Josephine van Dongen

Chapter 3 54 followed by a sustained reduction in rotavirus detections of 44–75% 10 .This reduced circulation of rotavirus may have influenced the number of rotavirus introductions into the Netherlands. We previously evaluated the cost-effectiveness of infant rotavirus vaccination in the Netherlands and considered three potential vaccination strategies: “no vaccination,” “universal vaccination,” and“targeted vaccination” 11 .The targeted vaccination strategy is a selective vaccination program, including only infants with medical risk conditions predisposing to severe or complicated rotavirus AGE, including prematurity, low bir th weight, and severe congenital pathology 10. No decision has been made yet on the preferred strategy for the Netherlands. Assuming the observed biennial pattern in rotavirus epidemics represents a new epidemiological equilibrium, a re-assessment of the national rotavirus disease burden and the cost-effectiveness of each of these infant rotavirus vaccination strategies is therefore required to inform policy makers. In addition, it is now widely recognized that rotavirus vaccination induces a small but increased risk of intussusception (IS). Because of this serious side effect, an evaluation of risk-benefit ratios has become an integral part of the decision-making process on rotavirus vaccination policy. Our aim was to update our previous model-based health economic evaluation of rotavirus vaccination in the Netherlands by both taking into account the present, lower endemic state, and expanding the analysis by including risk-benefit calculations. Methods Updated rotavirus disease burden The original economic evaluation 11 used data from three observational studies conducted in the Netherlands: (1) the Sensor cohor t study on community AGE 12 and (2) the Netherlands Institute of Primary Health Care (NIVEL) study on AGE in primary care 13 were conducted in 1999 and provided age-stratified data on AGE incidence and the propor tion rotavirus attributed; (3) the RoHo study quantified rotavirus community-acquired hospitalizations and nosocomial infections in children 0–15 years old for the years 2006–2010 with separate estimates for healthy children and those with medical risk conditions 11,14 . Incidence estimates by disease category (rotavirus in the community, GP visits, community-acquired hospitalizations, and nosocomial infections) and by risk status (healthy vs medical risk group) were derived from these studies and used as input for the original cost-effectiveness model.To represent the average rotavirus disease burden over the period 2013–2016, covering two high and two low epidemic seasons, we updated these input parameters using the data sources and methodology as discussed in the following sections. Rotavirus community incidence Virological laboratory surveillance in the Netherlands collects weekly numbers of rotavirus detections from 17 to 21 sentinel laboratories serving primary care, hospitals, and long-term