https://orcid.org/0009-0007-4032-3989
Figures
Abstract
Current immunosuppressive treatment is used for autoimmune, inflammatory, malignant, transplant-related, and other chronic conditions that may carry elevated cerebrovascular burden. Nationally representative evidence on current immunosuppressive treatment status and self-reported lifetime stroke history remains limited. This cross-sectional study used pooled 2023–2024 National Health Interview Survey Adult Sample data. The analytic sample included adults aged 18 years or older with valid survey weights and complete data on exposure, outcome, and selected covariates. The exposure was current self-reported use of medications that weaken the immune system. The outcome was self-reported lifetime stroke history. Survey-weighted logistic regression estimated odds ratios and 95% confidence intervals, accounting for National Health Interview Survey stratification, clustering, and adjusted sampling weights. The final analytic sample included 60,026 adults, representing approximately 250 million United States adults. The weighted prevalence of current immunosuppressive treatment was 4.6%. Stroke history was more common among adults receiving immunosuppressive treatment than among those not receiving treatment (6.0% versus 2.8%). In unadjusted analysis, immunosuppressive treatment was associated with higher odds of stroke history (odds ratio 2.24, 95% confidence interval 1.86–2.70, p < 0.001). After full adjustment for age, sex, race/ethnicity, hypertension, diabetes, smoking status, coronary heart disease, and income-to-poverty category, the association persisted (adjusted odds ratio 1.71, 95% confidence interval 1.40–2.08, p < 0.001). Current immunosuppressive treatment status, likely reflecting both medication exposure and underlying disease burden, was associated with higher odds of self-reported lifetime stroke history among United States adults. Because of the cross-sectional design, broad exposure definition, self-reported outcome, and potential confounding by indication, these findings should be interpreted as an association rather than evidence of causation.
Citation: Ansari MS, Nisar F, Reyaz U, Fatima M, Khan UA, Hadi B, et al. (2026) Current immunosuppressive treatment and self-reported lifetime stroke history among United States adults: A cross-sectional study. PLoS One 21(7): e0355164. https://doi.org/10.1371/journal.pone.0355164
Editor: Marwan Salih Al-Nimer, University of Diyala College of Medicine, IRAQ
Received: June 10, 2026; Accepted: July 18, 2026; Published: July 30, 2026
Copyright: © 2026 Ansari et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The datasets analysed during the current study are publicly available from the National Center for Health Statistics National Health Interview Survey repository: https://www.cdc.gov/nchs/nhis/. The 2023 and 2024 Adult Sample public-use data files and documentation were used for this analysis.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Stroke is a major cause of morbidity, disability, and mortality in the United States and remains a central focus of cardiovascular and neurologic prevention efforts [1]. Identifying populations with elevated cerebrovascular disease burden is important for risk-factor recognition and public health surveillance.
Adults receiving immunosuppressive treatment may represent one such population. Immunosuppressive medications are used for heterogeneous indications, including systemic autoimmune disease, inflammatory disorders, malignancy-associated treatment, and transplant-related care. These indications frequently overlap with chronic inflammation, immune dysregulation, high comorbidity burden, and intensive health-care utilization, processes that have independently been linked to cerebrovascular risk [2,4]. Nationally representative studies therefore need to distinguish carefully between medication exposure and the underlying disease burden that leads to treatment.
The biologic plausibility linking immune-mediated disease and cerebrovascular disease is supported by literature on inflammation, atherosclerosis, and immune dysregulation [2–5]. Large population-based evidence has also shown that autoimmune diseases are associated with increased risk across multiple cardiovascular outcomes, including cerebrovascular disease [6]. Disease-specific literature has reported elevated cardiovascular or cerebrovascular risk in rheumatoid arthritis, systemic lupus erythematosus, and inflammatory bowel disease [7–10]. Medication-specific observational studies in autoimmune populations suggest that cardiovascular risk may differ by drug class, disease severity, and treatment indication [11–14].
