Skip to main content
Download PDF
Download ePub

Minority health social vulnerability index and long COVID illness among a statewide, population-based study of adults with polymerase chain reaction-confirmed SARS-CoV-2

Archives of Public Health volume 83, Article number: 64 (2025) Cite this article

Abstract

Background

The COVID-19 pandemic has disproportionately affected socially vulnerable communities. Some individuals experience persistent symptoms and conditions of COVID-19 illness known as long COVID. As little research has examined how social vulnerability is related to long COVID, we studied this topic using Minority Health Social Vulnerability Index (MHSVI), specifically created for the COVID-19 pandemic in the U.S.

Methods

We merged county-level MHSVI data with population-based data of Michigan adults with PCR-confirmed SARS-CoV-2 infection between March 2020 and May 2022 based on respondents’ county of residence. We examined the relationship between county-level MHSVI (binary: high social vulnerability ≥ 75th percentile) and two long COVID measurements, assessed a median of 18.8 months after their initial infection: (1) ongoing long COVID (yes/no) and (2) long COVID diagnosis (yes/no). We conducted modified Poisson regression models with robust standard errors to estimate prevalence ratio (PR) between associations of MHSVI and long COVID overall and by six MHSVI themes (socioeconomic status, household composition/disability, minority/language, housing type/transportation, healthcare access, medical vulnerability), adjusting for individual-level and county-level covariates.

Results

Living in high MHSVI counties was not associated with ongoing long COVID or long COVID diagnosis. However, the associations differed by theme of MHSVI: respondents in highly socially vulnerable counties assessed by medical vulnerability had 1.32 times higher prevalence of long COVID diagnosis (95% CI:1.12 − 1.57). There were no statistically significant associations in other themes after the adjustment for covariates.

Conclusions

Our findings suggest the importance of upstream social determinants of health during public health emergencies and provide evidence that medically vulnerable communities need additional public health resources to cope with long COVID among their residents.

Peer Review reports

Text box 1. Contributions to the literature

• There is limited public health research on how area-level social vulnerability is related to long COVID.

• This study used population-based data of Michigan adults with PCR-confirmed COVID-19 to find that living in medically vulnerable areas was associated with long COVID diagnosis.

• Public health authorities should provide medically vulnerable communities with more resources and support.

Introduction

The COVID-19 pandemic has disproportionately affected communities that are economically and socially vulnerable [1, 2]. At the beginning of the pandemic, there were wide disparities in access to COVID-19 testing by geographic area, particularly in racial and ethnic minoritized communities, rural areas, and areas with low-income residents [3, 4]. Moreover, vulnerable communities with higher levels of poverty, household crowding, racial and ethnic minoritized populations, and economic segregation were more likely to have higher rates of COVID-19 cases, hospitalizations, and deaths compared to better resourced communities throughout the pandemic [3, 5,6,7,8]. The disproportionate burden of the COVID-19 pandemic among socially vulnerable communities has been explained in part by structural inequities, such as access to high-quality education and employment, safe working and neighborhood environments, transportation infrastructure, health insurance, and healthcare facilities [9].

The disproportionate burden has also been observed for post-acute sequelae of SARS-CoV-2 infection (PASC), which is also known as “long COVID”. Long COVID is a post-viral condition, defined as experiencing persistent signs, symptoms, and conditions of COVID-19 illness after the onset of an acute SARS CoV-2 infection [10]. Approximately, 15% of U.S. adults reported they have ever experienced long COVID according to the U.S. Household Pulse Survey (2022–2023) [11]. Long COVID symptoms may affect multiple organ systems [10] and limit an individual’s daily activity, psychophysical and occupational performance, and social roles [12,13,14,15]. Previous studies have highlighted that older age, being female, having lower educational attainment, living in rural areas, experiencing severe acute COVID-19 illness, having pre-existing health conditions, and lack of vaccination are risk factors for long COVID [16,17,18,19,20,21]. As these individual-level risk factors for developing long COVID are closely related to structural inequalities that could be measured at the area level [9, 16,17,18,19,20,21], it is necessary to study the relationship between area-level social vulnerability and long COVID.

Several studies have demonstrated that social vulnerability is related to COVID-19-related health outcomes using area-level measures such as the Area Deprivation Index or the Social Vulnerability Index (SVI). Communities with higher social vulnerability are more likely to have higher COVID-19 incidence [5, 9, 22], hospitalization for COVID-19 [23], new mobility disability after COVID-19 diagnosis [24], COVID-19 mortality rates [2, 8, 25], and COVID-19 vaccination hesitancy [26]. However, surprisingly little research has examined how social vulnerability is related to long COVID [27]. Individuals living in socially vulnerable communities, where COVID-19 incidence rates are higher [5, 9, 22] vaccination rates are lower [26] and high-quality healthcare resources are limited [9], may be more likely to experience long COVID compared to socially advantaged communities.

The current study examined the association between county-level social vulnerability and individual-level long COVID by merging data on the Minority Health Social Vulnerability Index (MHSVI) and with population-based data of adults with polymerase chain reaction (PCR)-confirmed SARS-CoV-2 infection between March 2020 and May 2022 in Michigan. We also studied the relationship between social vulnerability and long COVID by each of the six themes of MHSVI: (1) socioeconomic status, (2) household composition and disability status, (3) racial and ethnic minority status and language, (4) housing type and transportation, (5) healthcare infrastructure and access, and (6) medical vulnerability [24, 28]. We hypothesized that individuals residing in counties with a high level of MHSVI would have a higher prevalence of long COVID compared to individuals residing in counties with a low-to-moderate level of MHSVI. Our work extends existing knowledge about the social determinants of health and health equity and provides evidence about the relationship between social vulnerabilities and long COVID.

