Geographic influences on the uptake of infant immunisations: 1. Concepts, models, and aggregate analyses

This is the first of two papers in which the effects on the uptake of immunisation of transport, time - space, and gender-role constraints, among a wider range of influences, are assessed statistically. A critique of a paper by Jarman et al leads to the formulation of an improved conceptual and statistical framework for analyses of uptake. Within this framework, the possibility of explaining immunisation uptake by using readily available data at the District Health Authority scale is reevaluated. Results suggest that analyses solely at this highly aggregate scale are plagued by the statistical problem of overdispersion, and cannot provide reliable explanations of uptake. Rather, it is argued, disaggregate or, preferably, multilevel analyses are required. Such analyses form the subject matter of the second paper.