Income inequality and pandemics: insights from HIV/AIDS and COVID-19—a multicountry observational study

Objectives Assess the relationship between income inequality and HIV incidence, AIDS mortality and COVID-19 mortality. Design Multicountry observational study. Setting 217 countries for HIV/AIDS analysis, 151 countries for COVID-19 analysis. Participants Used three samples of national-level data: a sample of all countries with available data (global sample), a subsample of African countries (African sample) and a subsample excluding African countries (excluding African sample). Main outcome measures HIV incidence rate per 1000 people, AIDS mortality rate per 100 000 people and COVID-19 excess mortality rate per 100 000 people. The Gini index of income inequality was the primary explanatory variable. Results A positive and significant relationship exists between the Gini index of income inequality and HIV incidence across all three samples (p<0.01), with the effect of income inequality on HIV incidence being higher in the African sample than in the rest of the world. Also, a statistically positive association exists for all samples between income inequality and the AIDS mortality rate, as higher income inequality increases AIDS mortality (p<0.01). For COVID-19 excess mortality rate, a positive and statistically significant relationship exists with the Gini index for the entire sample and the excluding African sample (p<0.05), but the African sample alone did not deliver significant results (p<0.1). Conclusion COVID-19 excess deaths, HIV incidence and AIDS mortality are significantly associated with income inequality globally—more unequal countries have a higher HIV incidence, AIDS mortality and COVID-19 excess deaths than their more equal counterparts. Income inequality undercuts effective pandemic response. There is an urgent need for concerted efforts to tackle income inequality and to build pandemic preparedness and responses that are adapted and responsive to highly unequal societies, prioritising income inequality among other social determinants of health.


INTRODUCTION
Pandemics constitute a significant public health problem, posing threats to the health and well-being of substantial population segments across countries, especially marginalised populations. Estimates suggest over 300 excess COVID-19 deaths per 100 000 people as of July 2023 1 for a pandemic that recorded its first mortality in 2020. The UNAIDS data show that AIDS-related deaths have declined by 52% since 2010. Additionally,

WHAT IS ALREADY KNOWN ON THIS TOPIC
⇒ Earlier in the AIDS pandemic, a positive relationship between income inequality and HIV prevalence in sub-Saharan Africa was found. ⇒ The relationship was less clear outside African countries. ⇒ Some single-country studies suggest a link between income inequality and HIV transmission among specific groups. ⇒ Studies primarily from higher income contexts, where data are available, show a positive relationship between income inequality and COVID-19 cases or mortality.

WHAT THIS STUDY ADDS
⇒ Income inequality is globally associated with less effective pandemic response across at least two recent pandemics. ⇒ More unequal countries show higher HIV incidence, AIDS mortality and COVID-19 excess mortality. ⇒ The study highlights that both COVID-19 excess mortality and AIDS mortality are associated with income inequality at a global level for the first time, including low and middle-income countries often considered to have poor-quality data.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
⇒ There is an urgent need to craft more effective pandemic preparedness and response strategies for highly unequal contexts. ⇒ This study highlights the need for global policy responses to tackle income inequality, which is significantly associated with adverse pandemic outcomes. ⇒ Research at the regional or global level should consider income inequality as an essential variable in explaining and attenuating adverse pandemic outcomes.