Methods

Individual-level data

Michigan COVID-19 Recovery Surveillance Study (MI CReSS) is a statewide representative survey of adults 18 years and older with a PCR-confirmed SARS-CoV-2 test in the Michigan Disease Surveillance System (MDSS). A stratified probability sample of eligible adults was selected from 13 geographic strata, including six public health emergency preparedness regions (1, 3, 5, 6, 7, and 8) [29], six counties in southeast Michigan (Macomb, Oakland, Saint Clair, Monroe, Washtenaw, and Wayne [except Detroit]), and one city (Detroit). Sixteen sequential cross-sectional samples were drawn over time with a base number of 50 − 70 individuals from each geographic region, while the remainders of the sample were drawn proportionally based on overall case counts within each area.

Non-institutionalized adults were eligible for the sampling frame if they: (1) had a PCR-confirmed SARS-CoV-2 infection between March 2020 and May 2022, (2) were alive at the time of baseline survey, and (3) had a valid phone number and zip code or county information in the MDSS. Respondents completed the questionnaire either (1) online in English or (2) over the phone with an interviewer in English, Spanish, or Arabic. We did not include adults with SARS-CoV-2 infection based on at-home antigen tests. Respondents completed baseline surveys between June 2020 and December 2022, and follow-up surveys between January 2022 and November 2023. The median time from COVID-19 illness onset to baseline survey was 4.4 months (IQR = 3.4–5.7 months) and the median time from COVID-19 illness onset to follow-up survey was 18.4 months (IQR = 14.9–21.4 months). A total of 5,521 adults completed the baseline survey for a response rate of 32.1%, and 4,100 adults completed the follow-up survey for a response rate of 80.5% (American Association for Public Opinion Research response rate #6) [30]. All respondents provided consent to participate. The University of Michigan Institutional Review Board deemed this study exempt due to the use of secondary de-identified data.

Our analysis included respondents who completed both baseline and follow-up surveys. Of the 4,100 follow-up surveys completed, we excluded surveys missing information about counties (n = 157), outcome variables (n = 62), or covariates (n = 92). We further excluded ten surveys completed by proxy respondents due to mental capacity concerns (n = 7) or some other reasons (n = 1) at follow-up, leading to an analytic sample of n = 3,781 (Fig. 1).

Fig. 1
figure 1

Flowchart for the unweighted analytic sample, Michigan COVID-19 Recovery Surveillance Study, 2020–2023. a The numerator of the response rate includes both partial and complete surveys

Full size image

Outcome variable

We used self-reported measurements of long COVID during the follow-up survey, a median of 18.8 months after respondents’ initial infection using two items: (1) ongoing long COVID and (2) long COVID diagnosis. Prior to asking about long COVID, we provided the definition of long COVID: “The next set of questions ask you about potential long-term symptoms you experienced during your COVID-19 illness and your experiences seeking care for those symptoms. Persistent symptoms of COVID-19 are commonly referred to as Long COVID or Chronic COVID. For this section, we will refer to this condition as Long COVID.” Ongoing long COVID was assessed using two questions. Respondents were asked “At any point since your diagnosis with COVID-19, have you experienced long COVID?” If a respondent answered positively to the question, we also asked, “Have you recovered from long COVID to your usual state of health?” We created a binary variable of ongoing long COVID, which equals 1 if respondents had not recovered from long COVID to their usual state of health, and 0 if respondents had not experienced long COVID or had recovered from long COVD. We assessed a long COVID diagnosis using the question, “Has your doctor or other health professionals told you that you have long COVID, or that you were experiencing long-term symptoms of COVID-19?”

Social vulnerability data

We obtained county-level publicly available data on the MHSVI in 2020 from the U.S. Department of Health and Human Services Office of Minority Health (OMH) website (https://minorityhealth.hhs.gov/minority-health-svi). The MHSVI was specifically developed for the COVID-19 pandemic to improve existing resources to support the identification of racial and ethnic minoritized communities at highest risk for the disproportionate impact and adverse outcomes due to the pandemic [28]. Therefore, the MHSVI is a more comprehensive index to study COVID-19-related resources and outcomes compared to the standard SVI. Using 5-year estimates of demographic data from the U.S. Census Bureau’s American Community Survey 2016–2020, the MHSVI estimates the relative vulnerability of each county in the U.S. by subsuming 34 social factors in six themes: socioeconomic status, household composition and disability status, racial and ethnic minority status and language, housing type and transportation, healthcare infrastructure and access, and medical vulnerability [28, 31]. For each county, the 34 social factors were ranked and assigned a percentile-rank value ranging from 0 to 1, with higher percentiles indicating higher social vulnerability. Then, the score for each theme was obtained by summing and ranking the percentile values. The scores of six themes were summed and ranked to generate the overall MHSVI ranging from 0 to 1, with a higher score indicating higher social vulnerability. Additional details on MHSVI variable selection and methods are available elsewhere [31]. We merged the MHSVI data with data from the MI CReSS based on addresses recorded for each respondent in the MDSS at time of COVID-19 illness onset. Approximately 92% of respondents lived in the same county at baseline and follow-up, and 4.4% lived in different counties. The remaining 3.6% of respondents had missing information on counties either at baseline or follow-up and were excluded from the analysis. Then, we created a binary variable of the MHSVI using a cut-point of ≥ 75th percentile to indicate high social vulnerability overall and by each theme based on previous literature [24, 32,33,34,35,36,37]. Additionally, we used quintiles of the MHSVI as a sensitivity analysis (Q1 = lowest [reference], Q5 = highest) [38].

