This study provides a comprehensive analysis of real-world COVID-19 vaccine uptake, effectiveness and safety in immunocompromised people using population-based data. We estimate that over 93% of immunocompromised people in England had received at least one COVID-19 vaccine dose by April 2022, but there were disparities in uptake across ethnic and socioeconomic groups. We also estimated vaccine effectiveness of 91% against severe COVID-19 outcomes in immunocompromised people 2-6 weeks after the third dose, and that vaccination did not result in a higher risk of adverse events in immunocompromised people compared to the general population, confirming that they are safe to use in this clinically vulnerable group.
We observed a decline in vaccine uptake with successive doses in immunocompromised people, which may present concern for rollout of future booster vaccines for clinically vulnerable populations. A contributory factor may be reduced visibility of the COVID-19 vaccination campaign within public perception in context of the end of COVID-19 public health measures. We report significant and persistent health inequalities, both socioeconomic and ethnic, in COVID-19 vaccination uptake within the immunocompromised population in England, consistent with previous findings8. In our analysis, those in the most deprived quintile of deprivation were 20% less likely to receive at least one dose of vaccine compared to those in the most affluent quintile, and were 29% less likely to have a third dose. We also found that compared with people of White ethnicity, uptake of all four vaccine doses was lower in people of Pakistani, Bangladeshi, other Asian, Black Caribbean, and Black African ethnicities. National data in England show that ethnic disparities still persist in COVID-19 vaccine uptake during the 2023 booster rollouts32. These data suggest an urgent need for targeted communication to address specific barriers to COVID-19 vaccine uptake within different ethnic and social groups of the wider immunocompromised population.
While immunogenicity studies have consistently demonstrated that COVID-19 vaccines result in a poorer immune response in immunocompromised people compared to people who are immunocompetent 33-35, COVID-19 vaccines still offer promising protection against severe outcomes. Evidence indicates vaccine effectiveness of 32% to 83% against COVID-19-related hospitalisation and 62% to 95% against COVID-19-related death in immunocompromised people 16-19 27. However, these studies noted some uncertainties in findings, primarily due to the limited number of cases observed in the immunocompromised people. Our findings align with similar studies using electronic health record data in the UK and US reporting vaccine effectiveness against hospitalisation and/or death of around 90% following a third dose in people who are immunocompromised27 36. Consistent with the majority of studies, our results show that while COVID-19 vaccines provide a comparable level of protection, they are slightly less protective against hospitalisation and death in immunocompromised groups compared to the general population or other clinically vulnerable groups36 37.
The study cohort showed a higher proportion of COVID-19-related hospitalisations among immunocompromised individuals compared to the general population. Given that the outcome included outpatient hospital admissions and people with immunosuppressive conditions may have more frequent hospital visits, and have a higher mean age, they may be more likely to have incidental COVID-19 hospital admissions38. Previous reports found that using broader definitions of COVID-19 hospitalisation resulted in lower vaccine effectiveness estimates due to outcome misclassification39. The estimates of vaccine effectiveness may have also been affected by residual confounding due to shielding advice that was targeted towards clinically vulnerable groups in the UK6. We observed a progressive decrease in vaccine protection over time following each vaccine dose, supporting the need for booster doses in immunocompromised people. Further research including longer-term follow-up post-vaccination and assessing effectiveness of additional booster vaccine doses is required to determine the number of booster doses required and the optimum dosing interval.
Our findings showed that the risk of adverse events following a first, second or third dose of COVID-19 vaccine was similar in the immunocompromised population compared to the general population. Although we identified a potential increased risk of ankylosing spondyloarthritis following a second dose with ChAdOX1 and of multiple sclerosis following a third dose of BNT162b2 in immunocompromised people relative to the general population, these associations were no longer significant when adjusted for multiple comparisons. The safety of COVID-19 vaccination in immunocompromised populations has been debated due to previous studies linking COVID-19 vaccines to some immune-related adverse events, such as Guillain-Barre syndrome, myocarditis and immune thrombocytopenias, but these have generally been very rare occurrences22-24 40. Evidence suggests that booster vaccines are not associated with additional risk41. There have been case reports of multiple sclerosis presentation42-44 and rheumatic disease flares45 following COVID-19 vaccine exposure, however, there is little evidence for vaccination causing new-onset disease46. Furthermore, both ankylosing spondyloarthritis and multiple sclerosis are complex conditions typically diagnosed over a period of time, involving referral to a rheumatological47 or neurological specialist48, therefore any new diagnoses recorded within 28 days following vaccination are unlikely to be directly associated with the vaccine.
To our knowledge, this is the largest study of COVID-19 vaccine uptake, safety and effectiveness in immunocompromised populations. There are some limitations in our study. Firstly, we categorised our target population using hospital records with diagnostic codes or drug prescriptions related to having an immunocompromised condition for a 6-month period before the study start date (24 months for bone marrow transplant patients), which may have led to misclassification biases. Second, due to a limited number of severe outcome cases after the fourth dose within the immunocompromised population, our assessment of vaccine effectiveness and safety was primarily centred on the effectiveness and safety of the vaccine for the initial three doses, as reflected in our subgroup analyses. Third, despite conforming with pre-defined definitions used in national reports31, classification of outcomes according to presence of a positive COVID-19 test within a specified time period before a hospital admission or death may misclassify cases where COVID-19 is an incidental finding rather than the primary reason for admission or death. Fourth, we did not investigate adverse events related to worsening or exacerbation of pre-existing immune-related conditions, which may have differed from new presentations.
In summary, we found that overall uptake of COVID-19 vaccines in immunocompromised people in England was high, but that there were disparities in uptake across different ethnic and socioeconomic groups. We also found that two to three doses of any COVID-19 vaccine were protective against COVID-19-related hospitalisation, ICU admission and death in immunocompromised people, and that vaccine effectiveness was comparable to the general population . The safety profile of COVID-19 vaccines was found to be comparable in immunocompromised people and the general population. These findings underscore the importance of ongoing vaccination prioritisation for immunocompromised individuals to maximise the protection against severe COVID-19-related outcomes.