Disparities in SARS-CoV-2 Testing in Massachusetts During the COVID-19 Pandemic

Early deficiencies in testing capacity have contributed to poor control of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2),¹ particularly among minority (ie, Black and Latino/Latina) and socioeconomically vulnerable communities.^2,3 Allocating testing resources to locations of greatest need is important to mitigate subsequent waves of coronavirus disease 2019 (COVID-19).⁴ In the context of improved SARS-CoV-2 testing infrastructure, we examine the alignment of testing to epidemic intensity in Massachusetts.

This cohort study was designated as not human subjects research by the Mass General Brigham institutional review board because it used anonymous, publicly available data; thus, informed consent was not sought. This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

We compiled weekly SARS-CoV-2 molecular testing data from the Massachusetts Department of Public Health and Boston Public Health Commission for the period May 27 to October 14, 2020, following the initial COVID-19 wave. The Boston Public Health Commission reported tests of unique Boston residents, whereas the Massachusetts Department of Public Health reported total tests, including repeat testing of individuals. Consequently, we performed separate analyses for Massachusetts (351 cities and towns) and for Boston (15 neighborhoods).

We defined testing intensity as the number of SARS-CoV-2 tests performed weekly per 100 000 population and epidemic intensity as weekly test positivity. We considered optimal alignment of testing resources to be matching community ranks of testing and positivity. In communities with a testing gap (ie, the testing rank was lower than the positivity rank) in a given week, the gap was calculated as additional testing required to achieve matching ranks. For example, the testing gap for a community with the third highest positivity is the difference between its testing rate and that of the community with the third highest testing intensity.

Responses from the American Community Survey (2014-2018)⁵ were aggregated to characterize communities. Negative binomial models using robust sandwich estimators to account for repeated measures at the community level were fit to assess associations of the magnitude of the weekly testing gap with time (linearly by week), selected Centers for Disease Control and Prevention Social Vulnerability Index⁶ domains (eg, Socioeconomic Status and Minority Status/Language), and large university student population (>10% of residents). Owing to collinearity, the model of Boston neighborhoods only assessed associations with time and socioeconomic vulnerability. Two-sided Wald tests were used to assess significance at a threshold of P < .05. Data analysis was performed using R statistical software version 3.6.1 (R Project for Statistical Computing).

During the observation period, 4 262 000 tests were reported. COVID-19 incidence (median [range], 339 [0-6670] cases per 100 000) and testing intensity (median [range], 41 000 [5350-274 000] tests per 100 000) varied considerably between communities, with observed increased testing in less socioeconomically vulnerable localities, vacation regions, and areas near universities. Considerable overlap was observed between communities with the highest socioeconomic vulnerability and those with the largest testing gaps (Figure 1).

In a multivariable model of statewide testing, the relative testing gap increased by 9.0% per week (adjusted rate ratio [aRR], 1.09; 95% CI, 1.08-1.10; P < .001) (Figure 2). Increasing levels of socioeconomic vulnerability were associated with increased testing gaps (aRR, 1.35 per quartile; 95% CI, 1.23-1.49; P < .001). Communities with the highest quartile of minority status or language vulnerability had larger testing gaps after accounting for socioeconomic status, but the difference was not significant (aRR, 1.46; 95% CI, 0.96-2.23; P = .08). The presence of a large university student population was associated with decreased testing gaps (aRR, 0.21; 95% CI, 0.12-0.38; P < .001). Similar findings were observed within Boston, with increasing testing gaps (aRR, 1.08 per week; 95% CI, 1.03-1.13; P = .003) and larger gaps in more socioeconomically vulnerable neighborhoods (aRR, 2.51 per quartile increase; 95% CI, 1.56-4.03; P < .001).

These analyses indicate that, despite programs to promote equity and enhance epidemic control in socioeconomically vulnerable communities, testing resources across Massachusetts have been disproportionately allocated to more affluent communities. On the basis of this analysis, we do not know how much of the testing gap is due to underutilization of testing, which may be associated with fear, anticipated stigma, and/or loss of employment. Additional limitations include the use of test positivity with varying rates of asymptomatic testing, model assumption of independence between communities, and aggregation at the community level that may underestimate disparities. Worsening structural inequities in SARS-CoV-2 testing increase the risk of another intense wave of COVID-19 in Massachusetts, particularly among socioeconomically vulnerable communities.

Accepted for Publication: December 21, 2020.

Published: February 9, 2021. doi:10.1001/jamanetworkopen.2020.37067

Correction: This article was corrected on April 23, 2021, to fix an error in Figure 2.

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Dryden-Peterson S et al. JAMA Network Open.

Corresponding Author: Scott Dryden-Peterson, MD, MSc, Division of Infectious Diseases, Brigham and Women’s Hospital, 75 Francis St, Boston, MA 02115 (sldrydenpeterson@bwh.harvard.edu).

Author Contributions: Dr Dryden-Peterson had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: All authors.

Acquisition, analysis, or interpretation of data: Dryden-Peterson, Velásquez, Stopka.

Drafting of the manuscript: Dryden-Peterson, Velásquez, Davey, Lockman.

Critical revision of the manuscript for important intellectual content: Dryden-Peterson, Velásquez, Stopka, Ojikutu.

Statistical analysis: Dryden-Peterson.

Administrative, technical, or material support: Stopka, Ojikutu.

Conflict of Interest Disclosures: Dr Dryden-Peterson reported receiving grants from the National Institutes of Health during the conduct of the study. No other disclosures were reported.

Funding/Support: The work was supported by the Harvard University Center for AIDS Research, a funded program of the National Institutes of Health (grant P30 AI060354), the National Institute of Allergy and Infectious Diseases (grant K08 AI141740), the National Cancer Institute (grant R01 CA236546), the Dr Lynne Reid/Drs Eleanor and Miles Shore Fellowship at Harvard Medical School, and the Burke Global Health Fellowship at the Harvard Global Health Institute.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Disclaimer: The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the National Institutes of Health or the institutions with which the authors are affiliated.