Despite this background, nationally representative United States evidence on the association between current immunosuppressive treatment status and self-reported lifetime stroke history is limited. This study used pooled 2023–2024 National Health Interview Survey Adult Sample data to examine whether current self-reported immunosuppressive treatment was associated with self-reported lifetime stroke history among United States adults. We hypothesized that adults reporting current immunosuppressive treatment would have higher odds of reporting prior stroke after adjustment for demographic, socioeconomic, behavioral, and cardiometabolic factors. Because the National Health Interview Survey ascertains stroke history through a single self-reported item without physician confirmation or subtype classification, “stroke” in this study refers broadly to any self-reported lifetime cerebrovascular event and does not distinguish ischemic from hemorrhagic subtype.
Materials and methods
Study design and data source
This cross-sectional study used pooled 2023–2024 National Health Interview Survey Adult Sample data. The National Health Interview Survey is a nationally representative, stratified, multistage probability survey of the civilian, noninstitutionalized United States population conducted by the National Center for Health Statistics [15]. National Health Interview Survey sampling weights, strata, and primary sampling units are designed to support population-representative estimation and valid variance estimation when the complex survey design is incorporated [15]. The Strengthening the Reporting of Observational Studies in Epidemiology guidance was used to structure reporting [16].
The 2023 and 2024 public-use data files were accessed for research purposes on 31 May 2026. The authors accessed only de-identified public-use data and had no access to information that could identify individual participants during or after data collection.
Study population
The study population included adults aged 18 years or older. Participants were eligible if they had valid information on immunosuppressive treatment status and self-reported lifetime stroke history, a positive adult sample weight, and complete data for selected covariates. Survey weights were divided by two to account for pooling two survey years. The final complete-case analytic sample included 60,026 adults. A complete-case analysis was used because the primary regression models required nonmissing data for exposure, outcome, survey design variables, and all selected covariates. Of 62,034 eligible adults, 2,008 were excluded because of missing covariate data, leaving a final analytic sample of 60,026 adults. This represented approximately 3.2% of the eligible sample. Complete-case analysis was considered acceptable because the proportion excluded was small; however, possible selection bias from missing data was acknowledged as a limitation. Fig 1 summarizes sample selection.
NHIS, National Health Interview Survey.
Exposure
The exposure was current self-reported immunosuppressive treatment status, based on the National Health Interview Survey item regarding current use of medications that weaken the immune system (MEDRXTRT_A). The primary exposure was coded as yes versus no. The National Health Interview Survey immunosuppressive treatment item does not capture medication class, dose, treatment indication, or disease severity; therefore, medication-specific or indication-specific associations with stroke could not be examined, and the exposure should be interpreted as a broad marker of immunosuppressive treatment status rather than a specific therapeutic agent. Refused or “don’t know” responses for current immunosuppressive treatment were uncommon, occurring in 186 of 62,034 eligible participants (0.30%; refused: n = 36; don’t know: n = 150). These responses were grouped with the no-treatment category in the primary binary analysis to preserve a conservative reference group and avoid unstable estimates from sparse categories; a sensitivity analysis modeled uncertain responses separately to evaluate whether this coding decision affected the findings.
Outcome
The primary outcome was self-reported lifetime stroke history, defined using STREV_A. Participants reporting a prior stroke were classified as having stroke history. The analysis estimates the association between current treatment status and lifetime stroke history, not incident stroke or prospective stroke risk.
Covariates
Covariates were selected a priori based on known or suspected associations with both immunosuppressive treatment status and stroke history, and because they are consistently ascertained and validated within the National Health Interview Survey Adult Sample core module. These included age, sex, race/ethnicity, hypertension, type 2 diabetes, smoking status, coronary heart disease, and income-to-poverty category. Other established stroke-related factors, including body mass index, hyperlipidemia, physical activity level, atrial fibrillation, and chronic kidney disease, were not included in the primary adjustment set; their omission is addressed as a source of potential residual confounding in the Limitations. Race/ethnicity was categorized as non-Hispanic White, non-Hispanic Black, Hispanic, or Other. Smoking status was categorized as never smoker, current every day smoker, current some day smoker, or former smoker. Income was modeled as an ordinal income-to-poverty category, with higher values representing higher family income relative to the poverty threshold.
