key: cord-102515-ch6prsf3
authors: Moran, Elizabeth; Kubale, John; Noppert, Grace; Malosh, Ryan E; Zelner, Jon L
title: Inequality in acute respiratory infection outcomes in the United States: A review of the literature and its implications for public health policy and practice.
date: 2020-04-26
journal: nan
DOI: 10.1101/2020.04.22.20069781
sha: 
doc_id: 102515
cord_uid: ch6prsf3

Seasonal and pandemic respiratory viruses such as influenza and the novel coronavirus (SARS-COV-2) currently sweeping the globe have often been described as 'equal opportunity infectors', implying little socioeconomic disparity in susceptibility. However, early data from the COVID-19 pandemic has underscored that the burden of respiratory viruses actually reflect and magnify existing socioeconomic inequalities. We review the literature on socioeconomic and racial disparities in acute respiratory infection (ARI), as well as ARI-associated hospitalization and mortality. Our goal is to identify key principles of the relationship between socioeconomic inequality and ARI outcomes, as well as highlighting poorly understood areas that need to be addressed by research and policy in the wake of the COVID-19 pandemic. We find that there has been descriptive work in this area, but that there is a distinct lack of cohesive methodology in the literature exploring social determinants and ARI. We propose the fundamental cause theory is a useful framework for guiding future research of disparities in ARI and for the design of interventions to alleviate these disparities.

Acute respiratory infections (ARI) cause substantial morbidity and mortality worldwide 1 , both as the result of seasonal epidemics and pandemics, such as COVID- 19 . Respiratory viruses such as influenza and respiratory syncytial virus (RSV), have often been described as 'equal opportunity infectors', implying little socioeconomic disparity in susceptibility. However, early data from the COVID-19 pandemic has underscored that the burden of respiratory viruses actually reflect and magnify existing socioeconomic inequalities 2 . These trends have been made clear in the United States by the alarming disparities in the toll of severe disease and mortality experienced by African-Americans 3 . At the population level, viral and immune factors are necessary pre-conditions for the emergence and transmission of pathogens causing seasonal and pandemic ARI 4 . At the same time, socioeconomic inequalities are clearly key drivers of exposure, severe disease, and mortality. The rapid pace of transmission and mortality resulting from the COVID-19 pandemic has shone a bright light on these disparities. But they are no less acute in the context of other causes of seasonal and pandemic ARI, despite attracting less attention. In this paper, we review the literature on socioeconomic and racial disparities in ARI infection, hospitalization, and mortality in the United States. By focusing on the existing literature on influenza, RSV, and all-cause ARI we hope to identify key principles of the relationship between socioeconomic inequality and ARI outcomes, as well as highlighting poorly understood areas that need to be addressed by research and policy in the wake of the COVID-19 pandemic.

Disparities in ARI outcomes may result from 1) differential rates of exposure owing to occupation, housing, and other factors, 2) differential susceptibility to infection upon exposure, owing to both social and medical factors, e.g. access to vaccination where it is available, and 3) differential ARI-towards a more broadly protective influenza vaccine, increased use of antivirals to improve outcomes, and advances in genomics and mathematical modeling to understand transmission, can lead to significant strides against influenza and other endemic and emerging viral ARIs. However, such advances in population-level protection are often accompanied by a widening of disparities in outcomes, as new preventive measures and therapies become available first to the well-off. As recent discussions of the ability to practice social distancing, and whether a COVID-19 vaccine will become broadly available shows, these issues remain woefully unaddressed, and have significant implications for public health.

We searched PubMed for articles documenting socioeconomic disparities in ARI infection/disease risk, severe outcomes, and mortality. First, we attempted to capture articles investigating disparities in all-cause ARI, influenza-like illness (ILI), and RSV. Then, we identified articles documenting disparities in influenza infection risk and outcomes. We separated these searches to ensure each was sensitive to key differences in the literature, most notably around access to and uptake of influenza vaccination.

We used the PubMed special query for health disparities in conjunction with search terms relating to all-cause ARI, RSV, and influenza ( Figure 1 ). The health disparities query identifies articles evaluating disparities in health outcomes and healthcare access with inequities in dimensions of race/ethnicity, SES, gender identity and sexual orientation, insurance status, and other populations described as "vulnerable." We then screened studies using title and abstract review to identify articles with a specific focus on all-cause ARI, ILI, RSV, and influenza. Additional articles and reviews were added to our results upon recommendation from topic-area experts. Articles were not excluded based on

year of publication. Full search queries are available in the supplementary materials.

