Elsevier

Health Policy

Volume 112, Issues 1–2, September 2013, Pages 122-132
Health Policy

The international performance of healthcare systems in population health: Capabilities of pooled cross-sectional time series methods☆☆

https://doi.org/10.1016/j.healthpol.2013.05.023Get rights and content
Under a Creative Commons license
open access

Abstract

This paper outlines the capabilities of pooled cross-sectional time series methodology for the international comparison of health system performance in population health. It shows how common model specifications can be improved so that they not only better address the specific nature of time series data on population health but are also more closely aligned with our theoretical expectations of the effect of healthcare systems. Three methodological innovations for this field of applied research are discussed: (1) how dynamic models help us understand the timing of effects, (2) how parameter heterogeneity can be used to compare performance across countries, and (3) how multiple imputation can be used to deal with incomplete data. We illustrate these methodological strategies with an analysis of infant mortality rates in 21 OECD countries between 1960 and 2008 using OECD Health Data.

Keywords

Time series
Population health
Performance analysis
Error correction model
International comparison
Infant mortality

Cited by (0)

☆☆

Open Access for this article is made possible by a collaboration between Health Policy and The European Observatory on Health Systems and Policies.