Statistical analysis
All analyses incorporated the National Health Interview Survey complex survey design, including adjusted adult sample weights, strata, and primary sampling units. Descriptive statistics were calculated as survey-weighted proportions for categorical variables and survey-weighted means with standard errors for continuous variables. Group comparisons used survey-weighted Wald tests for continuous variables and Rao-Scott design-adjusted chi-square tests for categorical variables. Corresponding survey-weighted test statistics were reported for Table 1.
Survey-weighted logistic regression was used to estimate the association between immunosuppressive treatment status and self-reported lifetime stroke history. Logistic regression is appropriate for binary outcomes in cross-sectional data when interpreted as an odds ratio rather than a prospective risk ratio [17]. An unadjusted model included immunosuppressive treatment only. Sequential models then adjusted for potential confounders: Model 1 included age, sex, and race/ethnicity; Model 2 additionally included hypertension, diabetes, and smoking status; and Model 3 additionally included coronary heart disease and income-to-poverty category. Model 3 was considered the primary fully adjusted model. Exploratory interaction analyses examined whether the association differed by sex, age group, race/ethnicity, diabetes status, or hypertension status. No correction was applied for multiple comparisons in exploratory interaction analyses; therefore, subgroup findings were interpreted cautiously. Statistical significance was defined as two-sided p < 0.05. Data extraction, cleaning, and variable transformation were performed using the Lumono.ai research platform; statistical analyses were conducted in R version 4.5 using survey-weighted methods [18].
Results
Study population and exposure prevalence
The final analytic sample included 60,026 adults, representing approximately 250 million United States adults after application of survey weights. Overall, 2,975 participants reported current immunosuppressive treatment, corresponding to a weighted prevalence of 4.6%.
Baseline characteristics
Participants receiving immunosuppressive treatment were older than those not receiving treatment (weighted mean age 52.4 versus 47.9 years). The weighted prevalence of current immunosuppressive treatment was 5.3% among females and 3.9% among males. Adults receiving immunosuppressive treatment had higher comorbidity burden, including hypertension (42.1% versus 31.5%), diabetes (15.5% versus 9.6%), and coronary heart disease (9.4% versus 4.7%). Stroke history was also more common among adults receiving immunosuppressive treatment than among those not receiving treatment (6.0% versus 2.8%, p < 0.001). Baseline characteristics are shown in Table 1.
Sequential regression models
In the unadjusted survey-weighted logistic regression model, current immunosuppressive treatment was associated with higher odds of self-reported lifetime stroke history (odds ratio 2.24, 95% confidence interval 1.86–2.70, p < 0.001). After adjustment for age, sex, and race/ethnicity, the association was attenuated but remained statistically significant (odds ratio 2.00, 95% confidence interval 1.65–2.44, p < 0.001). Further adjustment for hypertension, diabetes, and smoking status yielded an odds ratio of 1.84 (95% confidence interval 1.51–2.24, p < 0.001). In the fully adjusted model, which additionally included coronary heart disease and income-to-poverty category, current immunosuppressive treatment remained associated with higher odds of stroke history (adjusted odds ratio 1.71, 95% confidence interval 1.40–2.08, p < 0.001). Sequential models are shown in Table 2.
Fully adjusted model
In the fully adjusted model, older age, hypertension, diabetes, current smoking, former smoking, coronary heart disease, and lower income-to-poverty category were associated with higher odds of stroke history. Coronary heart disease (odds ratio 2.41, 95% confidence interval 2.10–2.78) and hypertension (odds ratio 2.25, 95% confidence interval 1.95–2.59) showed strong independent associations with stroke history. Full model estimates are shown in Table 3.