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04. 22.20069781 doi: medRxiv preprint

Our all-cause ARI and RSV search returned 155 papers, of which 9 studies met the inclusion criteria. Of these, only 4 of the retained articles used data from the US. The influenza search returned 386 results, of which 27 were deemed relevant following title and abstract review. Of these, 12 addressed seasonal influenza in the US, with the remaining 15 focused on the 2009 H1N1 influenza pandemic. All selected articles, organized by etiology, dimension of disparity assessed, and measured outcome (incidence, severity, death, etc.) are listed in Table 1 .

A wide range of SES-related disparities in all-cause ARI were identified in studies from the US 14, 15, 19 , UK 16, 17, 20 , and Argentina 18 . Two studies reported increased ARI incidence 14, 15 among those with lower self-reported social status 14 and SES 15 . Race/ethnicity was also found to be associated with increased severity, as indicated by two studies reporting increased ARI-related hospitalization in a US study of children under 5 using data from New Vaccine Surveillance Network (NVSN) 19 and primary care seeking in UK using sentinel surveillance data 20 . SES disparities were also evident in all-cause ARI-related complications: two studies reported increased hospital admission for ARI 17 and bronchiolitis 16 and increased rates of medical intervention due to bronchiolitis 16 

Townsend deprivation score, index used to quantify material deprivation based on car and house ownership, household overcrowding, and employment, with a higher score showing more material deprivation 16, 17 . Finally, one study reported increased under-5 mortality due to ARI among those with proxy measures for poverty including having an adolescent mother, household crowding, and lacking running water in home 18 .

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04. 22.20069781 doi: medRxiv preprint Incidence: In a multi-state cohort study, female healthcare workers from Texas, Washington, and Oregon, with lower self-reported social status reported higher rates of ARI symptoms 14 . In an experimental study, higher subjective SES was protective against the development of cold symptoms after exposure to rhinovirus or influenza virus 15 .

Severity: Disparities in severity of ARI were indicated using hospitalization and care seeking rates in a variety of study designs. Using population-based surveillance data, the NVSN study found increased hospitalization rates for all-cause ARI among Black children as compared to white children; however, these results were not adjusted for SES 19 . Additional surveillance data from the Royal College of General Practitioners Research and Surveillance Centre in England found racial and ethnic minorities were more likely to seek care for cold/flu symptoms than whites 20 . In the UK, a case-control study matching children under 1 year old admitted to Sheffield hospital with clinical suspicion of bronchiolitis to healthy community controls identified from birth records found that children living in the two most deprived electoral wards, as defined using the Townsend index, were 1.5 times as likely to be admitted to the hospital and 1.7 times as likely to require a medical intervention than children living in other parts of the city 16 . An ecological study in the West Midlands Region of the UK found neighborhoods with higher Townsend scores of deprivation experienced higher rates of childhood hospital admissions due to ARI and pneumonia 17 .

Mortality: A case-control study in Buenos Aires found death due to ARI in children under 5 was associated with living in a crowded household, having an adolescent mother, and lacking running water in the home 18 . No other papers discussed mortality due to all-cause ARI.

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. 

Studies of disparities in seasonal influenza outcomes have focused primarily on the relationship between neighborhood SES and rates of hospitalization for severe disease in [8] [9] [10] [11] [12] with fewer studies of individual-level outcomes 21 . In addition, studies have investigated inequalities in influenza-related healthcare utilization (e.g. visits to outpatient clinics) 21, 22 , incidence 23 , and exposure 24 as they relate to SES, race/ethnicity 9, 19, 22 , and socioemotional stress 23 . SES and exposure indicators utilized in these studies include percentage of households living in poverty 8,10-12 , female headed households 9,10 , household crowding 9, 11 , and neighborhood population density 10 .

