Incidence rates of bullous pemphigoid, herpes zoster, and urticaria following the start of the COVID-19 vaccination programme in the UK: a population-based cohort study

Background

Several population-based studies have reported an increased risk of serious skin conditions following COVID-19 vaccinations, but their population samples were not representative. Accurately estimating these risks is important for informing vaccine hesitancy and clinical practice.

Objectives

To compare the incidence rates (IRs) of bullous pemphigoid (BP), herpes zoster (HZ), and urticaria before and after the UK's COVID-19 vaccination programme started.

Methods

We conducted a population-based cohort study using electronic healthcare records from the Clinical Practice Research Datalink (01/01/2019-31/05/2023). We estimated monthly IRs (cases/100 000 person-years) for BP, HZ, and urticaria. Using interrupted time series analysis, we compared changes in the IR slopes before and after the vaccine programme started (intervention). Sensitivity analyses adjusted for COVID-19 lockdowns and total GP appointments.

Results

16 156 639 adults (≥18) registered at a general practice between 01/01/2019 and 31/05/2023 (cases of BP: 3506, HZ: 129 668, urticaria 63 320).

BP IR (cases/100 000 person-years) did not change post-intervention compared to the pre-intervention period. We did not detect pre- and post-intervention slopes. Sensitivity analysis results were similar.

Pre-intervention, there was a downward slope for the HZ IR (-4.44; 95% CI -5.54, -3.35; p<0.005). HZ IR increased post-intervention compared to the pre-intervention period (5.15; 95% CI 3.43, 6.88; p<0.005). Despite the increase in HZ IR, we found no evidence of a post-intervention slope. Sensitivity analysis results were similar.

Pre-intervention, there was a downward slope for the urticaria IR (-3.42; 95% CI -4.13, -2.70; p<0.005). Urticaria IR increased post-intervention compared to the pre-intervention period (4.40; 95% CI 3.27, 5.53; p<0.005), resulting in an upward slope (0.98; 95% CI 0.43, 1.52; p<0.005). However, after accounting for lockdowns or GP appointments, the upward post-intervention slope was no longer apparent.

Conclusions

While the BP IR did not change following the vaccine programme, we detected increased HZ and urticaria IR. However, these increases were modest, and for many, the benefits of vaccination will likely outweigh the potential risks. Future research should use individual patient-level studies (e.g., self-controlled case series) to determine whether the increases in HZ and urticaria IRs are driven by COVID-19 vaccines.