Exploratory interaction and sensitivity analyses
Exploratory interaction analyses suggested possible effect modification by age and race/ethnicity. The association between immunosuppressive treatment and stroke history was stronger among adults aged 55 years or younger (odds ratio 2.67, 95% confidence interval 1.76–4.06) than among adults older than 55 years (odds ratio 1.35, 95% confidence interval 1.10–1.67; interaction p = 0.018). Evidence of possible interaction was also observed by race/ethnicity (interaction p = 0.041). No statistically significant interaction was observed by sex, diabetes status, or hypertension status. Because multiple subgroup tests were performed and no correction was applied for multiple testing, these findings should be interpreted as exploratory.
In sensitivity analysis using a three-level exposure definition, the association between definite immunosuppressive treatment and stroke history was nearly identical to the primary analysis (odds ratio 1.71, 95% confidence interval 1.41–2.09, p < 0.001). Participants with uncertain immunosuppressive treatment status also had higher odds of stroke history than those reporting no immunosuppressive treatment (odds ratio 2.03, 95% confidence interval 1.03–4.02, p = 0.041). Sensitivity analysis results are shown in Table 4. A supplementary analysis addressed outcome-ascertainment uncertainty specifically: of the 60,026 participants in the analytic sample, 22 (0.04%) had a refused or “don’t know” response for stroke history, coded as no stroke history in the primary analysis. Excluding these 22 participants did not change the fully adjusted association (odds ratio 1.71, 95% confidence interval 1.40–2.08), and recoding them as having stroke history, a conservative worst-case assumption, attenuated the association only slightly (odds ratio 1.69, 95% confidence interval 1.39–2.06). These findings indicate that the primary analysis’s treatment of uncertain stroke-history responses did not materially affect the conclusions. A related sensitivity analysis added the two of these factors that are measurable in the National Health Interview Survey, body mass index and hyperlipidemia, to the fully adjusted model. The association was essentially unchanged (odds ratio 1.72, 95% confidence interval 1.41–2.09, versus 1.71 in the primary model), indicating that these two factors do not materially confound the association. Physical activity and atrial fibrillation could not be assessed because the National Health Interview Survey does not collect these variables.
Discussion
Principal findings
In this nationally representative cross-sectional study of United States adults, current self-reported immunosuppressive treatment status was associated with higher odds of self-reported lifetime stroke history. Adults receiving immunosuppressive treatment had a higher weighted prevalence of stroke history than adults not receiving treatment. Sequential adjustment attenuated the association, suggesting partial confounding by demographic and cardiometabolic factors, but the association remained statistically significant after full adjustment.
Interpretation
The findings should be interpreted as evidence that current immunosuppressive treatment status identifies a medically complex population with higher cerebrovascular disease burden. They should not be interpreted as evidence that immunosuppressive medications caused stroke. The exposure likely captures both medication use and the underlying conditions requiring treatment. The observed association is consistent with evidence linking immune-mediated disease, chronic inflammation, and vascular outcomes [2–6]. It is also consistent with disease-specific literature reporting elevated cardiovascular or cerebrovascular risk among patients with rheumatoid arthritis, systemic lupus erythematosus, and inflammatory bowel disease [7–10].
The broad immunosuppressive treatment variable likely captures a heterogeneous group of underlying conditions, including autoimmune and inflammatory disorders, transplant-related immunosuppression, malignancy-associated treatment, and other chronic diseases requiring immune-modifying therapy. These conditions may differ substantially in their vascular risk profiles. For example, systemic autoimmune diseases may increase cerebrovascular burden through chronic inflammation, endothelial dysfunction, thrombosis, and accelerated atherosclerosis, whereas transplant- or malignancy-related immunosuppression may reflect different pathways of medical complexity. Because the National Health Interview Survey does not identify medication class, dose, treatment indication, disease activity, or timing relative to stroke in this analysis, the observed association should be interpreted as reflecting a broad marker of immunosuppressive treatment status and underlying disease burden rather than a medication-specific effect.
