% (wound swabs) between the trusts that performed the highest and the lowest number of tests, and from 300% to 1900% between the top and bottom 10% activity bands of general practices. Large and significant differences were found between the hospitals, irrespective of whether they belonged to the same trust, and irrespective of their geographical location. CONCLUSIONS: Large differences in microbiology testing exist within individual trust catchment areas in primary care, and there are also considerable differences between trusts. These inequalities may also introduce a selection bias into epidemiological and antibiotic resistance surveillance. This indicates a widespread need to examine and deal with the reasons responsible for these differences.