Area-level social and structural inequalities determine mortality related to COVID-19 diagnosis in Ontario, Canada: a population-based explanatory modeling study of 11.8 million people

  1. Linwei Wang
  2. Andrew Calzavara
  3. Stefan Baral
  4. Janet Smylie
  5. Adrienne K. Chan
  6. Beate Sander
  7. Peter C. Austin
  8. Jeffrey C. Kwong
  9. Sharmistha Mishra

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Importance

Social determinants of health (SDOH) play an important role in COVID-19 outcomes. More research is needed to quantify this relationship and understand the underlying mechanisms.

Objectives

To examine differential patterns in COVID-19-related mortality by area-level SDOH accounting for confounders; and to compare these patterns to those for non-COVID-19 mortality, and COVID-19 case fatality (COVID-19-related death among those diagnosed).

Design, setting, and participants

Population-based retrospective cohort study including all community living individuals aged 20 years or older residing in Ontario, Canada, as of March 1, 2020 who were followed through to March 2, 2021.

Exposure

SDOH variables derived from the 2016 Canada Census at the dissemination area-level including: median household income; educational attainment; proportion of essential workers, racialized groups, recent immigrants, apartment buildings, and high-density housing; and average household size.

Main outcomes and measures

COVID-19-related death was defined as death within 30 days following, or 7 days prior to a positive SARS-CoV-2 test. Cause-specific hazard models were employed to examine the associations between SDOH and COVID-19-related mortality, treating non-COVID-19 mortality as a competing risk.

Results

Of 11,810,255 individuals included, 3,880 (0.03%) died related to COVID-19 and 88,107 (0.75%) died without a positive test. After accounting for demographics, baseline health, and other SDOH, the following SDOH were associated with increased hazard of COVID-19-related death (hazard ratios [95% confidence intervals]) comparing the most to least vulnerable group): lower income (1.30[1.09-1.54]), lower educational attainment (1.27[1.10-1.47]), higher proportion essential workers (1.28[1.10-1.50]), higher proportion racialized groups (1.42[1.16-1.73]), higher proportion apartment buildings (1.25[1.11-1.41]), and larger vs. medium household size (1.30[1.13-1.48]). In comparison, areas with higher proportion racialized groups were associated with a lower hazard of non-COVID-19 mortality (0.88[0.85-0.92]). With the exception of income, SDOH were not independently associated with COVID-19 case fatality.

Conclusions and relevance

Area-level social and structural inequalities determine COVID-19-related mortality after accounting for individual demographic and clinical factors. COVID-19 has reversed the pattern of lower non-COVID-19 mortality by racialized groups. Pandemic responses should include prioritized and community-tailored intervention strategies to address SDOH that mechanistically underpin disproportionate acquisition and transmission risks and shape barriers to the reach of, and access to prevention interventions.

Key points

Question

Are area-level social determinants of health factors independently associated with coronavirus disease 2019 (COVID-19)-related mortality after accounting for demographics and clinical factors?

Findings

In this population-based cohort study including 11.8 million adults residing in Ontario, Canada and 3,880 COVID-19-related death occurred between Mar 1, 2020 and Mar 2, 2021, we found that areas characterized by lower SES (including lower income, lower educational attainment, and higher proportion essential workers), greater ethnic diversity, more apartment buildings, and larger vs. medium household size were associated with increased hazard of COVID-19-related mortality compared to their counterparts, even after accounting for individual-level demographics, baseline health, and other area-level SDOH.

Meaning

Pandemic responses should include prioritized and community-tailored intervention strategies to address SDOH that mechanistically underpin inequalities in acquisition and transmission risks, and in the reach of, and access to prevention interventions.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2022.03.14.22272368: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    EthicsIRB: Data use in this project was authorized under Section 45 of Ontario’s Personal Health Information Protection Act, which does not require review by a Research Ethics Board. Outcomes: Our primary outcome of interest was COVID-19-related death, defined as death within 30 days following, or 7 days prior to a positive SARS-CoV-2 test.
    Sex as a biological variablenot detected.
    Randomizationnot detected.
    Blindingnot detected.
    Power Analysisnot detected.

    Table 2: Resources

    Software and Algorithms
    SentencesResources
    SARS-CoV-2 test result and date were determined based on records in Ontario Laboratory databases, and the COVID-19 surveillance system (Case and Contact Management System (CCM)).
    Ontario Laboratory
    suggested: None

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    Limitations include the potential for misclassification biases due to lack of data on the cause of death. Based on the Ontario COIVD-19 surveillance data, 92% of recorded all-cause death among individuals diagnosed with COVID-19 occurred within 30 days following or 7 days prior to a positive test (Appendix-Figure-2). Other settings have adopted similar definition of COVID-19-related death to capture the immediate impact of COVID-19 on death(44). Individuals who do not have a health card number were not captured; and if they were more likely to be socially and structurally vulnerable, our estimates might have under-estimated the inequalities. We were restricted to area-level SDOH measures in the absence of individual-level SDOH measures, leading to the potential for residual confounders by SDOH. However, our findings were similar in pattern to that of other studies using individual-level SDOH measures(3, 15), suggesting a lower risk of ecological fallacy. Almost all areas with the highest quintile proportion racialized groups were urban areas. Therefore, our estimated associations between racialized groups and COVID-19-related mortality are subject to residual confounding by individual’s residence in rural vs urban areas. However, stratified analysis by rural/urban revealed that structural inequalities in COVID-19-related mortality by racialized groups were present in both settings (Appendix-Table-5). We did not examine the potential modifying effect by age or region, nor the ...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    Results from scite Reference Check: We found no unreliable references.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2022.03.14.22272368v1 on medRxiv