Covariates

We included individual-level sociodemographic factors at follow-up survey as covariates: age group (18 − 34, 35 − 54, 55 − 64, ≥ 65), sex at birth (male, female), race and ethnicity (Hispanic, non-Hispanic White, non-Hispanic Black, another non-Hispanic race or ethnicity), education (high school or less, some college, college graduate), household income (<$35,000, $35,000 − 74,999, ≥$75,000), and health insurance status (private insurance, Medicare/Medicaid/another type, none). We adjusted for survey-related factors including mode of completion (phone or online) and pandemic phase based on when the respondent was diagnosed with COVID-19 (phase 1: March 2020–September 2020, phase 2: October 2020–February 2021, phase 3: March 2021–September 2021, phase 4: October 2021–May 2022). We also added county-level covariates: population size from the 2020 U.S. decennial Census and rural-urban classification from the 2023 U.S. Department of Agriculture-Economic Research Service [9].

Statistical analysis

First, we calculated weighted descriptive statistics to characterize the analytic sample. Next, we calculated weighted prevalence estimates of ongoing long COVID and long COVID diagnosis by MHSVI and covariates. Statistical differences were evaluated with Pearson chi-square test with Rao’s correction. Then, we conducted unadjusted and adjusted modified Poisson regression models with robust standard errors to estimate the prevalence ratio (PR) for associations between a county-level binary variable of MHSVI and individual-level ongoing long COVID and long COVID diagnosis. The analyses were completed for overall MHSVI as well as each theme of MHSVI (socioeconomic status, household composition and disability status, racial and ethnic minority status and language, housing type and transportation, healthcare infrastructure and access, and medical vulnerability). For the sensitivity analysis, we conducted the same regression models using quintiles of MHSVI. All statistical analyses were completed using Stata, version 17.0. All estimates were weighted to account for nonresponse at baseline and attrition at follow-up [39]. For household income, missing information was imputed using the weighted sequential hot-deck method under a missing at random assumption [40]. We further estimated robust standard errors by clustering respondents within counties.

Results

Among our study respondents, 17.4% reported they had ongoing long COVID and 11.2% reported they had received a diagnosis of long COVID (Table 1). About 30.6% of respondents reported living in highly socially vulnerable areas. Respondents were predominately <55 years, female, non-Hispanic White, college graduate, had a household income ≥$75,000, insured, and lived in counties with ≥700,000 population and classified as urban.

Table 1 Characteristics of study participants, Michigan COVID-19 recovery surveillance study, 2022–2023 (n = 3,781)
Full size table

The prevalence of ongoing long COVID was 17.2% among respondents residing in highly socially vulnerable counties and 17.5% among respondents residing in the low-to-moderately socially vulnerable counties, which was not statistically different (p = 0.882) (Table 2). The prevalence of long COVID diagnosis was 12.2% among respondents in highly socially vulnerable counties, higher than the prevalence among respondents in low-to-moderately socially vulnerable counties (10.7%), but the difference was not statistically significant (p = 0.264). There were statistical differences in the weighted prevalence of both ongoing long COVID and long COVID diagnosis across age, sex, race and ethnicity, education, household income, and health insurance status categories. The prevalence of both ongoing long COVID and long COVID diagnosis was highest among respondents who were aged 55–64, female, non-Hispanic Black, had some college education, had a household income <$35,000, and had Medicare/Medicaid/another type of health insurance. The prevalence of ongoing long COVID was highest among respondents living in counties with a population of less than 180,000, and the prevalence of long COVID diagnosis was higher among respondents living in rural areas than those living in urban areas.

Table 2 Weighted prevalence of ongoing long COVID and long COVID diagnosis by minority health social vulnerability index and covariate among study participants, Michigan COVID-19 recovery surveillance study, 2022–2023 (n = 3,781)
Full size table

In unadjusted regression models, living in highly socially vulnerable counties was not associated with the prevalence of ongoing long COVID compared to living in low-to-moderately socially vulnerable counties (PR: 0.98, 95% confidence interval [CI]: 0.78–1.24) (Table 3). There continued to be no association between living in a highly socially vulnerable county and ongoing long COVID after adjustment for individual-level and county-level covariates (aPR: 0.94, 95% CI: 0.79–1.13). Moreover, residing in highly socially vulnerable counties, compared to residing in low-to-moderately socially vulnerable counties, was not associated with a prevalence of long COVID diagnosis in the unadjusted model (PR: 1.14, 95% CI: 0.91–1.43) or the adjusted model (aPR: 1.15, 95% CI: 0.93–1.42).

