Non- and semi-parametric estimation of age and time heterogeneity in repeated cross-sections: An application to self-reported morbidity and general practitioner utilization

Patterns of self-reported morbidity and general practitioner (GP) utilizati on exhibit complex age, sex and time heterogeneity. Underlying patterns are often obscured by data which are overly 'rough' because of noise associate d with adjacent year fluctuations. Ih this paper we describe methods to obt ain smoothed estimates of age, time and birth-cohort effects using data fro m the General Household Survey (GHS), covering the period 1984-1995/6 inclu sive. The methods outlined offer powerful analytic tools to research comple x profiles or trends, particularly over age or time. The relationships of the morbidity and GP utilization measures with age, se x and survey year characteristics are estimated non-parametrically using ro ughness penalized least squares (RPLS). A semi-parametric extension of this model is used to estimate the effect of the morbidity variables on GP util ization. Tests are employed for various forms of age and time heterogeneity including birth-cohort effects. Linear age specifications are rejected for all variables and evidence is found of time heterogeneity in one of the mo rbidity measures-limiting long-standing illness (LS)-and GP utilization. Th e advantages of employing non- and semi-parametric estimations in the prese nce of complex relationships such as those observed for age and time profil es are discussed. Adoption of these techniques by applied econometricians w orking in health economics is encouraged. Copyright (C) 1999 John Wiley & S ons, Ltd.