The maldistribution of general practitioners in England and Wales: 1974-2003.

BACKGROUND The geographical distribution of general practitioners (GPs) is a persistent policy concern within the National Health Service. Maldistribution across family health service authorities in England and Wales fell between 1974 and the mid-1980s but then remained, at best, constant until the mid-1990s. AIM To estimate levels of maldistribution over the period 1994-2003 and to examine the long-term trend in maldistribution from 1974-2003. DESIGN Annual snapshots from the GP census. SETTING One hundred 2001 'frozen' health authorities in England and Wales for 1994-2003 and 98 family health service authorities for 1974-1995. METHOD Ratios of GPs to raw and need-adjusted populations were calculated for each health authority for each year using four methods of need adjustment: age-related capitation payments, national age- and sex-specific consultation rates, national age- and sex-specific limiting long-term illness rates, and health authority-specific mortality. Three summary measures of maldistribution across health authorities in the GP to population ratio--the decile ratio, the Gini coefficient, and the Atkinson index--were calculated for each year. RESULTS Maldistribution of GPs as measured by the Gini coefficient and Atkinson index increased from the mid-1980s to 2003, but the decile ratio showed little change over the entire 1974-2003 period. Unrestricted GP principals and equivalents were more equitably distributed than other types of GP. CONCLUSION The 20% increase in the number of unrestricted GPs between 1985 and 2003 did not lead to a more equal distribution.