BMJ Global Health
in 2022, there were 1.3 million new HIV infections, the fewest since the 1980s, with the declines especially strong in regions with the highest HIV burdens. 2 Although marginalised populations and countries in the Global South are disproportionately affected, the reality that pandemics do not respect national boundaries calls for collective global action, paying attention to countries with weak infrastructure and vulnerable health systems. 3 They also necessitate significant solidarity, which is still challenging, as found with COVID-19 vaccines with substantial inequalities in vaccination rates, leaving less wealthy countries behind. 4 Governments and social entities across various sectors need swift and robust responses. Still, these responses often focus heavily on addressing proximate determinants of health, such as individual behaviours. Yet, it is vital to underline the significance of the broader social determinants of health inequalities which influence these behaviours and health outcomes because, as with many pandemics, including the HIV and COVID-19 pandemics, mortality rates in very deprived areas exceeded that in affluent areas. [5][6][7][8] These social determinants comprise broader policy environment and socioeconomic and environmental factors that indirectly or directly impact health by moulding individuals' living, schooling, working life, and ageing choices and conditions that are actionable with effective responses. 9 Despite being less immediately apparent compared with proximate determinants, these social determinants exert considerable influence on health outcomes, often shaping the trajectory of pandemics by affecting disease spread and impact within and across communities. 8 10 Therefore, their inclusion in pandemic responses is crucial to tackling the root causes of health inequalities, as any policy to change health behaviours cannot substantially reduce health inequalities without tackling the underlying causes outside the health sector. 11 Several pathways, often linked to class or power structure (including economic, political, social and cultural) within and between societies, exist on how income inequality generally affects health, health outcomes and health inequalities. [12][13][14][15] In the context of pandemics, a likely causal pathway from income inequality to higher rates of pandemic disease within and between countries is evident. Since the 1990s, a large and robust literature comprising several hundred studies links income inequality to health outcomes. [16][17][18][19][20][21] A systematic review showed that this evidence includes data meeting epidemiological causality criteria. 22 Knowing that pandemics exacerbate income and economic inequalities, 8 this paper highlights three relevant causal pathways from the literature linking income inequality and pandemic infectious diseases. First, inequality can be linked to deprivation among a significant portion of the population in areas from nutrition to education, increasing their vulnerability to infection and disease. In Malawi, for example, higher income inequality was linked to HIV mainly through limited individual choice, higher risk sex and violence, 23 increasing vulnerability among key populations. Second, inequality is linked to social factors limiting effective pandemic response, including lower social cohesion and trust. 22 Third, inequality is linked to political factors undermining health, which makes it harder to coordinate an effective response to HIV and COVID-19, 24 and this weakens the solidarity needed to tackle pandemics.
Increased inequality is hypothesised to be linked to worse pandemic health outcomes, with research highlighting the impact of pandemics on widening inequality 8 and the ways inequality shaped pandemic responses and subjected specific populations to greater risk and lesser protection in certain countries and regions. [25][26][27][28] However, studies showing how income inequality, a critical social determinant of health inequalities, is associated with major recent pandemics on a global scale are lacking. For the HIV pandemic, most of these studies concentrate on sub-Saharan Africa, 29 while COVID-19 studies leave out lower income countries 30 because of claims of data quality for COVID-19 deaths. However, reliable modelled data on COVID-19 excess mortality are now available, providing an avenue to extend the analysis to countries that have been previously excluded. Thus, this paper adds to this literature by assessing the relationship between income inequality and health outcomes (HIV/ AIDS and COVID-19, the two most devastating recent pandemics) globally and regionally. It seeks to answer the research question of whether income inequality, measured using the Gini index, is significantly associated with HIV incidence or AIDS mortality and COVID-19 excess mortality. It also highlights the need for concerted efforts to address income inequality and its detrimental effects on pandemic outcomes.

Study design and data sources
Data used for analysis include AIDS mortality rate per 100 000 people, HIV incidence per 1000 people, COVID-19 excess deaths per 100 000 people, Gini index of income inequality, current health expenditure per capita in US$, World Bank income categories and the UNAIDS regions. The data sources included the World Bank, 31

