The dynamics of health in the British Household Panel Survey

Abstract

SUMMARYThis paper considers the dynamics of a categorical indicator of self-assessed health using eight waves (1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998) of the British Household Panel Survey (BHPS). Our analysis has three focal points: the relative contributions of state dependence and heterogeneity in explaining the dynamics of health, the existence and consequences of health-related sample attrition, and the investigation of the effects of measures of socioeconomic status, with a particular focus on educational attainment and income. To investigate these issues we use dynamic panel ordered probit models. There is clear evidence of health-related attrition in the data but this does not distort the estimates of state dependence and of the socioeconomic gradient in health. The models show strong positive state dependence and heterogeneity accounts for around 30% of the unexplained variation in health.

Should we brush twice a day? Determinants of dental health among young adults in Finland

“…These studies have shown that state dependence is highly significant, implying that there is a strong continuance or persistence in health care use from one period to the next. Having explored the dynamics of self-assessed health in the BHPS, Contoyannis et al [54] introduce evidence of persistence in health explained by state dependence and individual heterogeneity. All these findings, as regards our study, suggest that caries might tend to be repeated or recurring throughout many periods of an individual's life and regular dental care use would play an important role in reducing caries.…”

Informal care and labour force participation among middle-aged women in Spain

“…Lastly, we use the inverse of the predicted probabilities for each individual 1=p it ð Þ; to weight the contributions of each observation to the log-likelihood function. In this respect, as mentioned by Contoyannis et al (2004), the IPW estimator can be applied in situations where the objective function is additive in the contribution of each observation. This is why this estimator cannot be used in models such as the random effects ordered probit model, for which-as can be seen in expression (5)-there is a term consisting of the product of the contributions of the observations of any given individual for different time periods.…”

Early Retirement Among Men in Britain and Germany: How Important is Health?

“…Subsequent transitions could be analysed but, due to the small sample sizes, we have chosen not to pursue this. 39 See Contoyannis et al (2004). 40 See Jones and Wildman (2008).…”