Incidence: Two papers described disparities in rates of influenza susceptibility and exposure in older African Americans 24 and college students with higher perceived stress 23 . Black Americans in nursing homes were less likely to be vaccinated against seasonal influenza, and were less likely to have vaccinated contacts, due to racial and socioeconomic segregation in nursing home care. The authors of this study argue that the results from nursing homes may provide clues to residential segregation's effects on seasonal influenza risk at the population level 24 . In a cohort study of U.S. college students, an increase in perceived stress score was associated with a 25% greater rate of self-reported ILI, after adjustment for demographic factors, behaviors and flu vaccination 23 .

Severity: Hospitalization was the most frequently studied dimension of severity used to characterize influenza disparities, followed by overall healthcare utilization 21, 22 . Area-level poverty and race/ethnicity were used as predictors of risk in a number of studies examining disparities in rates of influenza-related hospitalization. Two ecological studies found that living in neighborhoods with a high percentage of residents below the poverty line and in female-headed households experienced higher rates of influenza . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. 8, 9 . Similarly, census-tract rates of influenza-associated hospitalization surrounding Nashville, Tennessee, increased with population density and the percentage living in poverty, female headed/crowded households 10 . In New Haven, Connecticut, the annual incidence of pediatric influenzaassociated hospitalizations in high-crowding census tracts (defined as >5% of households with > one occupant per bedroom), and high-poverty census tracts (defined as >20% of households below the poverty line), was > 3 times higher than in low-crowding/low-poverty census tracts 11, 12 . This disparity remained even after adjustment for comorbid medical conditions and influenza vaccination 11, 12 . In a study of individual-level outcomes using data from FluServ-NET, across multiple US counties and influenza seasons, African American and Latino adults had increased odds of influenza-related hospitalization than whites 9 . One study using population-based surveillance data found significantly higher hospitalization rates for laboratory-confirmed influenza for Black children than for white children 19 .

Viral and bacterial pneumonias are common ARI-related complications that reflect SES-based disparities in baseline health status as well as access to care. Several studies identified by our search focused specifically on disparities in hospitalization 13, 17, 25, 26 and mortality 27 associated with diagnosis of pneumonia and influenza (P&I).

Severity: In a study examining socio-demographic factors increasing risk of P&I hospitalization in elderly populations using Medicare data from 1991-2004, counties with high proportions of live-in grandparent caregivers and lower median income experienced higher rates of pneumonia-related hospitalization 26 . In a study of Alaska Natives using hospitalization records from 2000-2004, regions . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04. 22.20069781 doi: medRxiv preprint with a lower proportion of households with in-home water service, indicating higher poverty rates, also had higher hospitalization rates for P&I 13 .

Mortality: There is some evidence that disparities in P&I associated mortality have decreased over time.

One study using age-adjusted mortality rates for P&I show a decreasing ratio and absolute difference between Black and white Americans from 1950 to 2000 27 . The authors attribute this reduction in disparity due to widely available treatment to guard against P&I complications and mortality 27 .

Literature examining RSV-specific disparities was limited to hospitalization/severity. Disparities were identified in dimensions of race/ethnicity 19, 28, 29 , household socioeconomic status 13, 30 , high population density 30 , maternal unemployment 31 , younger maternal age 32 , and household crowding 33 . A study using population-based surveillance data found that hospitalization rates for laboratory confirmed RSV were similar for Black and white children under 12 months of age, but for children > 12 months, hospitalization rates for lab-confirmed RSV were higher for Black children than for white children 19 . A study utilizing surveillance and hospital admittance data in Southwestern Alaska identified villages with lower proportion of houses plumbed water, higher proportion of household crowding, and higher proportion of families living below the poverty line as risk factors for hospital admission for RSV 30 .

Similar risk factors were also identified in non-US studies 292831 32 .

Pandemics of influenza and other viral respiratory pathogens, such as COVID-19, are characterized by higher overall attack rates, elevated morbidity and mortality, unevenly adopted countermeasures, and increased strain on health systems. All of which may magnify inequalities in 1 0 patterns of disease, despite a complete or near-complete lack of protective immunity. Because of this lack of acquired immunity in the population, the age distribution of infection is likely to be different than seasonal epidemics, resulting in differential patterns of healthcare utilization and mortality risk.