Statistical analysis
Analytically, this paper assesses the relationship between income inequality (where the Gini index ranged from 0, a case of perfect equality, to 1 for perfect inequality) and HIV incidence, AIDS mortality and COVID-19 excess mortality using the linear regression model. 35 The general model can be written as follows: where H corresponds to the primary health outcomes (ie, HIV incidence per 1000 people, AIDS mortality rate per 100 000 people or COVID-19 excess mortality per 100 000 people) in country i in year t. IN it is the Gini index of income inequality and X is the vector of control variables. For the HIV analysis, H it is replaced with lnH it+1 , the natural logarithm, because this was non-negative, where t+1 signifies that values of the following year were used because it is hypothesised that current income inequality is associated with future health outcomes. β 1 is the coefficient associated with our primary indicator of income inequality (Gini index) in country i in year t . β 2 is a vector of coefficients of several alternative factors in country i in year t that impact health outcomes: the country's income level or category, per capita health expenditure and UNAIDS region. Although economists would argue against allowing income to determine people's access to health services, it is the case that income level, at the country level and between countries, is a critical determinant of health outcomes and differences in health outcomes between countries. 36 Higher income typically provides better access to resources needed for a healthier lifestyle, including quality food, housing, education and healthcare services. It can also mitigate the impact of stressors that can negatively impact health. By controlling for income levels using the World Bank income categories of countries, we can separate the effect of income (wealth) from that of inequality since they are inter-related but distinct factors influencing health outcomes. Health expenditure per capita directly measures the resources allocated for health in a country at an individual level. It captures aspects related to the availability and quality of health services, which are crucial factors in health outcomes. Health expenditure per capita may be associated with inequality and health outcomes, 37 so including it as a covariate prevents omitted variable bias and provides a more accurate estimate of the effect of inequality on health.
The HIV/AIDS models included regional and year fixed effects to control the average differences across regions and years in unobservable predictors. 38 The COVID-19 model contains only regional fixed effects as the complete data included information on COVID-19 deaths between 2020 and 2021 because the latest available data on per capita health expenditure were in 2021, 31 and many countries did not record significant COVID-19 mortality until mid-2020. Africa is most severely affected by HIV and AIDS, 2 39 and the reportedly low COVID-19 mortality in Africa was described as a paradox. 40 So, even though modelled COVID-19 excess mortality data were used 1 in the case of COVID-19 estimations, to avoid skewed regression estimates, three separate regression models were estimated for each dependent variable-a global sample, an Africa-only sample and a sample excluding Africa.
All analyses were done in Stata V.17, 41 and the paper follows the Strengthening the Reporting of Observational Studies in Epidemiology cross-sectional reporting guidelines. 42 Patient and public involvement This study analyses secondary data sets and does not directly involve patients or the public. Although patients were not involved, the findings from the BMJ Global Health paper have been presented to a broader audience from many countries.

Basic and descriptive statistics
Over the period covered in the analysis, the descriptive statistics in

Income inequality and HIV/AIDS outcomes
The regression results in table 2 are for two broad models. The first segment is for the HIV incidence model, while the second is for the AIDS mortality model. The three separate analyses for each model included the global sample, excluding the African countries and only African countries. The results show a positive and significant relationship between income inequality and the natural logarithm of HIV incidence per 1000 people in the following year across all three samples (p<0.01). As predicted, higher levels of income inequality, measured using the Gini index, are significantly associated with an increased incidence rate of HIV per 1000 people in the following year (p<0.01). Loosely speaking, overall and in Africa, higher levels of income inequality in a year are associated with higher HIV incidence in the next year, all things being equal.
Specifically, the effect of income inequality on HIV incidence was higher in the African subsample than in the rest of the world, probably due to higher HIV rates in the sub-Saharan Africa region. Applying Duan's smearing estimator 43 to the results in table 2, a 25 percentage point reduction in the Gini index corresponds to the HIV incidence rate for the next year, significantly dropping by 0.14 per 1000 people for the global sample (p<0.01). For the African subsample, this will significantly reduce the HIV incidence rate by 2.11 per 1000 people in the next year (p<0.01). Similarly, for AIDS mortality, a 25% reduction in the Gini index is associated with a significant decline in AIDS mortality rate by 6.58 (p<0.01), 11.45 (p<0.01) and 17.39 (p<0.01) per 100 000 people in the next year for the global sample, the African sample and the sample excluding Africa, respectively. Figure 1 shows the positive relationship between income inequality and the natural logarithm of HIV incidence per 1000 people, with a steeper slope for the African subsample.
Similar patterns emerge when examining the relationship between income inequality and the AIDS mortality rate per 100 000 people. A positive association between income inequality, as measured by the Gini index, and the natural logarithm of AIDS deaths per 100 000 people (figure 2) exists as higher levels of income inequality in a year are linked to an increase in next year's AIDS mortality rate (p<0.01). Unlike the results for HIV incidence, the effect size was not highest in the African subsample even though it is statistically significant at the 1% significance level. Reducing income inequality (ie, the Gini index by 25%) is significantly associated with a reduction in the AIDS mortality rate by 6.58 (p<0.01) and 11.45 (p<0.01) deaths per 100 000 people in the following year for the entire sample and the African subsample, respectively.