Table 3 Associations of minority health social vulnerability index with ongoing long COVID and long COVID diagnosis, Michigan COVID-19 recovery surveillance study, 2022–2023 (n = 3,781)
Full size table

The associations, however, differed by theme of MHSVI (Table 4). The medical vulnerability theme presented robust results: respondents in highly socially vulnerable counties had 1.44 times higher prevalence of long COVID diagnosis (95% CI: 1.29–1.60) in the unadjusted model, which was slightly attenuated to 1.32 in the adjusted model but remained statistically significant (95% CI: 1.12–1.57). In other themes, there were no statistically significant associations between living in highly socially vulnerable counties and long COVID measurements after the adjustment for individual-level and county-level covariates. In the socioeconomic status theme, respondents residing in highly socially vulnerable counties had 1.15 times higher prevalence of ongoing long COVID (95% CI: 1.00–1.31) and 1.38 times higher prevalence of long COVID diagnosis (95% CI: 1.23–1.56), which were not statistically significant in the adjusted model. Respondents residing in highly socially vulnerable counties for the minority status and language theme had 1.24 times higher prevalence of long COVID diagnosis (95% CI: 1.05–1.47), but it was not statistically significant after the adjustment for covariates.

Table 4 Associations of each theme of minority health social vulnerability index with ongoing long COVID and long COVID diagnosis, Michigan COVID-19 recovery surveillance study, 2022–2023 (n = 3,781)
Full size table

The sensitivity analysis indicated that the overall results were similar, but statistical significance was more pronounced when we used quintiles of the MHSVI (Supplementary Table S1). Respondents living in the highest socially vulnerable counties (Q5) had 1.48 times higher prevalence of long COVID diagnosis (95% CI: 1.01–2.17) than those living in the lowest socially vulnerable counties (Q1) with adjustment for covariates. By theme of the MHSVI, respondents in the more socially vulnerable counties had a higher prevalence of ongoing long COVID (aPR: 1.14, 95% CI: 1.02–1.27 for Q3) and long COVID diagnosis (aPR: 1.26, 95% CI: 1.01–1.56 for Q3; aPR: 1.41, 95% CI: 1.10–1.79 for Q4; aPR: 1.43, 95% CI: 1.16–1.76 for Q5) in the socioeconomic status theme. In the household composition and disability theme, respondents in the more socially vulnerable counties had a higher prevalence of ongoing long COVID (aPR: 1.24, 95% CI: 1.00–1.54 for Q2) and long COVID diagnosis (aPR: 1.41, 95% CI: 1.06–1.86 for Q2; aPR: 1.35, 95% CI: 1.07–1.71 for Q3; aPR: 1.64, 95% CI: 1.17–2.30 for Q5). The trend was similar in in the housing and transportation theme for ongoing long COVID (aPR: 1.19, 95% CI: 1.05–1.35 for Q2) and long COVID diagnosis (aPR: 1.32, 95% CI: 1.11–1.57 for Q2; aPR: 1.40, 95% CI: 1.12–1.76 for Q3; aPR: 1.36, 95% CI: 1.06–1.75 for Q4). Our results for the medical vulnerability theme were robust, suggesting that respondents in highly socially vulnerable counties had a higher prevalence of long COVID diagnosis compared to those in the lowest socially vulnerable counties (aPR: 1.18, 95% CI: 1.00–1.38 for Q3; aPR: 1.36, 95% CI: 1.10–1.68 for Q4). In contrast, living in highly socially vulnerable counties in terms of healthcare infrastructure and access was associated with a lower prevalence of long COVID diagnosis (aPR: 0.76, 95% CI: 0.57–1.00 for Q3; aPR: 0.71, 95% CI: 0.54–0.93 for Q5).

Discussion

This study examines how social vulnerability is related to long COVID illness by using population-based data of adults with COVID-19 infection. With the overall MHSVI score, living in highly socially vulnerable counties was not associated with the prevalence of ongoing long COVID or long COVID diagnosis compared to living in low-to-moderately socially vulnerable counties. However, the theme analysis indicated that living in counties with high medical vulnerability (i.e., higher rates of cardiovascular disease, chronic respiratory disease, obesity, diabetes, no internet access) was associated with a higher prevalence of long COVID diagnosis, which was robust in both main and sensitivity analysis. Our findings highlight the importance of studying how different aspects of area-level social vulnerability contribute to health inequity.

Among our analytic sample of adults with COVID-19 in Michigan, 17.4% reported ongoing long COVID and 11.2% received a diagnosis of long COVID. As previous studies used different study populations, methods, and definitions of long COVID, it is difficult to compare the prevalence of long COVID across studies. One comparable estimate was from the Household Pulse Survey, which suggested that roughly 10% of U.S. adults with COVID-19 reported ongoing long COVID as of October 2023 [41]. The prevalence of ongoing long COVID was higher in our study, possibly because we included only PCR-confirmed COVID-19 cases and defined long COVID as potential long-term symptoms without a specific time frame, whereas the Household Pulse Survey identified COVID-19 cases based on any type of COVID-19 tests and defined long COVID as long-term symptoms lasting 3 months or longer. For example, individuals with severe symptoms of COVID-19 illness, who are more likely to experience long COVID [18], may also be more likely to seek medical care and get PCR-confirmed tests.