Nevertheless, in a pattern echoing findings for seasonal flu, RSV and all-cause ARI, inequalities in the most severe ARI pandemic outcomes have consistently been documented. For example, historical data from Chicago illustrated disparities in 1918 influenza pandemic mortality by neighborhood SES and racial composition 34, 35 , with some of this disparity explained by differential rates of transmission in low-SES, overcrowded neighborhoods 34 . In 2009, a number of studies documented social race/ethnic outcome disparities [36] [37] [38] [39] [40] [41] , with lower SES 41 associated with increased exposure risk 36 , as well as overall incidence 37 , hospitalization 38, 39, 41 , complications 36 , and death 39,40 due to pandemic H1N1.

Incidence: A nationally representative survey from the U.S. measured risk factors for exposure during the 2009 pandemic, including living in an apartment building, relying on public transportation, and difficulty finding day care that was separate from other children. This study found that these risks were significantly more common among non-white participants 36 . Another US study identified a lack of access to sick-leave and increased number of children in the household as risks for higher ILI incidence during the H1N1 pandemic 37 . The authors determined that this increased risk was associated with Hispanic ethnicity even after controlling for education and income 37 , suggesting that additional, unmeasured SES-related risk factors contributed to this disparity.

Severity: In a study using surveillance data from Illinois, hospitalization rates were twice as high for Hispanics and Blacks compared to whites during the 2009 H1N1 pandemic 38 . An analysis using data from multiple nationally representative surveillance networks investigated disparities in hospitalization . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04. 22.20069781 doi: medRxiv preprint and death found substantial differences in morbidity and mortality between race/ethnic groups 39 . Ageadjusted hospitalization rates for minorities were two times higher than those for non-Hispanic whites.

Mortality: One study using nationally representative surveillance networks found non-Hispanic Black and Hispanic children had disproportionately higher mortality than non-Hispanic whites 39 . During the H1N1 pandemic, American Indians and Alaskan Natives had a mortality rate 4 times higher than all other racial/ethnic groups 40 . Conversely, a study using hospital discharge and census data found the living in a higher poverty census tract (>6% below poverty line) and Hispanic ethnicity was associated with lower risk of ICU admission for H1N1 related hospitalizations in Massachusetts 41 , although it is unclear if this result reflects differential severity or disparities in access to care.

Outside the United States: Research outside of the United States has identified similar relationships between inequality and pandemic ARI outcomes. For example, in a case-control study in Ontario, Canada of persons hospitalized for ARI during the 2009 H1N1 pandemic, the odds of H1N1 infection were higher among adults identifying as East/Southeast Asian, South Asian, and Black compared to whites 42 . In Spain, individuals belonging to an ethnic minority group were more than twice as likely than whites to be admitted to the hospital, and individuals with a secondary or higher education were 0.54 times less likely to be admitted to the hospital as individuals with less education 43 . In England, the 2009 H1N1 pandemic caused 3.1 and 2.0 times higher mortality rates in the two most socio-economic deprived quintiles compared to the least deprived quintile, and this disparity persisted after adjustment for underlying medical conditions and age 44 .

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. The COVID-19 pandemic has shown in stark relief the ways in which SES and racism structure ARI outcomes at every level from exposure to disease severity and mortality. However, despite the large toll of mortality exacted by seasonal and pandemic ARI and the clear evidence of disparities, the number of studies investigating disparities in ARI outcomes and incidence is too limited to help chart a way towards more equitable outcomes. Notably, most of the analyses discussed in this review come from data collected by studies in which identifying disparities was not a primary goal, and thus was not explicitly incorporated into study design and the enrollment of participants. Further, much of the previous work has documented disparities, for example by race/ethnicity, but has not explored the history of policies and practices that specifically marginalized these racial/ethnic minorities and led to disparate outcomes 45 . This echoes earlier findings that studies focusing on identifying and addressing social determinants are underrepresented relative to the importance of social factors in structuring infectious disease risk 46 .

The lack of studies dedicated to identifying and ameliorating ARI disparities guarantees that the analyses reviewed here provide a decidedly incomplete picture of the drivers of disparities in ARIrelated outcomes. For example, many of the studies identified in our search is that they characterized disparities among the population of cases presenting at a point of care, ranging from routine medical visits to hospitalization. This reliance on clinical data is likely to under-count risk in populations without access to care, who are disproportionately likely to have low SES, and/or come from marginalized groups, e.g. undocumented immigrants, for whom real and perceived overlaps between medical and legal systems may serve as a disincentive to care-seeking. Finally, most ARI cohort studies, which allow for granular examination of individual-level risk and protective factors, drew largely on geographically concentrated study populations that are largely homogeneous in terms of SES and race/ethnicity. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. 

