From February 2017, information about the work of the MRC/CSO Social and Public Health Sciences Unit, University of Glasgow is available and updated on the University of Glasgow website.

Data Linkage

Record linkage and Scottish Health Survey

We were part of the team involved in the original record linkage of the repeat cross-sectional Scottish Health Surveys (SHeS) to data on hospitalisations, cancer registrations, and mortality. Details of this linkage and future data resource has been described in a cohort profile.
 
Linked data from other countries provide opportunities not available within the UK. For example, using information on employer and social security grants for the Norwegian population, we showed that moderate (30-59%) downsizing in firms was associated with a 25% increase in the uptake of disability pensions, often for musculoskeletal and psychiatric diagnoses. We used linked Finnish data to examine socioeconomic disparities in mortality amenable to health care intervention. We noted a slower rate of improvement in mortality due to such causes of death in the lowest income groups and found inequalities in mortality attributable to primary care to be larger than those attributable to specialised care. In New South Wales, Australia, we are investigating the validity of potentially preventable hospitalisations as a means of assessing the effectiveness of primary care services in a NHMRC-funded study, which in turn has led to a CSO-funded project on avoidable hospitalisations in Scotland. We have used the system of linked Scottish Morbidity Records to extend our understanding of the patterning of coronary heart disease.
 
Record linkage and Indigenous health outcomes
 
Many routinely-captured variables are subject to measurement or recording error, particularly for potentially sensitive variables such as ethnicity. In NHMRC-funded work on Aboriginal health in New South Wales, we considered the impact of different algorithms for coding Aboriginal status on estimated disparities in admissions and mortality between Aboriginal and non-Aboriginal patients. Following hospital admission for AMI, Aboriginal patients have a similar risk of 30-day mortality as non-Aboriginal patients but higher risk of dying within a year. Rates of first AMI events were more than twice as high among the Aboriginal population, with the disparity being strongest in more disadvantaged and remote areas. We found Aboriginal patients to have a revascularisation rate 37% lower than non-Aboriginal patients of the same age, sex and AMI type; about half of this difference was accounted for by Aboriginal patients being more likely to present at small rural hospitals. Age- and sex-adjusted rates of cataract surgery were lower in Aboriginal people, with greater disparities in major cities and less disadvantaged areas. Higher rates of road transport injuries in Aboriginal people could be explained by higher rates in remote and deprived areas.
 
Other uses of linked data
 
Linked data from other countries provide opportunities not available within the UK. For example, using information on employer and social security grants for the Norwegian population, we showed that moderate (30-59%) downsizing in firms was associated with a 25% increase in the uptake of disability pensions, often for musculoskeletal and psychiatric diagnoses. We used linked Finnish data to examine socioeconomic disparities in mortality amenable to health care intervention. We noted a slower rate of improvement in mortality due to such causes of death in the lowest income groups and found inequalities in mortality attributable to primary care to be larger than those attributable to specialised care [305,330,777]. In New South Wales, Australia, we are investigating the validity of potentially preventable hospitalisations as a means of assessing the effectiveness of primary care services in a NHMRC-funded study [238,763], which in turn has led to a CSO-funded project on avoidable hospitalisations in Scotland. We have used the system of linked Scottish Morbidity Records to extend our understanding of the patterning of coronary heart disease.