The main finding on the association between overall MHSVI and long COVID was not statistically significant, while the sensitivity analysis indicated that only respondents in the highest MHSVI quintile had a higher prevalence of long COVID diagnosis than those in the lowest MHSVI quintile. However, previous studies found statistically significant associations of higher MHSVI with COVID-19 incidence [2], mobility disability after a COVID-19 diagnosis [24], and COVID-19 mortality [2]. Additionally, a few studies have found areas with vulnerable socioeconomic status (e.g., households in poverty) or household composition and disability (e.g., households with children or older adults) are more likely to have worse COVID-19-related outcomes, such as COVID-19 incidence [9] or mobility disability after a COVID-19 diagnosis [24]. Our findings on the robust relationship between higher medical vulnerability and long COVID diagnosis are aligned with a recent study, which suggested that counties with greater medical vulnerabilities had lower COVID-19 vaccination coverage [42] given that COVID-19 vaccines protect against long COVID [20]. This could be because the risk of having long COVID illness is better explained by medical vulnerability than other aspects of social vulnerability when individual-level and area-level covariates were adjusted. Our findings may reflect that communities with vulnerable medical conditions (e.g., cardiovascular or chronic respiratory disease) have (1) lower capacities and fewer resources to recover from disasters such as COVID-19 pandemic [43, 44], (2) less access to healthcare services with high quality to receive treatment [9], and (3) a higher burden of other chronic conditions and comorbidities that worsen long-term effects of COVID illness [45, 46]. While the sensitivity analysis for most themes indicated a positive association between social vulnerability and long COVID, living in socially vulnerable counties assessed by the healthcare infrastructure and access theme was associated with a lower prevalence of long COVID diagnosis. This might be because individuals in these areas have limited access to medical services needed to receive a long COVID diagnosis, resulting in under-reporting. It has been documented that restricted healthcare access can lead to underestimating cases of infectious diseases [47]. Further research should be done to identify the mechanism underlying the observed association.

Our study has several limitations. First, our data included only adults who had a positive PCR test for SARS-CoV-2, were recorded in the MDSS with valid contact and geographic information,

were alive when the survey sample was drawn, and agreed to participate at follow-up. Thus, our sample may have selection bias as we did not include individuals who had COVID-19 but were never tested or those who tested positive at home and individuals who died from severe COVID-19 illness, which limits generalizability. Second, we used self-reported assessments of ongoing long COVID and long COVID diagnosis by health professionals, which lack medical confirmation. As the long COVID definitions have been evolving due to highly heterogeneous and complex symptoms of long COVID [48, 49], our long COVID measurements might not correctly assess ongoing long COVID or long COVID diagnosis. Additionally, responses to our long COVID questions could be different due to differences between individuals, such as health knowledge, health-seeking behaviors, access to healthcare, and availability of clinicians [50, 51]. Third, the response rate for the baseline survey was 32.1%, which is consistent with other large probability surveys [52, 53]. To account for nonresponse, we used sampling weights that matched the age and sex distribution of the sampling frame. Additionally, a nonresponse bias analysis demonstrated few differences between respondents and nonrespondents [54]. Finally, causal inference is limited because this study examined cross-sectional associations between social vulnerability and long COVID using observational data.

Despite these limitations, our study contributes to the literature by identifying understudied associations between area-level social vulnerability and long COVID, using a population-based study, which provides timely and representative data during the public health emergency. Using the MHSVI, a specific and comprehensive social vulnerability index to study COVID-19 pandemic, we found that living in medically vulnerable areas was associated with long COVID diagnosis. Community-based support is needed for vulnerable areas with high concentrations of households with pre-existing chronic conditions. Support could include improving access and quality of medical care services and healthcare providers for long COVID illness for households with chronic conditions [43]. Additionally, social support programs collaborating with social workers (e.g., group wellness program, care coordination) may be helpful to reduce stress, facilitate healthcare utilization, and ease burdens of comorbidities [55]. As existing inequalities are magnified during times of crisis, governments and public health authorities should pay close attention to upstream social determinants of health and provide medically vulnerable communities with more resources particularly during pandemics or other public health emergencies.

Data availability

No datasets were generated or analysed during the current study.

References

  1. Islam SJ, Nayak A, Hu Y, Mehta A, Dieppa K, Almuwaqqat Z, et al. Temporal trends in the association of social vulnerability and race/ethnicity with county-level COVID-19 incidence and outcomes in the USA: an ecological analysis. BMJ Open. 2021;11(7):e048086.

    Article  PubMed  Google Scholar 

  2. Tipirneni R, Schmidt H, Lantz PM, Karmakar M. Associations of 4 geographic social vulnerability indices with US COVID-19 incidence and mortality. Am J Public Health. 2022;112(11):1584–8.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Tan TQ, Kullar R, Swartz TH, Mathew TA, Piggott DA, Berthaud V. Location matters: geographic disparities and impact of coronavirus disease 2019. J Infect Dis. 2020;222(12):1951–4.

    Article  CAS  PubMed  Google Scholar 

  4. Asabor EN, Warren JL, Cohen T. Racial/Ethnic segregation and access to COVID-19 testing: Spatial distribution of COVID-19 testing sites in the four largest highly segregated cities in the united States. Am J Public Health. 2022;112(3):518–26.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Hawkins RB, Charles EJ, Mehaffey JH. Socio-economic status and COVID-19–related cases and fatalities. Public Health. 2020;189:129–34.

