Herpes zoster incidence in Germany - an indirect validation study for self-reported disease data from pretest studies of the population-based German National Cohort


  • M. Caputo
  • J. Horn
  • A. Karch
  • M.K. Akmatov
  • H. Becher
  • B. Braun
  • H. Brenner
  • S. Castell
  • B. Fischer
  • G. Giani
  • K. Günther
  • B. Hoffmann
  • K.H. Jöckel
  • T. Keil
  • B. Klüppelholz
  • L. Krist
  • M.F. Leitzmann
  • W. Lieb
  • J. Linseisen
  • C. Meisinger
  • S. Moebus
  • N. Obi
  • T. Pischon
  • S. Schipf
  • B. Schmidt
  • C. Sievers
  • A. Steinbrecher
  • H. Völzke
  • R. Mikolajczyk


  • BMC Infectious Diseases


  • BMC Infect Dis 19 (1): 99


  • BACKGROUND: Until now, herpes zoster (HZ)-related disease burden in Germany has been estimated based on health insurance data and clinical findings. However, the validity of self-reported HZ is unclear. This study investigated the validity of self-reported herpes zoster (HZ) and its complication postherpetic neuralgia (PHN) using data from the pretest studies of the German National Cohort (GNC) in comparison with estimates based on health insurance data. METHODS: Data of 4751 participants aged between 20 and 69 years from two pretest studies of the GNC carried out in 2011 and 2012 were used. Based on self-reports of physician-diagnosed HZ and PHN, age- and sex-specific HZ incidence rates and PHN proportions were estimated. For comparison, estimates based on statutory health insurance data from the German population were considered. RESULTS: Eleven percent (95%-CI, 10.4 to 12.3, n = 539) of the participants reported at least one HZ episode in their lifetime. Our estimated age-specific HZ incidence rates were lower than previous estimates based on statutory health insurance data. The PHN proportion in participants older than 50 years was 5.9% (1.9 to 13.9%), which was in line with estimates based on health insurance data. CONCLUSION: As age- and sex-specific patterns were comparable with that in health insurance data, self-reported diagnosis of HZ seems to be a valid instrument for overall disease trends. Possible reasons for observed differences in incidence rates are recall bias in self-reported data or overestimation in health insurance data.