Discussion
Our study evidences the extent of stark levels of absolute and relative inequality, using range and distributional measures, in Scotland in 2019, prior to the COVID-19 pandemic. The overall disease burden was double that in the most deprived areas in Scotland, largely driven by inequalities in mortality. We found that the gap in the burden of disease is so wide that the fatal burden of disease (YLL) in the most deprived areas was higher than the total burden of disease (DALYs) in the least deprived areas. We already know from life expectancy figures that people in the most deprived communities die at younger age. They also spend more than a third of their lives in poor health compared with the least deprived communities.4 5 Our findings further strengthen our understanding of the cause-specific diseases driving the interaction between stalling mortality improvements, healthy life expectancy and health inequalities.26 27
Many of the leading causes of disease burden in 2019—heart disease, drug use disorders, lung cancer and chronic obstructive pulmonary disease (COPD)—were also the leading drivers of absolute and relative inequalities in the disease burden. Causes such as: alcohol, and drug use disorders; liver cirrhosis; COPD; lung cancer; and, self-harm and suicide, exhibited the largest relative inequalities in disease burden which is consistent with results published elsewhere in the literature.2 13 15 The absolute and relative inequalities in the burden of disease in Scotland from these causes underline that the health of the population, and health inequalities within the population, are shaped by wider determinants of health such as economic, social, and environmental factors and not solely due to individual health behaviours.28 29 On the other hand, there were health conditions where there were smaller, to no, inequalities, for example, sensory conditions, headaches and some musculoskeletal conditions indicating that burden does not vary across deprivation groups for these causes. Triangulating these results with estimates of inequality-attributable DALY estimates for COVID-19 indicates that although COVID-19 infection has likely increased all-cause inequalities, there are several other causes of disease which have a much larger impact in driving levels of all-cause inequality in the burden of disease.8
The COVID-19 pandemic has caused almost unprecedented change across health, education and the economy in Scotland and across the world. COVID-19 has been shown to have a substantial impact on the population health in Scotland; however, this impact was not shared equally across areas experiencing different levels of socioeconomic deprivation; in Scotland, the marked inequalities in COVID-19 YLL reported in 2020 were further exacerbated in 2021, to the extent that approximately half of COVID-19 YLL was attributable to inequalities in area deprivation. This widening of inequalities in 2021 was confirmed across all measures of absolute and relative inequality.7 8 Furthermore, health problems that existed before the COVID pandemic, have not gone away, and in many cases have been exacerbated through the impact of the economic lockdown and the unintended consequences of the policy responses to the pandemic, and unless future harm is prevented, this trend will continue.30 31 Integrating this impact, quantitively, is difficult, and trends may not be comparable due to the major disruptions, and changes in access to, and delivery of, services, and the competition between causes of mortality.32 These inequalities have emerged through the syndemic nature of COVID-19—as it interacts with and exacerbates existing social inequalities in the disease burden and the social determinants of health33—prior to the pandemic, improvements in life expectancy and healthy life expectancy in Scotland had stalled since 2012, with a slowdown in the overall progress of reducing mortality and widening of socioeconomic inequalities in mortality.5 34–37 The results presented in this paper are, therefore, likely to be representative of wider pre pandemic trends given that YLD does not change quickly over time and that YLL estimates ultimately drive changes in the overall DALYs. Our estimates help us to understand the state of prepandemic inequalities, providing important baseline positions that can be triangulated with robust, and locally relevant, emerging evidence, to assess the likely impact of the pandemic, and other public health crisis such as the current cost-of-living crisis, on non-COVID and COVID-19-related health inequalities.
Strengths and limitations
This study has several important strengths. First, this study assesses the impact of inequalities using DALYs for over 70 individual diseases, conditions and injuries. DALYs provide a composite, internally consistent measure of population health loss which can be used to evaluate the proportionate burden of different diseases and injuries and compare population health by geographical region and over time. By combining information on fatal and non-fatal burden, BOD studies allow planners and policy-makers to have a better understanding of the contribution that different diseases and injuries make to the total burden of disease and how this burden varies by levels of deprivation. This in turn supports decisions about where prevention and service activity should be focused to address health inequalities in Scotland.
YLD estimates are derived from surveys, and routine administrative data and as such may not be fully representative of the underlying burden by deprivation decile, for example, administrative data are reflective of demand rather than need. This is also relevant for severity distributions, which are internationally derived with cancers being the main exception. In terms of impact of study results by deprivation decile, on the absolute scale this is most relevant for the causes with the largest YLD estimates, or those where YLD is the main contributor to disease-specific DALYs. For those causes which have a larger YLL component, it is plausible that any biases in YLD from unmet need or survey-specific bases will have a smaller impact on the DALY estimate if this unmet need causes disease-specific excess mortality.
Our measure of deprivation is area-based rather than individual measure and is, therefore, likely to misclassify many individuals into categories that do not reflect their individual experiences.38 Although the SIMD includes health indicators in its range of domains (thereby raising the theoretical possibility of reverse causality whereby people are ordered by the health outcome rather than socioeconomic deprivation), the employment-income deprivation index (ie, excluding the health measures) is very highly correlated with the overall SIMD index and so this is unlikely to have changed the results.39 40 Furthermore, the SII and RII have the advantage that they are based on data about the whole population, rather than just the extremes, and so take into account inequalities across the entire distribution of inequality. They do, however, require a reasonably linear relationship between outcomes and exposures.41
Some burden of disease studies attempt to propagate the uncertainty around DALY estimates by estimating uncertainty intervals, which combine sampling error and non-sampling errors arising from study design. This approach was developed to avoid misleading users over how precise confidence intervals were, given they only capture the extent of sampling error. However, other approaches have been developed to give a sense of how uncertain an estimate might be, such as developing a scoring system over how certain, or otherwise, an estimate may be.42 In this context, the results which we have found for neck and low back pain and headaches have the highest levels of uncertainty when using the uncertainty measured estimated via the SBOD study.