It is evident from our review that the lack of a coherent theoretical framework for understanding relationships between social inequality and ARI incidence, severity, and mortality is a significant factor in the dearth of research in this area. The evidence turned up by our review, as well as the issues of racially and socioeconomic disparate exposure risks, coupled with unequal access to testing and treatment made clear by the COVID-19 pandemic, demonstrate that the principles underlying Link & Phelan's theory of SES and race as fundamental causes of health and illness apply to ARIs as much as many non-communicable diseases as well as infections more classically understood has having social antecedents, such as tuberculosis and diarrheal disease. Fundamental cause theory (FCT) posits SES drives disparities through its impact on the financial and social power to marshal the material and social resources that are protective of health 47 . FCT is a mainstay of the literature on SES-related disparities in non-communicable diseases, but has enjoyed very limited application to infectious diseases 48 . A key tenet of FCT is that these disparities are persistent and are likely increase even as population-level risk goes down. For example, medical innovation is likely to increase disparities because new interventions typically become available first to the well-off and well-insured. This strongly suggests that progress in the treatment and prevention of ARIs using antivirals, monoclonal antibodies 49 , influenza vaccination, and molecular genotyping to inform surveillance and control, are likely to result in increased inequality in both ARI incidence and severe outcomes. These issues have once again been front-and-center during the COVID-19 pandemic, as well-off individuals in the United States enjoyed preferential access to testing in the early days of the pandemic. Similarly, the ability to participate in 'social distancing' by working from home, having food and other essential delivered to one's home, and engage other protective measures are clearly a function of power and material resources. In the absence of effective vaccines and antivirals, this is the key technology available to protect oneself from an emerging . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04. 22.20069781 doi: medRxiv preprint infection. As preventive and therapeutic interventions against COVID-19 become available, we are likely to see these also distributed preferentially to those with greater means, thereby only exacerbating the yawning disparities in exposure, severity, and mortality that are already evident.

Surveillance and large network studies can estimate burden of disease and influenza vaccine effectiveness (VE), but may under enroll low-SES and non-white populations and are by design limited to those with access to care since participants are enrolled at the point of care. Ideally, large longitudinal cohort studies with study samples that are representative of the US population would be available to estimate differences in incidence and risk and VE. A meta-analysis found that 97% of observational studies on influenza severity were cohort studies, but very few of these studies collected information on race/ethnicity and the authors do not report on measures of SES in their analysis 55 . To capture the components of risk associated with social disparities, these studies would ideally span multiple seasons and account for known factors associated with risk such as age, influenza infection and vaccination history, household composition, crowding, and other potential exposure and susceptibility related factors. In the absence of such studies, greater attention is needed to increase the socioeconomic diversity of the populations that studies of seasonal ARI risk and VE draw upon, as well as the potential impact of SES on infection outcomes and vaccine effectiveness. Where such data are not available, careful statistical modeling is essential. For example, approaches such as multiple regression with poststratification (MRP), which can be used to make unbiased population-level predictions from nonrepresentative data, may help address some of these questions while we wait for more detailed data 56 .

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04. 22.20069781 doi: medRxiv preprint FCT highlights why identifying and intervening only on mediators of the relationship between SES and disease is unlikely to be effective: risk of ARI infection, severe disease, and mortality cannot be reduced to the impact of a single variable, e.g. vaccination. Instead, it reflects a spectrum beginning with comorbid health conditions, the ability to avoid exposure, access to preventive and therapeutic treatment, and other factors. These issues are arguably even more acute for infectious diseases than for non-communicable ones, as the risks of our friends and family directly impact our own risks via transmission.

