Population figures for capitation formulas need to be designed differently

Editor—The organisation of health care through primary care groups with unified cash limited budgets raises several questions concerning the methods of resource allocation that will be used. The presumption is that there will be some form of weighted capitation, and this will obviously generate considerable agonising—and eventual compromise—among the doyens of resource allocations formulas¹,²; but there is one basic point which has to be resolved even before those discussions can start: what is the correct population base?

The formulas in the hospital and community health services sector are based on the resident population,³ based on estimates from the Office for National Statistics. Weighted capitation formulas in the primary sector are also based on these figures, but are adjusted to reflect patients registered with the general practitioners responsible to the health authority in question.

However, allocation to sub-health authority units such as primary care commissioning groups—and eventually to practices—have to take into account the registered list sizes compiled for each practice by the responsible health authority. These lists are on average 6% higher than the estimates by the Office for National Statistics; comparison with data provided by the General Medical Services—Statistics branch of the NHS Executive shows that this varied in 1997 between −8% in Morecambe and 23% in Ealing, Hammersmith, and Hounslow.

It has been assumed that there are “reasonable” explanations of this “inflation”: for example, mobility, mostly of young adults (ages 15-24), and delay in removing patients who have died or emigrated from practice lists.

Not unsurprisingly, health authorities have been attempting to “rationalise” the lists and “resolve” the conflicts between the two estimates. Notwithstanding these efforts, the values of list inflation have remained about the same and have been differentiated in the same way between age groups and between authorities over several years.

We are currently investigating the differences between the two population figures, and early results indicate that counting errors by health authorities account for only a minority of the differences. The two figures are defined, collected, and administered very differently. It is essential that, whatever the decision finally made as to which is the most appropriate and practical to be used in designing weighted capitation formulas, both the Office for National Statistics and the health authorities ensure that both sets of figures are accurate and that they are based on the most up to date information possible. Whichever is chosen, there will be substantial shifts in the target budgets for some authorities and practices.