No previous analysis has investigated the determinants of screening uptake for breast and cervical cancer screening for possible spillover effects from one type of screening examination to the other type of screening examination with a dynamic bivariate panel probit model. For our analysis, we used a dynamic random effects bivariate panel probit model with initial conditions (Wooldridge-type estimator) and dependent variables were the participation of breast and cervical cancer screening in the recent year. The balanced panel sample consisted of 844 women from the British Household Panel Survey (BHPS) from the time period 1992 to 2008. Our analysis showed the high relevance of past screening behaviour and the importance of state dependency for the same and the other type of cancer screening examinations even after controlling for covariates and unobserved heterogeneity. The uptake for breast and cervical cancer screening was higher when the same screening examination was done one or three years earlier. This result is in accordance with the medical screening programmes in Great Britain. With regard to breast and cervical cancer screening positive spillover effects existed between screening examinations in the third order lags. Women with a previous visit to a general practitioner and individuals in the recommended age groups had a higher uptake for breast and cervical cancer screening. Other socioeconomic and health related variables had non-uniform results in both screening examinations. Promoting the uptake of one female prevention activity could also enhance the uptake of the other prevention activity.
Dynamic panel probit model Bivariate probit Preventive medicine State dependence Screening Health check-ups