Governmental intervention in Hospital Information Exchange (HIE) diffusion: a quasi-experimental ARIMA interrupted time series analysis of monthly HIE patient penetration rates

This study examines changes in the monthly penetration rates of a Health Information Exchange (HIE) in a large Health Maintenance Organisation (HMO) in Israel after its successful adoption, and how those rates changed in anticipation of a government policy to turn this HIE into a national system. Penetration rate is the proportion of patients whose data have been accessed through the HIE. We apply the Bass model to the penetration data and estimate an ARIMA interrupted time series analysis on the resulting dependent variable. In the Bass model, the diffusion of new products or services over time follows an S curve, where the proportion of non-users who take up the new technology is assumed to be a linear function of the proportion who are already users. The results indicate (1) that also HIE penetration shows a Bass model pattern, thus extending previous research that indicated that the adoption rate of an HIE (i.e. its initial installation) shows a Bass model pattern, and (2) that there was a significant one-time increase in penetration in this HMO when hospitals in other HMOs started training towards adopting this HIE in preparation for it becoming the national system. Implications are discussed.