University of Glasgow
Measuring health inequalities in Malta: Development and application of a small area-based deprivation index
Abstract
dc:description.abstractBackground - Improvement in population health requires that health systems focus on social determinants as important risk factors. Despite some limitations, small area-based deprivation indices have been useful in the analysis of health inequalities. These indices can be linked to data sources through small area-based geographical information. The aim of this thesis was to develop and validate a small area-based deprivation index for Malta using census data and apply it to the analysis of health inequalities. Methods - Data from two censuses were used to create the index. Nineteen area level items were considered. After Z-standardisation, principal component analysis was used for item reduction and a weighted additive score was constructed. Quintiles of deprivation from least to most deprived were produced. The index was applied to the analysis of deprivation gradients for mortality and cancer incidence using age-standardised rates. The Relative Index of Inequality (RII) was estimated. Multi-level logistic regression models were used to analyse the association between individual socioeconomic factors and the area deprivation measure on two self-reported health outcomes and mortality. Results - The final index contained five items within the domains of education, employment and living conditions. Premature mortality rates declined between the two periods, however the linear gradient for premature mortality by deprivation quintile, increased. This was most notable for males in the most deprived areas, where the premature mortality rate in the most recent period increased. Results for old-age mortality were less consistent. The relationship between deprivation and cancer incidence was not consistent, with negative and positive gradients observed for different sites. Results for bronchus-lung cancer incidence presented the strongest deprivation gradient. In the multi-level models, a gradient was still seen with increasing deprivation for all health outcomes, though the odds were attenuated when including individual predictors. The strongest relationships were maintained for the most deprived quintile. Discussion - The small area-based deprivation index presented deprivation gradients for mortality and cancer incidence. Even when including individual predictors of socioeconomic status, the area measure still contributed to the analysis of health inequalities. The findings were generally in line with what was found elsewhere. The index developed has wider implications for research, evidence-based policy making and practice in Malta. Conclusion- Area-based indices assign a level of deprivation to an area and not individuals, and thus may be prone to ecological fallacy. Despite this, there is continued value in their use in health inequalities research.
Degree
thesis:*- Level dc:type.qualificationlevel
- PhD
- Grantor dc:publisher.institution
- University of Glasgow
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Galea Gauci, Dorothy
Subjects
dc:subject × 1Rights
- Language dc:language
- en
Identifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.5525/gla.thesis.84253