Lifetime prevalence of herpes zoster and simplex in electronic health records compared with self-reported lifetime prevalence from population surveys and cohorts in England

<ns5:p>Background Herpesviruses are common viruses that persist lifelong after infection. Reactivation of varicella zoster leads to herpes zoster, while herpes simplex virus reactivation can cause cold sores or genital ulcers. It is unclear how well herpesvirus reactivation is recorded in electronic health records (EHRs). This study compared lifetime prevalence of herpes zoster and herpes simplex diagnoses in EHRs with self-reported lifetime prevalence from nationally representative survey and cohort data to inform future studies. Methods The survey/cohort samples included i) adults aged ≥16 years from Health Survey for England (HSE), 2015 with self-reported data on herpes zoster; ii) HSV-1 seropositive adults aged 40 to 89 years from EPIC-Norfolk with data on frequent cold sores from 2002 to 2004; iii) individuals aged 16 to 74 years from England participating in Natsal-3 from 2010 to 2012 with self-reported data on genital herpes. For each sample, comparable populations were identified using EHR data from i) Clinical Practice Research Datalink (CPRD), and ii) CPRD linked with Hospital Episode Statistics. The lifetime prevalence of herpes zoster, cold sores and genital herpes was calculated, with 95% confidence intervals. Results Lifetime prevalence of herpes zoster among HSE participants was 11.5% (95% CI: 10.7–12.4%) compared with 3.5% (95% CI: 3.5–3.5%) in CPRD. Among EPIC-Norfolk HSV-1 seropositive participants, 17.6% (95% CI: 16.6–18.7%) reported frequent cold sores versus 0.9% (95% CI: 0.9–0.9%) in the CPRD sample. In Natsal-3, lifetime genital herpes prevalence was 1.0% (95% CI: 0.7–1.4%) among males and 1.4% (95% CI: 1.1–1.7%) among females compared with 0.2% (95% CI: 0.2–0.2%) and 0.6% (95% CI: 0.6–0.6%) among comparable males and females respectively in CPRD. Conclusions EHR data show lower lifetime prevalence of alpha herpesviruses, especially cold sores and genital herpes, than self-reported figures from population surveys and cohorts. Careful choice of data sources is needed for studies of these viruses.</ns5:p>