Income inequality and COVID-19
The relationship between income inequality (using the Gini index) and COVID-19 excess mortality rate per 100 000 people, as shown in figure 3, was not initially positive for the entire sample. However, a positive relationship emerged using the regression model presented in table 3.
After controlling health expenditure per capita, regions and income groups, a positive and statistically significant relationship between the Gini index and COVID-19 excess mortality rate per 100 000 people was found for the entire sample and the subsample excluding Africa (p<0.05), as shown in table 3. This result means that more unequal countries tend to report more COVID-19 excess mortality than their more equal counterparts, all other things being equal. The results for the African subsample were not statistically significant (p>0.1), even though a positive relationship was found. This may be partly due to a smaller sample of countries 44 and the near homogeneity in the distribution of COVID-19 excess mortality

BMJ Global Health
for the African subsample in figure 3. A sensitivity analysis was conducted for Africa's subsample by excluding Seychelles' data point with −151.1 COVID-19 excess mortality, and the results remain remarkably similar, still not statistically significant.

DISCUSSION
This paper highlights the significant relationship between income inequality measured using the Gini index and key pandemic variables, including HIV incidence, AIDS mortality and COVID-19 excess mortality. It emerged that unequal societies are burdened more by HIV/AIDS and COVID-19 excess deaths than their more equal counterparts, with the results remaining similar for Africa and the rest of the world. These findings show that income inequality is significantly and adversely associated with pandemic outcomes. The robust association between higher levels of inequality and critical indicators such as HIV incidence per 1000 people, AIDS mortality per 100 000 people and COVID-19 excess mortality suggests income inequality is a factor that may limit effective pandemic response because income inequality creates conditions that perpetuate and exacerbate pandemics, leaving marginalised populations more vulnerable to the devastating consequences of pandemics.

Strengths and weaknesses of this study
A major strength of the analyses in this paper is its multicountry nature, with data from all the UNAIDS regions with available data. Also, the paper uses a standardised measure of income inequality, the Gini index, to assess the relationship between income inequality and pandemic health outcomes. The paper was not limited to one pandemic outcome but considered three outcomes. As a limitation, this paper focused solely on income inequality and its relationship with pandemic outcomes. It did not delve into other essential dimensions of inequality, such as disparities based on gender, race, sexuality and other key population characteristics, which often intersect with income inequality, exacerbating the challenges of marginalised communities and increasing the burden of health disparities. 45 46 Another limitation of the paper is that it did not assess causality, as the results imply association.

BMJ Global Health
But the significance of the relationships in many cases signifies the strength of the association between income inequality and pandemic health outcomes. In turn, the results represent an initial step for future investigations that delve into the robustness of the relationship between inequality and pandemic outcomes.
Although this study was about global phenomena like pandemics, requiring comparable information and data from diverse countries and societies, data availability and quality posed a significant challenge. It is important to note that this analysis cannot discount how disparities in data collection methods, reporting accuracy, and scope across different regions and periods could impact the reliability and comparability of results. For example, the COVID-19 analysis relied on excess mortality data based on epidemiological modelling because of the paucity of COVID-19 mortality data in Africa and Asia, where reported COVID-19 mortality was scanty and not comprehensive. While the HIV/AIDS analyses were based on time series data, the COVID-19 analysis uses almost cross-sectional data, limiting the analysis power. It is undisputed that numerous factors (at the microlevel, mesolevel and macrolevel) other than those controlled in this paper's analyses influence pandemic outcomes. For example, health system capacities, country-level policy responses and heterogeneities, governance quality, cultural behaviours and pre-existing health conditions play substantial roles in determining pandemic responses and outcomes. These variables may interact with income inequality in complex ways, adding complexity to the analysis. These notwithstanding, this paper serves as a starting point to take this research forward to understand how complex factors interact and the pathways through which income inequalities affect pandemic outcomes.

