COVID Vaccines Cut Infection and Death Risk Sharply in Dialysis Patients, Nationwide Study Finds

People with end-stage kidney disease who rely on dialysis to stay alive have long been counted among the most vulnerable victims of the COVID-19 pandemic. Their kidneys have failed, their immune systems are blunted by chronic inflammation and the dialysis procedure itself, and their treatment schedule forces them into crowded clinics several times a week, making isolation virtually impossible. Early in the pandemic, studies from several countries reported that roughly two-thirds of dialysis patients who caught SARS-CoV-2 ended up in hospital, and about one in four died within a month of infection. Now, one of the largest real-world analyses ever conducted in this population offers strong reassurance that vaccination works for them too, even if their antibody responses are weaker than those of healthy people.

A nationwide retrospective cohort study from South Korea, published in the journal Heliyon, followed essentially the entire population of dialysis patients in the country during the first year vaccinations became available. The researchers, led by Jae Young Kim and Tae Ik Chang of the National Health Insurance Service Ilsan Hospital, drew on a linked database combining the Korea Disease Control and Prevention Agency’s COVID-19 records with the National Health Insurance Service, the single compulsory insurer covering every Korean citizen. Because the insurance system assigns permanent special codes to patients receiving maintenance dialysis for at least three months, the team could identify the full national population of people on hemodialysis or peritoneal dialysis with remarkable reliability.

The starting cohort included 105,323 adults on maintenance dialysis as of February 26, 2021, the day COVID-19 vaccinations began in South Korea. After excluding patients already infected with SARS-CoV-2 or lacking complete baseline data, 103,130 individuals remained under observation through December 31, 2021. Strikingly, 82.3 percent of them — 84,895 people — completed a full vaccine series during that window, a testament to the country’s policy of prioritizing dialysis patients for immunization. The vaccines used included the Pfizer-BioNTech mRNA vaccine BNT162b2, the AstraZeneca vector vaccine ChAdOx1 nCov-19, the Moderna mRNA vaccine mRNA-1273, and a tiny fraction of the single-dose Janssen vaccine Ad26.COV2.S; 42.5 percent of second doses were heterologous, meaning they mixed vaccine platforms.

To make a fair comparison, the investigators used propensity score matching, a statistical technique that pairs vaccinated and unvaccinated patients with nearly identical profiles of age, sex, income, residence, dialysis modality, dialysis duration, and a long list of comorbidities including hypertension, diabetes, heart failure, cerebrovascular disease, dementia, and cancer. Before matching, the unvaccinated group was older and sicker, reflecting hesitancy and access problems among the frailest patients. After nearest-neighbor matching with a caliper of 0.10, the final analytical cohort comprised 14,498 fully vaccinated and 14,498 unvaccinated patients with standardized differences below 0.1 on every measured characteristic — a near-perfectly balanced comparison. The average age was 63.2 years, just over half were men, roughly 89 percent were on hemodialysis, and about half had been on dialysis for five years or more.

The results were unambiguous. Over a median follow-up of 150 days, 375 participants — 1.3 percent of the matched cohort — experienced a first laboratory-confirmed SARS-CoV-2 infection, verified by positive reverse transcription polymerase chain reaction testing. Among vaccinated patients, 0.88 percent were infected, compared with 1.71 percent of unvaccinated patients. In Cox proportional hazards regression, vaccination was associated with a 53 percent lower risk of infection, corresponding to a hazard ratio of 0.47 with a 95 percent confidence interval of 0.38 to 0.58. Kaplan-Meier curves showed the cumulative incidence of infection diverging steadily between the groups throughout the observation period.

The protection against severe disease was even more pronounced. The study defined serious COVID-19 as a composite of severe infection — hospitalization requiring any form of respiratory support such as noninvasive ventilation, high-flow oxygen, or invasive mechanical ventilation, or treatment with extracorporeal membrane oxygenation or continuous renal replacement therapy — together with COVID-19-related death, meaning death during hospitalization or within 28 days of a positive test. Serious outcomes occurred in 0.26 percent of vaccinated patients versus 0.74 percent of unvaccinated patients, a hazard ratio of 0.32. Analyzed separately, severe COVID-19 was reduced by 70 percent (hazard ratio 0.30) and COVID-19-related death was likewise reduced by 70 percent (hazard ratio 0.30). In other words, even when breakthrough infections occurred after two doses, they were far less likely to turn deadly in vaccinated dialysis patients.