Medication-specific studies support caution in interpretation. Observational research among rheumatoid arthritis and systemic lupus erythematosus populations suggests that cardiovascular event risk may vary by medication class, disease severity, and patient selection [11,12]. Studies of autoimmune disease and immune-related therapies further show that cardiovascular outcomes can reflect complex interactions among underlying disease, treatment, and baseline risk profile [13,14].
Clinical and public health implications
These findings do not support changing immunosuppressive therapy decisions. Rather, they suggest that adults reporting current immunosuppressive treatment may warrant careful attention to conventional stroke-related factors, including hypertension, diabetes, smoking, coronary heart disease, and socioeconomic barriers to care. Clinicians should interpret current immunosuppressive treatment status as a possible marker of medical complexity and elevated cerebrovascular burden, not as a medication-specific causal risk factor.
Strengths and limitations
Strengths of this study include use of a large nationally representative United States sample, incorporation of National Health Interview Survey weights and design variables, adjustment for major demographic and cardiometabolic confounders, sequential modeling to assess attenuation across adjustment steps, and sensitivity analysis using an alternative exposure definition. The reporting structure followed Strengthening the Reporting of Observational Studies in Epidemiology guidance for observational studies [16].
This study also has important limitations. First, the cross-sectional design prevents assessment of temporality. Current immunosuppressive treatment may have started before or after a participant’s stroke; therefore, this study estimates an association with lifetime stroke history, not incident stroke risk. Second, both exposure and outcome were self-reported, which may introduce misclassification. Third, the exposure was broad and did not distinguish medication class, dose, duration, indication, or disease activity. Fourth, confounding by indication and residual confounding are major concerns because the analysis could not adjust for specific autoimmune diagnoses, cancer history, transplant status, chronic kidney disease, body mass index, hyperlipidemia, physical activity level, atrial fibrillation, disease severity, corticosteroid dose, medication duration, or health-care utilization. Fifth, refused or “don’t know” responses for stroke history were treated as negative in the primary analysis, which may have underestimated stroke prevalence. Sixth, complete-case analysis may introduce selection bias if missing covariate data were related to treatment status or stroke history, although the proportion excluded for missing confounder data was relatively small. Finally, the National Health Interview Survey excludes institutionalized adults, which may underestimate stroke burden among adults with severe disability or long-term care needs.
Conclusions
In pooled 2023–2024 National Health Interview Survey Adult Sample data, current self-reported immunosuppressive treatment status was associated with higher odds of self-reported lifetime stroke history among United States adults. The association persisted after adjustment for demographic, socioeconomic, behavioral, and cardiometabolic factors. These findings suggest that adults receiving immunosuppressive treatment may represent a population with elevated cerebrovascular disease burden. However, because of the cross-sectional design, broad exposure definition, self-reported outcome, and likely confounding by indication, causality cannot be inferred. Longitudinal studies with medication-specific exposure data, clinical stroke ascertainment, treatment indication, disease severity, and timing of exposure relative to stroke are needed.
Supporting information
S1 File. R code used for statistical analysis.
R script implementing all descriptive statistics, regression models, interaction analyses, and sensitivity analyses reported in this manuscript, including the sequential models in Table 2 and the sensitivity analyses in Table 4. Instructions for downloading and pooling the raw National Health Interview Survey files are included in the script.
https://doi.org/10.1371/journal.pone.0355164.s001
(R)
S2 File. STROBE checklist.
Completed Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist for this cross-sectional study, with page references to the manuscript.
https://doi.org/10.1371/journal.pone.0355164.s002
(DOCX)
References
- 1. Virani SS, Alonso A, Aparicio HJ, Benjamin EJ, Bittencourt MS, Callaway CW. Heart disease and stroke statistics—2021 update: a report from the American Heart Association. Circulation. 2021;143:e254–743.
