Understanding risk factors for herpes zoster and postherpetic neuralgia in UK primary care: investigations to inform vaccine policy.

Background: Herpes zoster affects millions of people worldwide each year and many go on to suffer long-term pain, called postherpetic neuralgia (PHN). As zoster is common and PHN is difficult to treat, preventing zoster through vaccination is important. This thesis aims to better understand risk factors for zoster and PHN, in order to inform vaccination policy. Methods: Three large observational studies were carried out using primary care data from the UK Clinical Practice Research Datalink and linked secondary care data from the Hospital Episodes Statistics. First, a matched case-control study quantified the effects of possible risk factors for zoster and explored whether their effects differed by age group. Second, a descriptive study looked at antiviral prescription patterns and patient characteristics associated with antiviral receipt after zoster diagnosis. Third, a cohort study assessed risk factors for PHN and investigated whether their effects were modified by antiviral use. Results: The case-control study of zoster risk factors included 144,959 zoster patients and 549,336 controls and found an increased risk of zoster among patients with rheumatoid arthritis, systemic lupus erythematosus, inflammatory bowel disease, chronic obstructive pulmonary disease, asthma, chronic kidney disease, depression and type 1 diabetes; odds ratios ranged from 1.14 to 1.72. In general, the relative effects of these risk factors on zoster decreased with increasing age. In the descriptive study of antiviral use, of 142,216 zoster cases, only 58.1% received an antiviral prescription at zoster diagnosis. Antivirals were even under-prescribed among the immunosuppressed and older individuals, for whom guidelines recommend routine treatment. The cohort study of PHN risk factors identified 119,413 zoster patients, 5.8% of whom developed PHN. An increased risk of PHN was found among patients with rheumatoid arthritis, systemic lupus erythematosus, inflammatory bowel disease, chronic obstructive pulmonary disease, asthma, depression, type 2 diabetes, lower socioeconomic status, smoking and under- or overweight; odds ratios ranged from 1.13-1.82. Antiviral use was not associated with PHN risk overall. The zoster case-control and PHN cohort study showed that patients with severely immunosuppressive conditions were at greatest risk of both zoster and PHN. Conclusions: A number of patient characteristics and comorbidities were associated with increased zoster and PHN risks. Patients at highest risk of zoster and PHN are those of older age and those with immunosuppression; currently, patients with immunosuppression are not eligible for vaccination, highlighting a need for alternative risk reduction strategies in this group. Low antiviral use at zoster diagnosis suggests treatment guidelines be revised to encourage greater use, especially among the immunosuppressed and older individuals who are recommended, but not routinely given, antivirals. Research on the cost-effectiveness of vaccinating patients with specific risk factors is needed.