    Article  CAS  PubMed  Google Scholar 

  6. Melvin SC, Wiggins C, Burse N, Thompson E, Monger M. The role of public health in COVID-19 emergency response efforts from a rural health perspective. Prev Chronic Dis. 2020;17:E70.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Gaynor TS, Wilson ME. Social vulnerability and equity: the disproportionate impact of COVID-19. Public Adm Rev. 2020;80(5):832–8.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Krieger N, Waterman PD, Chen JT. COVID-19 and overall mortality inequities in the surge in death rates by zip code characteristics: Massachusetts, January 1 to May 19, 2020. Am J Public Health. 2020;110(12):1850–2.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Karaye IM, Horney JA. The impact of social vulnerability on COVID-19 in the U.S.: an analysis of spatially varying relationships. Am J Prev Med. 2020;59(3):317–25.

    Article  PubMed  PubMed Central  Google Scholar 

  10. U.S. Centers for Disease Control and Prevention. Long COVID Basics [Internet]. 2024. Available from: https://www.cdc.gov/covid/long-term-effects/index.html?s_cid=SEM.GA:PAI:RG_AO_GA_TM_A18_C-CVD-AfterCOVID-Brd:after%20covid%20symptoms:SEM00001%26;utm_id=SEM.GA:PAI:RG_AO_GA_TM_A18_C-CVD-AfterCOVID-Brd:after%20covid%20symptoms:SEM00001%26;gad_source=1

  11. U.S. Centers for Disease Control and Prevention, Long COVID. Household Pulse Survey [Internet]. 2023 [cited 2024 Jan 22]. Available from: https://www.cdc.gov/nchs/covid19/pulse/long-covid.htm

  12. Spence NJ, Russell D, Bouldin ED, Tumminello CM, Schwartz T. Getting back to normal? Identity and role disruptions among adults with long COVID. Sociol Health Illn. 2023;45(4):914–34.

    Article  PubMed  Google Scholar 

  13. Gualano MR, Rossi MF, Borrelli I, Santoro PE, Amantea C, Daniele A, et al. Returning to work and the impact of post COVID-19 condition: A systematic review. Work. 2022;73(2):405–13.

    PubMed  Google Scholar 

  14. Ireson J, Taylor A, Richardson E, Greenfield B, Jones G. Exploring invisibility and epistemic injustice in long Covid-A citizen science qualitative analysis of patient stories from an online Covid community. Health Expect. 2022;25(4):1753–65.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Burdorf A, Porru F, Rugulies R. The COVID-19 pandemic: one year later– an occupational perspective. Scand J Work Environ Health. 2021;5VL–47(4):245–7.

    Article  Google Scholar 

  16. Perlis RH, Santillana M, Ognyanova K, Safarpour A, Lunz Trujillo K, Simonson MD, et al. Prevalence and correlates of long COVID symptoms among US adults. JAMA Netw Open. 2022;5(10):e2238804–2238804.

    Article  PubMed  PubMed Central  Google Scholar 

  17. MacCallum-Bridges CL, Hirschtick JL, Allgood KL, Ryu S, Orellana RC, Fleischer NL. Cross-sectional population-based estimates of a rural-urban disparity in prevalence of long COVID among Michigan adults with polymerase chain reaction-confirmed COVID-19, 2020–2022. J Rural Health. 2023 Nov 16.

  18. Hirschtick JL, Titus AR, Slocum E, Power LE, Hirschtick RE, Elliott MR, et al. Population-Based estimates of Post-acute sequelae of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (PASC) prevalence and characteristics. Clin Infect Dis. 2021;73(11):2055–64.

    Article  CAS  PubMed  Google Scholar 

  19. Khullar D, Zhang Y, Zang C, Xu Z, Wang F, Weiner MG, et al. Racial/Ethnic disparities in Post-acute sequelae of SARS-CoV-2 infection in new York: an EHR-Based cohort study from the RECOVER program. J Gen Intern Med. 2023;38(5):1127–36.

    Article  PubMed  PubMed Central  Google Scholar 

  20. MacCallum-Bridges C, Hirschtick JL, Patel A, Orellana RC, Elliott MR, Fleischer NL. The impact of COVID-19 vaccination prior to SARS-CoV-2 infection on prevalence of long COVID among a population-based probability sample of michiganders, 2020–2022. Ann Epidemiol. 2024;92:17–24.

    Article  PubMed  Google Scholar 

  21. Hirschtick JL, Xie Y, Slocum E, Hirschtick RE, Power LE, Elliott MR, et al. A statewide population-based approach to examining long COVID symptom prevalence and predictors in Michigan. Prev Med. 2023;177:107752.

    Article  PubMed  Google Scholar 

  22. Dasgupta S, Bowen VB, Leidner A, Fletcher K, Musial T, Rose C, et al. Association between social vulnerability and a County’s risk for becoming a COVID-19 Hotspot–United States, June 1-July 25, 2020. MMWR Morb Mortal Wkly Rep. 2020;69(42):1535–41.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  23. Landman JM, Steger-May K, Joynt Maddox KE, Hammond G, Gupta A, Rauseo AM, et al. Estimating the effects of race and social vulnerability on hospital admission and mortality from COVID-19. JAMIA Open. 2021;4(4):ooab111.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Allgood KL, Whittington B, Xie Y, Hirschtick JL, Ro A, Orellana RC, et al. Social vulnerability and new mobility disability among adults with polymerase chain reaction (PCR)-confirmed SARS-CoV-2: Michigan COVID-19 recovery surveillance study. Prev Med. 2023;177:107719.

    Article  PubMed  Google Scholar 

  25. Kim SJ, Bostwick W. Social vulnerability and Racial inequality in COVID-19 deaths in Chicago. Health Educ Behav. 2020;47(4):509–13.

