On health inequality in the past and present

Qiao, Siqi (2025) On health inequality in the past and present. PhD thesis, University of Glasgow.

Abstract

This thesis explores socioeconomic gradients in health outcomes through evidence from the past and present. Studying health inequality is fundamental to revealing how socioeconomic factors create unfair variations in health outcomes across populations, which is esssential for developing targeted interventions to promote equal opportunity for well-being. Historical analysis provides crucial insights into the long-run implications of socioeconomic inequalities on health, with particular significance during the mortality transition—a period characterised by unprecedented improvements in population health alongside profound transitions that established enduring patterns of health disparities that persisted for generations. Chapters 1 and 2 examine the past by analysing mortality dynamics in London and its constituent areas from the 1850s to the 1950s. Chapter 3 investigates present-day health inequality by focusing on a key health risk factor—physical (in)activity.

The first two chapters draw on novel datasets I constructed by digitising and compiling information and data from historical administrative records, including Medical Officer of Health reports, censuses, and Registrar General’s reports. The final chapter leverages the Understanding Society survey, a nationally representative panel dataset that captures comprehensive and nuanced measures of socioeconomic status across UK households.

Chapter 1 utilises a new time series dataset of London mortality spanning 1841–1964, encompassing four mortality measures: crude mortality, corrected death rate (adjusted for age and sex composition), infant mortality, and non-infant mortality. The annual, long-term, and continuous mortality data illuminate an S-shaped decline curve, characterised by an initial period of slow reductions before faster reductions and ultimately a slow-down in improvements. This pattern is observed across all mortality measures, enabling analysis through a four-parameter logistic model. This formal statistical method yields key parameters that capture the underlying dynamics effectively. The inflection point, occurring around 1896 for crude mortality, indicates the point of the most rapid mortality reduction. To explain the sigmoid pattern, an economic model is developed where higher socioeconomic groups adopt health technologies earlier, thereby experiencing mortality transitions earlier. The model underscores how socioeconomic inequality shapes the sigmoid dynamics of mortality decline.

Chapter 2 expands the analysis to constituent areas in London. The new panel data-set comprises three fundamental components: first, geographic areas whose boundaries remained stable for a long enough period; second, three mortality measures: crude mortality, infant mortality, and non-infant mortality for these areas; third, measures of socioeconomic inequality for these areas, including, for example, the proportion of professional occupations and middle-class residents. Application of the four-parameter logistic model reveals substantial spatial heterogeneity in mortality dynamics, with inflection points varying by decades across areas. The economic model developed in Chapter 1 predicts that this variation stems from socioeconomic differences—areas with more high socioeconomic residents adopted health technologies earlier, facilitating earlier mortality transitions. Empirical testing through an augmented logistic model confirms this hypothesis, demonstrating a significant negative relationship between an area’s socioeconomic status and the inflection point of the mortality dynamics. For example, areas that had a 1% higher proportion of individuals in professional classes experienced an inflection point in mortality declines around a year and a half earlier. These findings indicate that the mortality transition proceeded through a prolonged period of mortality divergence across socioeconomic groups before eventual convergence, rather than immediate convergence across social groups.

Taken together, the first two chapters provide the first formal examination of how socioeconomic inequality shaped the sigmoid pattern of mortality decline and generated the “divergence-convergence” pattern of mortality inequality during the period of mortality transition. This historical analysis yields twofold implications: it reveals how uneven access to health innovations historically perpetuated and amplified health disparities in today’s developed economies, whilst providing an analytical framework for contemporary developing economies where ongoing epidemiological transitions, absent policies addressing socioeconomic inequalities, risk exacerbating existing health inequalities.

Shifting focus from historical London to contemporary UK, Chapter 3 investigates the distinct mechanisms through which education and household income shape physical activity (PA) patterns among working-age adults in the UK from 2015–2019, using Understanding Society survey data. Employing an ordered logit model to distinguish three PA levels—zero physical activity, engagement in at least some physical activity, and meeting WHO-recommended guidelines—the study yields three principal findings, contributing to the understudied aspects of the relationship between socioeconomic status and physical activity behaviour. First, while both education and income positively influence PA, their relative magnitudes differ substantially—the effect of upgrading from GCSE to degree-level education equals that of a four-and-a-half-fold increase in household disposable income, which suggests that educational interventions may be more effective at promoting physical activity than income support policies. Second, household disposable income has approximately twice the impact of household gross labour earnings on physical activity engagement, highlighting the necessity of using correct income measures and the effectiveness of redistributive policies. Third, women’s physical activity is more strongly associated with educational attainment, while household income demonstrates a greater influence on men’s activity patterns.

Drawing from this thesis, socioeconomic inequality functions as a crucial determinant of health outcomes, operating both through contemporaneous mechanisms and by shaping the long-run dynamics of health disparities. These findings underscore the importance of examining the persistent effects of socioeconomic conditions across temporal horizons to fully understand their implications for population health.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Subjects: H Social Sciences > HA Statistics; H Social Sciences > HB Economic Theory; H Social Sciences > HM Sociology
• Colleges/Schools: College of Social Sciences > Adam Smith Business School > Economics
• Supervisor's Name: Angelopoulos, Professor Konstantinos and Mancy, Dr Rebecca
• Date of Award: 2025
• Depositing User: Theses Team
• Unique ID: glathesis:2025-85085
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 23 Apr 2025 13:21
• Last Modified: 23 Apr 2025 13:30
• Thesis DOI: 10.5525/gla.thesis.85085
• URI: https://theses.gla.ac.uk/id/eprint/85085