- 2. Libby P. Inflammation in atherosclerosis. Nature. 2002;420(6917):868–74. pmid:12490960
- 3. Hansson GK. Inflammation, atherosclerosis, and coronary artery disease. N Engl J Med. 2005;352(16):1685–95. pmid:15843671
- 4. Parikh NS, Merkler AE, Iadecola C. Inflammation, Autoimmunity, Infection, and Stroke: Epidemiology and Lessons From Therapeutic Intervention. Stroke. 2020;51(3):711–8. pmid:32078460
- 5. Shoenfeld Y, Gerli R, Doria A, Matsuura E, Cerinic MM, Ronda N, et al. Accelerated atherosclerosis in autoimmune rheumatic diseases. Circulation. 2005;112(21):3337–47. pmid:16301360
- 6. Conrad N, Verbeke G, Molenberghs G, Goetschalckx L, Callender T, Cambridge G, et al. Autoimmune diseases and cardiovascular risk: a population-based study on 19 autoimmune diseases and 12 cardiovascular diseases in 22 million individuals in the UK. Lancet. 2022;400(10354):733–43. pmid:36041475
- 7. Avina-Zubieta JA, Thomas J, Sadatsafavi M, Lehman AJ, Lacaille D. Risk of incident cardiovascular events in patients with rheumatoid arthritis: a meta-analysis of observational studies. Ann Rheum Dis. 2012;71(9):1524–9. pmid:22425941
- 8. Hak AE, Karlson EW, Feskanich D, Stampfer MJ, Costenbader KH. Systemic lupus erythematosus and the risk of cardiovascular disease: results from the nurses’ health study. Arthritis Rheum. 2009;61(10):1396–402. pmid:19790130
- 9. Xiao Z, Pei Z, Yuan M, Li X, Chen S, Xu L. Risk of Stroke in Patients with Inflammatory Bowel Disease: A Systematic Review and Meta-analysis. J Stroke Cerebrovasc Dis. 2015;24(12):2774–80. pmid:26360973
- 10. Maradit-Kremers H, Nicola PJ, Crowson CS, Ballman KV, Gabriel SE. Cardiovascular death in rheumatoid arthritis: a population-based study. Arthritis Rheum. 2005;52(3):722–32. pmid:15751097
- 11. Solomon DH, Avorn J, Katz JN, Weinblatt ME, Setoguchi S, Levin R, et al. Immunosuppressive medications and hospitalization for cardiovascular events in patients with rheumatoid arthritis. Arthritis Rheum. 2006;54(12):3790–8. pmid:17136752
- 12. Choi MY, Li D, Feldman CH, Yoshida K, Guan H, Kim SC, et al. Comparative risks of cardiovascular disease events among SLE patients receiving immunosuppressive medications. Rheumatology (Oxford). 2021;60(8):3789–98. pmid:33369672
- 13. Lee C, Drobni ZD, Zafar A, Gongora CA, Zlotoff DA, Alvi RM, et al. Pre-existing autoimmune disease increases the risk of cardiovascular and noncardiovascular events after immunotherapy. JACC CardioOncol. 2022;4:660–9.
- 14. Asenjo-Lobos C, González L, Bulnes JF, Roque M, Muñoz Venturelli P, Rodríguez GM. Cardiovascular events risk in patients with systemic autoimmune diseases: a prognostic systematic review and meta-analysis. Clin Res Cardiol. 2024;113(2):246–59. pmid:37650912
- 15. Moriarity C, Parsons V, Jonas K, Schar B, Bose J. Sample design and estimation structures for the National Health Interview Survey, 2016–2025. Vital Health Stat 2. 2022;(191). https://stacks.cdc.gov/view/cdc/115394
- 16. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet. 2007;370(9596):1453–7. pmid:18064739
- 17. Reichenheim ME, Coutinho ESF. Measures and models for causal inference in cross-sectional studies: arguments for the appropriateness of the prevalence odds ratio and related logistic regression. BMC Med Res Methodol. 2010;10:66. pmid:20633293
- 18. Skinner C. Analysis of categorical data for complex surveys. Int Stat Rev. 2019;87:S64–78.