Subgroup analyses reinforced the message across clinically meaningful strata. The association between vaccination and lower risks of both infection and serious outcomes held consistently for patients younger and older than 65, for men and women, for those with and without diabetes or ischemic heart disease, and for those on dialysis for fewer or more than five years. This robustness matters because dialysis patients are extraordinarily heterogeneous: many carry a heavy burden of cardiovascular and metabolic disease, and longer dialysis vintage, anemia, and immunosuppressive therapy have all been linked to poorer antibody responses after mRNA vaccination.

The findings address a persistent scientific worry. Kidney failure induces immune alterations and chronic low-grade systemic inflammation, dialysis itself has immunomodulatory effects, and many patients take immunosuppressive drugs — all of which can interfere with the generation of protective immunity. Previous studies reported post-vaccination antibody levels in dialysis patients ranging from as low as 7 percent up to 66 percent of those seen in healthy controls, and a meta-analysis of 32 studies found an overall immunogenicity rate in dialysis patients of 86 percent relative to non-dialysis individuals. Yet the Korean study, which did not have antibody measurements available, demonstrates that this blunted immunogenicity does not necessarily translate into weaker clinical protection. Its results align with real-world effectiveness studies from the United States and Canada, suggesting that cellular immunity and other mechanisms may compensate for lower antibody titers. The protection observed was nonetheless more modest than the 94 percent efficacy against symptomatic infection reported in randomized trials of the general population, which had excluded nearly all patients with moderate to advanced chronic kidney disease — only 0.7 percent of trial participants had such disease.

The study’s strengths are considerable. By covering the entire national ESKD dialysis population, including home-based peritoneal dialysis patients often omitted from clinic-based research, it avoided the selection biases of single-network studies. Its time-to-event design started follow-up only after completion of the vaccine series plus a 14-day immune-response window, so the estimated risks reflect the period when vaccination would be expected to confer protection. Defining the primary outcome as laboratory-confirmed infection rather than symptomatic illness also sidestepped a practical problem: some dialysis patients cannot reliably complete symptom questionnaires because of language or cognitive limitations. Frequent RT-PCR screening, mandatory for in-center hemodialysis patients before entering their units, made undetected infections rare.

Limitations remain, and the authors are candid about them. As an observational study, residual confounding cannot be excluded; data on obesity, immunosuppressive use, dialysis adequacy, and residual kidney function were unavailable, and unmeasured risky health behaviors among the unvaccinated could have influenced censoring. The follow-up was too short to assess how long protection lasts, viral variant data were absent, and the entire study period preceded both the Omicron wave and the rollout of third booster doses, so the findings speak to the pre-Omicron era. The cohort was also entirely Korean, warranting caution when generalizing to other populations. Still, the core conclusion stands: complete COVID-19 vaccination substantially reduced the risk of SARS-CoV-2 infection, severe disease, and death in one of the pandemic’s most susceptible populations. The authors argue the evidence supports health authorities’ policies to promote and prioritize vaccination for dialysis patients, and to combat the vaccine hesitancy that has kept uptake low in some countries — a population with much to gain from a vaccine, regardless of type.

Subject of Research: Effectiveness of COVID-19 vaccination in preventing SARS-CoV-2 infection and severe outcomes in patients with end-stage kidney disease on maintenance dialysis

Article Title: Impact of COVID-19 vaccination on infection risk and clinical outcomes in patients with end-stage kidney disease undergoing dialysis: A nationwide retrospective study in South Korea

Article References: Kim, J. Y., Son, K. J., Park, P. G., & Chang, T. I. (2026). Impact of COVID-19 vaccination on infection risk and clinical outcomes in patients with end-stage kidney disease undergoing dialysis: A nationwide retrospective study in South Korea. Heliyon, 12(15), Article e45541. https://doi.org/10.1016/j.heliyon.2026.e45541

Image Credits: AI Generated

DOI: 10.1016/j.heliyon.2026.e45541

Keywords: COVID-19 vaccination, end-stage kidney disease, dialysis, SARS-CoV-2, vaccine effectiveness, South Korea, nationwide cohort study, hemodialysis, peritoneal dialysis, propensity score matching, severe COVID-19, immunocompromised patients