Abstract
AIMS: To examine whether variations in pathology test requesting between different general practices can be accounted for by sociodemographic or other descriptive indicators of the practice. METHOD: This was a comparative analysis of requesting patterns across a range of pathology tests representing 95% of those requested in general practice, in 22 general practices in a single district, serving a population of 165 000. Spearman correlation coefficients were calculated and both the top and bottom fifths of activity were displayed graphically to detect trends at the extremes of the ranges. RESULTS: The proportion of women of childbearing age, median practice Townsend scores, or the existence of specialist miniclinics within the practice did not have a demonstrable impact on requesting patterns. A weak correlation was found between the proportion of elderly patients and creatinine/electrolyte testing but not for the other two tests examined for this patient group. CONCLUSIONS: The large differences observed in general practice pathology requesting probably result mostly from individual variation in clinical practice and are therefore potentially amenable to change.