An FCT perspective suggests that effective policies must address inequality in access to prevention and treatment, as well as inequity in the ability to avoid exposure. For example, policies limiting or eliminating paid sick leave have been shown to increase rates of influenza transmission in the workplace, and to disproportionately affect racial/ethnic minorities 37 , with similar patterns evident in the COVID-19 pandemic. In order to be able to reduce transmission, symptomatic persons should be treated with antivirals (when applicable) and practice home isolation 57 . This cannot be achieved without policies that support rather than penalize such behavior. The lack of supportive workplace policies has been suggested to result in a population-attributable risk of 5 million additional cases of ILI in the US with a disproportionate burden on Hispanic Americans 37 . The lack of a health system ensuring equal access to prevention and care is clear a key factor in exacerbating SES-related disparities in seasonal ARI incidence, as well as undermining pandemic preparedness. However, in the absence of sweeping changes that would address the root causes of these disparities , targeted vaccination for influenza 58, 59 and RSV 60 based on age specific contact and transmission patterns has been theorized to improve both the efficacy of vaccines. Identifying high priority groups for vaccination has been shown to be effective in reducing disease burden when target groups are those with high-risk of infection or high transmission potential for both RSV and influenza [58] [59] [60] [61] . The expansion of targeted programs to improve access to . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. vaccination among those of lower SES could be beneficial in reducing disparities. When identifying groups for targeted vaccination, especially those of disadvantaged backgrounds, it is important to address past abuses that may lead to mistrust of these programs. Finding ways to engage with organizations that serve low SES and racial minority communities to address concern and reduce vaccine hesitancy will be paramount to the success of these programs 36 

In sum, our findings highlight wide disparities in ARI outcomes by socioeconomic status and race/ethnicity in the U.S. Though the COVID-19 pandemic has brought many of these concerns to the forefront of the public and scientific consciousness, it has also shown how poorly developed our set of tools for addressing both the root and proximal causes of these disparities is. This study represents a first step towards the development a coherent framework for identifying disparities in ARI and their causal antecedents. This will hopefully increase clarity around when and where socially focused interventions (i.e. healthcare and workplace policies) should be prioritized, as well as how biomedical innovations (e.g. vaccination, antivirals) should be distributed to shrink group-level disparities at the same time that they minimize population risk.

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04.22.20069781 doi: medRxiv preprint 1 7 To accomplish this, a sustained effort to build studies and create standardized measures purposively aimed at understanding and curbing SES-related disparities in ARIs is urgently needed.

However, even with this lack of consensus, the body of evidence suggests there are disparities in ARI incidence, severity, and mortality. With potential new technologies on the horizon to prevent and better treat ARIs from RSV to influenza as well as emerging infections with pandemic potential, such as COVID-19, the time to make a plan for how to ensure the equitable distribution of benefits from these and other technologies is now.

. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. Regions with lower proportion (<80%) of inhome water service had a 1.9 times higher rate of hospitalization due to pneumonia and influenza and a . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. Māori or Pacific ethnicity was associated with over 3 times the risk of . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. Living in a census tract with higher poverty levels (6.0-11.9% and 16% or more vs. 0-5.9% households living below poverty) and Hispanic ethnicity was associated with lower risk of ICU admission in H1N1 related hospitalizations. When stratified by race, living in a census tract with higher poverty levels (6.0-11.9% vs. 0-5.9% households living below poverty) was associated with lower risk of ICU admission . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. . CC-BY-NC 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 26, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 26, 2020. . https://doi.org/10.1101/2020.04. 22.20069781 doi: medRxiv preprint 

Global, regional, and national causes of under-5 mortality in 2000-15: an updated systematic analysis with implications for the Sustainable Development Goals

Why inequality could spread COVID-19

African Americans are at higher risk of death from coronavirus. The Washington Post

Drivers of airborne human-to-human pathogen transmission

Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a systematic review and modelling study

Global burden of respiratory infections due to seasonal influenza in young children: a systematic review and meta-analysis

Reasons for the Decline of Mortality in England and Wales during the Nineteenth Century

Influenza-Related Hospitalizations and Poverty Levels -United States

Social determinants of influenza hospitalization in the United States. Influenza Other Respi Viruses

Socioeconomic Disparities and Influenza Hospitalization

Neighborhood socioeconomic status and influenza hospitalizations among children

Influenza-related hospitalization of adults associated with low census tract socioeconomic status and female sex

The Relationship Between In-Home Water Service and the Risk of Respiratory Tract , Skin , and Gastrointestinal Tract Infections Among Rural Alaska Natives