    Article  PubMed  Google Scholar 

  26. Kiefer MK, Mehl R, Rood KM, Germann K, Mallampati D, Manuck T, et al. Association between social vulnerability and COVID-19 vaccination hesitancy and vaccination in pregnant and postpartum individuals. Vaccine. 2022;40(44):6344–51.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Qureshi I, Gogoi M, Al-Oraibi A, Wobi F, Pan D, Martin CA, et al. Intersectionality and developing evidence-based policy. Lancet. 2022;399(10322):355–6.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Office of Minority Health. Minority Health Social Vulnerability Index. [Internet]. Centers for Disease Control and Prevention, Agency for Toxic Substances and Disease Registry. 2023. Available from: https://www.minorityhealth.hhs.gov/minority-health-svi/

  29. Michigan Department of Health and Human Services. Regional trauma networks [Internet]. 2022 [cited 2023 May 10]. Available from: https://www.michigan.gov/mdhhs/doing-business/trauma/regional

  30. American Association for Public Opinion Research. Standard definitions: Final dispositions of case codes and outcome rates for surveys [Internet]. 2016. Available from: https://aapor.org/wp-content/uploads/2022/11/Standard-Definitions20169theditionfinal.pdf

  31. U.S. Centers for Disease Control and Prevention. CDC SVI Documentation 2020 [Internet]. Atlanta, GA: Centers for Disease Control and Prevention; 2020 Feb [cited 2024 Jan 16]. Available from: https://www.atsdr.cdc.gov/placeandhealth/svi/documentation/SVI_documentation_2020.html

  32. Basile Ibrahim B, Barcelona V, Condon EM, Crusto CA, Taylor JY. The association between neighborhood social vulnerability and cardiovascular health risk among Black/African American women in the intergen study. Nurs Res. 2021;70(5S):S3.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Cardenas MGP, Otalvaro-Acosta L, et al. Social vulnerability and appendicitis: single-institution study in a high insurance state. J Surg Res. 2022;275:35–42.

    Article  PubMed  Google Scholar 

  34. Carmichael H, Dyas AR, Bronsert MR et al. Mar 9 2022. Social vulnerability is associated with increased morbidity following colorectal surgery. Am J Surg. 2022.

  35. Paro A, Hyer JM, Diaz A, Tsilimigras DI, Pawlik TM. Profiles in social vulnerability: the association of social determinants of health with postoperative surgical outcomes. Surgery. 2021;170(6):1777–84.

    Article  PubMed  Google Scholar 

  36. Roth SE, Govier DJ, Marsi K, Cohen-Cline H. Differences in outpatient health care utilization 12 months after COVID-19 infection by race/ethnicity and community social vulnerability. Int J Environ Res Public Health 2022; 19 (6).

  37. Hyer JM, Tsilimigras DI, Diaz A, Dalmacy D, Paro A, Pawlik TM. Patient social vulnerability and hospital community Racial/ethnic integration: do all patients undergoing pancreatectomy receive the same care across hospitals? Ann Surg. 2021;274(3):508–15.

    Article  PubMed  Google Scholar 

  38. Tetzlaff F, Sauerberg M, Grigoriev P, et al. Age-specific and cause-specific mortality contributions to the socioeconomic gap in life expectancy in Germany, 2003-21: an ecological study. Lancet Public Health. 2025;10(1):e10. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/S2468-2667(24)00297-4.

    Article  Google Scholar 

  39. Brown M, Goodman A, Peters A, Ploubidis GB, Sanchez A, Silverwood R, et al. COVID-19 survey in five National longitudinal studies: waves 1, 2 and 3 user guide (Version 3). London: UCL Centre for Longitudinal Studies and MRC Unit for Lifelong Health and Ageing; 2021.

    Google Scholar 

  40. Mayer B. Hot deck propensity score imputation for missing values. Sci J Med Clin Trials. 2013.

  41. US Centers for Disease Control and Prevention, Long COVID. Household Pulse Survey. Accessed December 29, 2024. https://www.cdc.gov/nchs/covid19/pulse/long-covid.htm

  42. Saelee R, Chandra Murthy N, Patel Murthy B, Zell E, Shaw L, Gibbs-Scharf L, et al. Minority health social vulnerability index and COVID-19 vaccination coverage — The united States. Vaccine. 2023;41(12):1943–50. December 14, 2020–January 31, 2022.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Zakour MJ, Harrell EB. Access to disaster services: social work interventions for vulnerable populations. J Social Service Res. 2003;30(2):27–54.

    Article  Google Scholar 

  44. Flanagan BE, Gregory EW, Hallisey EJ, Heitgerd JL, Lewis B. A social vulnerability index for disaster management. J Homel Secur Emerg Manage. 2011;8(1):0000102202154773551792.

    Google Scholar 

  45. Tisminetzky M, Delude C, Hebert T, Carr C, Goldberg RJ, Gurwitz JH, Age. Multiple chronic conditions, and COVID-19: A literature review. Journals Gerontology: Ser A. 2022;77(4):872–8.

  46. Arjun MC, Singh AK, Pal D, Das K, Venkateshan GA. Characteristics and predictors of long COVID among diagnosed cases of COVID-19. PLoS ONE. 2022;17(12):e0278825.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Giorgi Rossi P, Riccardo F, Pezzarossi A, Ballotari P, Dente MG, Napoli C, Chiarenza A, Velasco Munoz C, Noori T, Declich S. Factors influencing the accuracy of infectious disease reporting in migrants: A scoping review. Int J Environ Res Public Health. 2017;14(7):720. https://doiorg.publicaciones.saludcastillayleon.es/10.3390/ijerph14070720.

