Nicolau Martin-Bassols is a Post-doctoral Research Fellow at the Department of Economics at the University of Bologna. He obtained his PhD from Monash University’s Centre of Health Economics in 2022. He specialized in applied microeconomics and micro-econometrics, with a focus on health, labor economics, and social genomics. His research explores the impact of genetics, familial investments, and policy interventions on the construction of health and human capital, as well as economic disparities.

Abstract:

This paper introduces a novel method for detecting and quantifying demographic selection effects using genetic data. Specifically, we leverage polygenic index (PGI) data to examine how the distribution of genetic predispositions shifts in populations before and after major policy changes. Our primary case study focuses on the introduction of the National Health Service (NHS) in the UK in 1948, which provided universal healthcare free at the point of use. Following the NHS implementation, pregnant mothers became more likely to give birth in hospitals, leading to significant health benefits for newborns.

Using data from the Registrar General's Statistical Review, the English Longitudinal Study of Ageing (ELSA), and the UK Biobank (UKB), we first estimate the short-term effects of the NHS on infant mortality rates (IMR) through a regression discontinuity design (RDD). Our findings show a 5.48 per 1,000 birth reduction in IMR, a 16.1% decline relative to the mean. 

Building on these substantial effects, we explore whether the NHS left a genetic footprint on affected cohorts. Across multiple datasets (UKB and ELSA), we consistently find that NHS implementation increased PGI values for conditions that can be considered qualitatively detrimental, such as depression, COPD, and schizophrenia, while reducing those associated with more beneficial traits, such as life satisfaction, self-rated health, and IQ. These shifts are substantial, with effect sizes reaching 10% of a standard deviation in UKB and up to 40% in ELSA.

To account for this survival bias, we examine long-term health outcomes, including self-reported health and BMI, among individuals aged 45 to 80. In the nationally representative ELSA dataset, we find that the NHS improved these measures, and controlling for PGIs appears to partially correct for survival bias, increasing estimated effect sizes by up to 43% and improving precision. However, these findings are not consistent when using the UKB, where we find no evidence of long-term health effects following the NHS implementation.

Our findings highlight the importance of considering demographic selection effects when analyzing the long-term impact of public health interventions and suggest that genetic data can provide valuable insights for mitigating such biases.