Poor health, crime, and social-welfare dependency cluster in a small segment of the population

Analyses of administrative data from two nations reveal opportunities to reduce health and social inequalities

Income disparities are at the forefront of economic and political debate, and are common knowledge among citizens. The gap between the rich and the poor has grown in most advanced economies, with income becoming increasingly concentrated in a small, fortunate segment of the population. Comparably less public attention has been paid to resources that are concentrated among the less fortunate. Healthcare researchers worry about the hospital revolving-door problem, labor economists are concerned about welfare dependency, and criminologists know that a small percentage of offenders commit the majority of crimes. My colleagues and I observed, however, that no one had brought these literatures together. In the current paper, we show that these different fields have been studying the same people. 

We leveraged an exciting data resource. In 2011 Statistics New Zealand established the Integrated Data Infrastructure (IDI), a collection of de-identified administrative data on the people of New Zealand that are linked at the individual level. This represented a significant feat of data linkage: the IDI contains information from a wide range of government sectors, including health, social services, education, justice, migration and movements, the Census, and national surveys. In 2018 I traveled to the Centre of Methods and Policy Application in the Social Sciences at the University of Auckland to work with the IDI data, in collaboration with Drs. Barry Milne and Stephanie D’Souza.

Using a measure of statistical dispersion from economics (the Gini coefficient), we measured the distributions of public hospital nights, prescription drug fills, criminal convictions, social welfare months, and injury insurance claims in 1.7 million New Zealanders of working age. In each sector, service use was highly concentrated in a small segment of citizens. Further, the same small segment of citizens accounted for a disproportionate share of services used and public-service costs across multiple different sectors. With the assistance of Dr. Signe Hald Andersen, Deputy Chief of Research at the Rockwool Foundation Research Unit, we replicated these findings in administrative registers linked to 2.3 million Danish citizens.

What, then, differentiates this high-need population segment from other individuals? To address this question, we linked the New Zealand administrative databases to the Dunedin Longitudinal Study, a matched, representative cohort of 1,037 babies born in Dunedin, New Zealand who have been followed from birth into their mid-forties. We focused on early-life factors that impact individuals’ ability to obtain education and training, which are becoming increasingly important for the workforce. High-need service-users had low childhood brain health, which limits primary-school readiness; poor adolescent mental health, which impacts secondary-school learning; and were likely to drop out of secondary school. They also reported low levels of life satisfaction in adulthood.

There is growing international debate about how to best support the needs of citizens struggling to meet changing workforce demands. Our findings point to early cognitive health, mental health, and school retention as potential targets for intervention. Our study cannot resolve whether these risk factors are causal. However, our analysis suggests that if associations are causal, investing in young people’s education and training potential could substantially reduce health- and social-service dependence, mitigate their economic costs, and improve population wellbeing.

Our study also highlights the discovery value of linked administrative data. Without the ability to integrate information across multiple health and social sectors, we would not have been able to uncover such a high-priority segment of the population. Some might wonder how these findings apply to countries where linked administrative databases do not yet exist. However, efforts to create these databases are ongoing. In the U.S., integrated health-information systems are becoming increasingly widespread at the regional and state levels. In both the U.S. and the U.K., there are efforts to link administrative data across systems, including education, health, and criminal-justice systems. Our study lays a road map for what could be achieved in these and other countries.

Our labor force is rapidly evolving to require more specialized technical skills. A portion of the population is being left behind, and is depending on public services. Efforts to support the educability and trainability of young citizens could benefit all members of society.

You can find the paper here.


Please sign in or register for FREE

If you are a registered user on Behavioural and Social Sciences at Nature Portfolio, please sign in