All GPs are different and some are more different than others

The causes of the wide variation in GPs' behaviour referring their patients to specialists is an under-researched subject, but is likely to become more scrutinised with the advent of cost conscious NHS initiatives like ‘practice-based commissioning’ and ‘payment by results’. O'Sullivan et al's study¹ showed that most of the variation in referrals remains unexplained, and suggests that 30% is attributable to differences in morbidity.

This study was dependent on the choice of diagnostic codes by GPs. Is this a important flaw in the study? There is a widespread belief that GPs often choose the diagnostic label after they decide how to manage a patient, and the diagnostic label will support the action (such as referral to specialist care) that they have already determined. In the absence of objective diagnostic codes for all clinical encounters, information derived from these codes informs us more about variation in GPs' behaviour than patient morbidity. Future research should focus on why GPs seem to manage similar problems very differently.