Comparison with other studies
This study breaks new ground on several frontspresenting truly global associations between income inequality and multiple pandemic outcomes, using updated AIDS mortality and COVID-19 excess mortality data to enable wide coverage of low and middle-income countries, and doing so with data accounting for major shifts in recent decades in the AIDS response.
Most studies on HIV from earlier in the pandemic's history did not focus on inequality per se, but rather on poverty and wealth. This focus was particularly driven by surprising evidence from some population surveys suggesting that HIV infection was not consistently higher among the poor; in fact, it was often more prevalent among wealthier individuals 44 47 48 and is higher in Africa's wealthiest countries. 49 Some smaller studies using data from sub-Saharan Africa 29 50-52 show a significant relationship between income inequality and HIV prevalence, and occasionally transmission. Income inequality, for example, has been linked to the spread of HIV among women in Malawi at the neighbourhood and district levels 23 and persons who inject drugs in Vietnam at the community level. 53 In a smaller area setting, New York City in the USA, the relationship between income inequality and adverse HIV health outcomes has been reported. 45 In contrast, a study in Asia (one of the only analyses outside Africa) found no association between income and HIV prevalence. 48 Most use HIV prevalence, a less meaningful measure we would argue than HIV incidence. Meanwhile, there is a paucity of literature on the income inequality and AIDS mortality nexus. There is evidence of higher risk of mortality among persons of low socioeconomic status in South Africa and the USA, 54 55 but to the best of our knowledge, there is no literature that discusses the relationship between income inequality and AIDS mortality, particularly at the global level. In addition, studies have not considered the most recent decade of data, during a time a significant change in the AIDS pandemic. Our paper uses the most recent data to bridge these gaps. This paper demonstrates a statistically significant relationship between inequality and both new HIV infections and AIDS deaths through to the present time and does so beyond a country or region to present evidence at a global level.
While the COVID-19 pandemic could exacerbate income and economic inequality, 56 a significant positive relationship between income inequality and COVID-19 cases or mortality has also been reported. 30 46 57 58 Most studies have focused on a smaller subset of countries including Organisation for Economic Co-operation and Development member countries where data are robust. These findings largely align those reported in this paper for COVID-19 excess mortality. Notably, there is a dearth of studies using COVID-19 excess mortality, considered to reflect better the impact of COVID-19 deaths over and above the expected death rates within countries. 59 In particular, many low and middle-income countries have been excluded from previous studies assessing the association between COVID-19 health outcomes and income inequality mainly because of data paucity. Although the relationship between income inequality and COVID-19 excess mortality was not significant for Africa, the overall similarity of this paper's findings regarding the role of income inequality in aggravating the effects of the COVID-19 pandemic is critical to note.

