P. Scuffham et al., The cost-effectiveness of introducing a varicella vaccine to the New Zealand immunisation schedule, SOCIAL SC M, 49(6), 1999, pp. 763-779
This study examined the cost-effectiveness of adding a varicella vaccine to
an existing, childhood immunisation schedule relative to a counterfactual
where the varicella vaccine is available on a user-pays basis (the current
New Zealand situation). The costs and consequences of chickenpox in an annu
al cohort of 57,200, ii-month old children were simulated for a 30-year per
iod. The cohort simulation design captures the 'phasing-in' effects of rout
ine varicella vaccination on the population.
From a health care payer's perspective (medical costs only) every dollar in
vested in a vaccination programme would return NZ $0,67, However, from a so
cietal point of view (which includes the value of work-loss), a vaccination
programme would return NZ $2,79 for every dollar invested. To implement a
varicella vaccination programme covering 80% of 15-month old children in Ne
w Zealand would add more than NZ $1 million in net direct (health care) cos
ts each year. However, the indirect cost savings from reduced losses of wor
k-time exceed NZ $2 million annually.
The net average health care cost per child vaccinated over the 30-year mode
lling period was $54 whereas the cost-savings from work-loss averted averag
ed $101 per child vaccinated. Total cost-savings to society of $47 per chil
d vaccinated, on average, could be gained from a vaccination programme.
The finding that the addition to vaccination costs resulting from a routine
programme (including the cost of complications from the vaccine) were grea
ter than the offsetting health cart: cost savings from reduced incidence of
chickenpox were robust to a sensitivity analysis on all assumptions within
plausible ranges. Overall cost-effectiveness estimates were most sensitive
to assumptions regarding lost work-time, the discount rate, and the price
and efficacy of the vaccine. Estimates were relatively insensitive to chang
es in assumptions regarding health care utilisation. (C) 1999 Elsevier Scie
nce Ltd. All rights reserved.