    Article  PubMed  PubMed Central  Google Scholar 

  48. Fernández-de-las-Peñas C, Long COVID. Current definition. Infection. 2022;50:285–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s15010-021-01696-5.

    Article  CAS  PubMed  Google Scholar 

  49. Chou R, Herman E, Ahmed A, et al. Long COVID definitions and models of care: A scoping review. Ann Intern Med. 2024;177:929–40. https://doiorg.publicaciones.saludcastillayleon.es/10.7326/M24-0677.

    Article  PubMed  Google Scholar 

  50. Knuppel A, Boyd A, Macleod J, Chaturvedi N, Williams DM. The long COVID evidence gap in England. Lancet. 2024;403(10440):1981–2.

    Article  PubMed  Google Scholar 

  51. Huijts T, Gage Witvliet M, Balaj M, Andreas Eikemo T. Assessing the long-term health impact of COVID-19: the importance of using self-reported health measures. Scand J Public Health. 2023;51(5):645–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1177/14034948221143421.

    Article  PubMed  PubMed Central  Google Scholar 

  52. Meyer BD, Mok WKC, Sullivan JX. Household surveys in crisis. J Economic Perspect. 2015;29:199–226. https://doiorg.publicaciones.saludcastillayleon.es/10.1257/jep.29.4.199.

    Article  Google Scholar 

  53. Krieger N, LeBlanc M, Waterman P, Reisner SL, Testa C. Decreasing survey response rates in the time of COVID-19: implications for analyses of population health and health inequities. Am J Public Health. 2023;113:667–70. https://doiorg.publicaciones.saludcastillayleon.es/10.2105/AJPH.2023.307267.

    Article  PubMed  PubMed Central  Google Scholar 

  54. Hirschtick JL, Xie Y, Whittington B, Patel A, Elliott MR, Allgood K, Coyle J, Fleischer NL. Methodology for a COVID-19 Recovery Surveillance Study conducted through an academic–state partnership. Accepted at Public Health Reports.

  55. Bellizzi KM, Fritzson E, Ligus K, Park CL. Social support buffers the effect of social deprivation on comorbidity burden in adults with Cancer. Ann Behav Med. 2024 Jun 27.

Download references

Acknowledgements

The authors thank the Michigan COVID-19 Recovery Surveillance Study participants and interviewers for making this study possible, as well as the study’s Community Advisory Committee, including Ghada Aziz, Rev. Sarah Bailey, Vicki Dobbins, Carlton Evans, Adnan Hammad, Chuqui King, Marta Larson, Roquesha O’Neal, and LaKila Shea Salter. The authors also thank Blair Whittington for geocoding the MI CReSS data and Akash Patel for his data management of the MI CReSS data.

Funding

The Michigan COVID-19 Recovery Surveillance Study has received funding from the Michigan Department of Health and Human Services, the Michigan Public Health Institute, the University of Michigan Institute for Data Science, the University of Michigan Rogel Cancer Center, and the University of Michigan Epidemiology Department. This manuscript is supported by the Centers for Disease Control and Prevention of the U.S. Department of Health and Human Services (HHS) funded by CDC/HHS through grant number 6 NU50CK000510-02-04 and 1 NH75OT000078-01-00. The contents are those of the authors and do not necessarily represent the official views of, nor an endorsement, by CDC/HHS, or the U.S. Government.

Author information

Authors and Affiliations

  1. Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, USA

    Soomin Ryu, Yanmei Xie & Nancy L. Fleischer

  2. Department of Epidemiology and Biostatistics, School of Public Health, Texas A&M University, College Station, TX, USA

    Kristi L. Allgood

  3. CDC Foundation, Atlanta, GA, USA

    Robert C. Orellana

  4. Michigan Department of Health and Human Services, Lansing, MI, USA

    Robert C. Orellana

Authors
  1. Soomin Ryu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Kristi L. Allgood
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Yanmei Xie
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Robert C. Orellana
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Nancy L. Fleischer
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

SR conceptualized the study, conducted data analysis, and wrote the main manuscript text. KLA conceptualized the study and reviewed and revised the manuscript. YX suggested methodology and reviewed and revised the manuscript. RCO reviewed and revised the manuscript. NLF conceptualized the study, reviewed, and revised the manuscript, supervised the study, and acquired funding. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Soomin Ryu.

Ethics declarations

Ethics approval and consent to participate

Ethical approval for this analysis was considered exempt by the University of Michigan Institutional Review Board (HUM00207685) due to the use of a de-identified secondary dataset. Informed consent was obtained from all individual participants included in this study. All research was completed in accordance with the Declaration of Helsinki.

Consent for publication

All authors of the manuscript have read and agreed to its content and are accountable for all aspects of the accuracy and integrity of the manuscript in accordance with ICMJE (International Committee of Medical Journal Editors) criteria.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ryu, S., Allgood, K.L., Xie, Y. et al. Minority health social vulnerability index and long COVID illness among a statewide, population-based study of adults with polymerase chain reaction-confirmed SARS-CoV-2. Arch Public Health 83, 64 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13690-025-01553-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13690-025-01553-z

Archives of Public Health

ISSN: 2049-3258