The Relationship Between Health and Homelessness in Scotland

Main Article Content

Andrew Waugh
David Rowley
Auren Clarke


Homeless people are among the most vulnerable and socially excluded in society, exhibiting higher premature mortality and multiple morbidity. This study links health and homelessness data for the first time at a national level in Scotland.

This study adds to the evidence base on the relationship between homelessness and health needs, including how homeless people exhibit multiple or complex needs.

This study matches those who have experienced homelessness in Scotland since 2002 (n=430,000) with two people of the same age and sex – at random, one control is chosen from the 20% least deprived areas in Scotland and another from the 20% most deprived. This gives 1.3 million people in total. Each cohort is linked to a wide range of health datasets. More detail on the methodology is presented in a separate talk (Challenges of Analysing Case-Control Datasets: A Health and
Homelessness Case Study) at this conference.

At the 2018 ADRN Health and Wellbeing session, we wish to present on our findings that we are publishing shortly before the conference. Large and complex differences were found between the cohorts in health service use and outcomes, particularly for the homeless. Analysis is presented on the different health datasets used, including acute and mental-health admissions, A&E attendances, outpatient appointments, prescriptions and drugs misuse assessments. We also discuss findings for activity related to particular issues including drugs and alcohol. Furthermore, notable differences were also found between once only and repeat homeless individuals.

The range of linked health data used in this study has enabled invaluable evidence to be collected on the scale, severity and complexity of the health needs of homeless populations in Scotland.

Article Details

How to Cite
Waugh, A., Rowley, D. and Clarke, A. (2018) “The Relationship Between Health and Homelessness in Scotland”, International Journal of Population Data Science, 3(2). doi: 10.23889/ijpds.v3i2.492.