Implications of findings
The correlation between income inequality and other social variables demonstrates the ways that inequality undermines pandemic response. More unequal countries are likely to have higher HIV and COVID-19 rates both because of unequal access to essential health services and other relevant services outside the health sector, including education and shelter. 23 Beyond reporting the association between income inequality and pandemic outcomes, which is crucial, it is imperative to understand the mechanisms more deeply through which income inequality influences health outcomes amidst a pandemic. For the AIDS pandemic, the means of influence likely operate on multiple levels. At a social and political level, widening income inequality may foster greater BMJ Global Health othering in pandemic response both within and between countries, which undermines universal responses and the deprioritisation of services, disproportionately serving a disfavoured portion of the population, as has been shown in some contexts. 60 As those with good healthcare access, living in well-off communities, get pre-exposure prophylaxis through private means, for instance, HIV rates for such communities may fall rapidly. At the same time, the political priority for reaching the rest diminishes. At an individual level, income levels and inequality may reduce choices for preventing HIV transmission-for example, by pushing more people into higher HIV risk or transactional sex. In other words, the relationship between income inequality and pandemics means that despite the widespread availability of HIV testing, prevention and treatment options, societies with greater inequality have struggled to use these resources to reduce rates of HIV infection and AIDS-related deaths effectively.
Meanwhile, the overall health of people, which has been concretely tied to various social determinants, 23 likely makes people in unequal societies more vulnerable to HIV mortality and less able to access good HIV care and cutting edge medicines. 61 Similar mechanisms likely apply to the COVID-19 pandemic. The COVID-19 pandemic presented a unique challenge as it led to many drastic measures undertaken by governments. Wealthier countries with more resources could provide relief and better policy responses than their less affluent counterparts. 3 The relationship between COVID-19 excess mortality and income inequality likely means that more unequal countries were challenged to ensure access to preventive measures, testing and treatment, including vaccination, 62 leading to elevated COVID-19 excess mortality rates in these countries relative to less unequal countries. In the subsample of African countries, the insignificant relationship between income inequality and COVID-19 excess deaths may not be surprising due to the smaller number of observations and relatively low variability in the Gini index of income inequality compared with the subsample that excludes African countries. This further supports inequality's role in exacerbating COVID-19 excess mortality, especially in a sample of unequal countries. Indeed, the data seem to suggest that economic inequality is a more significant driving force than net wealth.
Based on the findings of this study, to effectively respond to future health crises, it is crucial to prioritise policies and interventions that aim to reduce income inequality. But pandemics will not wait for changes in economic distribution. As such, it is crucial to consider how pandemic responses in highly unequal societies could act to counter, rather than reinforce, those inequalities. Although the results from this study show that income inequality is bad for effective pandemic responses using HIV/AIDS and COVID-19 pandemics, as noted in the limitations of this paper, focusing on income inequalities is just one aspect of broader social inequality. Future research should incorporate additional dimensions of inequality to fully understand the complex dynamics of pandemics and their societal impacts. This comprehensive approach will help develop more effective and equitable pandemic responses, promote social justice and improve health outcomes across diverse segments of society. By taking a holistic approach, we can build a more resilient and equitable society better equipped to mitigate the devastating effects of future pandemics.

CONCLUSION
Income inequality significantly correlates with HIV incidence, AIDS mortality rate and COVID-19 excess mortality rate using a cross-country analysis involving the UNAIDS regions. The findings reported in this paper underscore the urgent need for concerted efforts to tackle income inequality and its detrimental effects on pandemic outcomes and to craft pandemic responses that work more effectively in highly unequal contextscountering rather than reinforcing inequality. We argue that this sits at the heart of the global call to leave no population behind in attaining development goals.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Patient consent for publication Not applicable.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available in a public, open access repository. All data used in this paper are available in the public domain from different sources. HIV/AIDS data are available from the UNAIDS (https://aidsinfo. unaids.org/), COVID-19 data from The Economist (https://www.economist.com/ graphic-detail/coronavirus-excess-deaths-estimates), health expenditure data from the World Bank (https://data.worldbank.org/indicator/) or WHO, and the Gini index of income inequality from the World Inequality Database (https://wid.world/data/).

Open access
This is an open access article distributed in accordance with the Creative Commons Attribution 4.0 Unported (CC BY 4.0) license, which permits others to copy, redistribute, remix, transform and build upon this work for any purpose, provided the original work is properly cited, a link to the licence is given, and indication of whether changes were made. See: https://creativecommons.org/ licenses/by/4.0/.