Subjective social status predicts wintertime febrile acute respiratory illness among women healthcare personnel

Objective and Subjective Socioeconomic Status and Susceptibility to the Common Cold

Deprivation and bronchiolitis

Social deprivation and hospital admission for respiratory infection: Ecological study

Mortality Associated With Acute Respiratory Infections Among Children at Home

Disparities Between Black and White Children in Hospitalizations Associated With Acute Respiratory Illness and Laboratory-confirmed Influenza and Respiratory Syncytial Virus in 3 US Counties -2002 -2009

RCGP Research and Surveillance Centre Annual Report 2014-2015: disparities in presentations to primary care

Socio-economic disparities in the burden of seasonal influenza: The effect of social and material deprivation on rates of influenza infection

Increased influenza-related healthcare utilization by residents of an urban aboriginal community

Stress, adherence to preventive measures for reducing influenza transmission and influenza-like illness

Health Care Segregation and Race Disparities in Infectious Disease

A spatial analysis of the determinants of pneumonia and influenza hospitalizations in Ontario (1992-2001)

Disparities in influenza mortality and transmission related to sociodemographic factors within Chicago in the pandemic of 1918

Pandemic influenza and socioeconomic disparities: Lessons from 1918

Racial disparities in exposure, susceptibility, and access to health care in the US H1N1 influenza pandemic

The Impact of Workplace Policies and Other Social Factors on Self-Reported Influenza-Like Illness Incidence During the 2009 H1N1 Pandemic

Disparities among 2009 pandemic influenza A (H1N1) hospital admissions: A mixed methods analysis -Illinois

Racial and Ethnic Disparities in Hospitalizations and Deaths Associated with 2009 Pandemic Influenza A (H1N1) Virus Infections in the United States

Deaths related to 2009 pandemic influenza A (H1N1) among American Indian/Alaska Natives 12 states

Effect of race/ethnicity and socioeconomic status on pandemic H1N1-related outcomes in Massachusetts

Ethnic disparities in acquiring 2009 pandemic H1N1 influenza: a case-control study

Social factors related to the clinical severity of influenza cases in Spain during the A (H1N1) 2009 virus pandemic

Socio-economic disparities in mortality due to pandemic influenza in England

Neighborhood-Level Mass Incarceration and Future Preterm Birth Risk among African American Women

Analyses of infectious disease patterns and drivers largely lack insights from social epidemiology: contemporary patterns and future opportunities

Social Conditions as Fundamental Causes of Health Inequalities: Theory, Evidence, and Policy Implications

Contemporary Social Disparities in TB Infection and Disease in the USA: a Review

Cost-effectiveness of Palivizumab for Respiratory Syncytial Virus: A systematic review

Results From the US Hospitalized Adult Influenza Vaccine Effectiveness Network (HAIVEN)

Measurement of Vaccine Direct Effects Under the Test-Negative Design Measurement of Vaccine Direct Effects Under the Test-Negative Design

Influenza Vaccination Coverage --United States

Interim Estimates of 2017 -18 Seasonal Influenza Vaccine Effectiveness -United States

Interim Estimates of 2016 -17 Seasonal Influenza Vaccine Effectiveness -United States

Populations at risk for severe or complicated influenza illness: A systematic review and meta-analysis

Struggles with Survey Weighting and Regression Modeling

The Impact of Illness on Social Networks: Implications for Transmission and Control of Influenza

On the relative role of different age groups in influenza epidemics

OR "medically underserved area"[MeSH Terms] OR minority group[TIAB] OR "minority groups

OR Alaska Natives[TIAB] OR people of color[TIAB] OR "poverty areas"[MeSH Terms] OR poverty area[TIAB] OR poverty areas[TIAB] OR "rural health"[MeSH Terms] OR rural health[TIAB] OR "rural health services"[MeSH Terms] OR "rural population"[MeSH Terms] OR rural population[TIAB] OR rural populations[TIAB] OR slum[TIAB] OR slums[TIAB] OR "urban health"[MeSH Terms] OR "urban health services"[MeSH Terms] OR "urban population

OR "minority health

AND ((((human influenza[MeSH Terms]) OR human influenzas[MeSH Terms]) OR human influenza[Title/Abstract]) OR human influenzas