[PIN54] Adaptation & calibration of a UK model of meningococcal disease to the US setting

[PIN54] Adaptation & calibration of a UK model of meningococcal disease to the US setting

2008 Value in health

Aballea, S. | McGarry, L.J. | Taylor, D.C. | Hill, G. | Pawar, V. | Clements, K.M. | Thompson, D. | Volume: 11, Issue: 6, Pages: A445-A446, Disease, model, Vaccination, methods, vaccines, Risk,

OBJECTIVES: To evaluate vaccination policies for meningococcal
disease in the US, we adapted and calibrated a dynamic model
of meningococcal disease originally developed for the UK.
METHODS: We adapted an age-stratified mathematical model
of meningococcal transmission, carriage, and disease from the
UK to the US setting. The model focus was expanded from
serogroup C to serogroups A, C, W, and Y disease to reflect
newer vaccines. The most recent available US cross-sectional
carriage data were used to extract serogroups A, W, and Y
carriage from the non-C (“other”) serogroup data used in the UK
model. We used recently-published contact matrix data for transmission
calculations and performed formal calibrations to published
US surveillance data. We first calibrated age-specific force
of infection to ACWY carriage data and solved for transmission
rates (b) at equilibrium for a cohort of 75,000 individuals representative
of the US population. We then calibrated the risk ofmeningococcal disease
conditional on carriage (q) to US incidencedata. Calibration used the
maximum likelihood method;
goodness-of-fit was assessed using deviance. RESULTS: The calculated
b for infant-to-infant transmission is 1.86 effective contacts
per 100,000 infants per year. b peaks at 29.7/100,000 for
transmission between 14 year olds and 11-14 year olds, and
ranges from 0 to 16.3/100,000 for adult-adult transmission. The
q is highest in infants (0.015), declines steeply and plateaus at
0.00013 in those aged >20 years. The model predicted yearly
cases of ACWY-related meningococcal disease are 132 among
infants, 439 among 1-20 year-olds, and 202 among those aged
>20, compared with 129, 433, and 211 for US benchmarks
(deviance = 184, with 72 degrees of freedom). CONCLUSIONS:
Calibration techniques can be used to adapt a UK meningococcal
disease model to the US using modified serogroup and contact
data such that a reasonable fit is achieved with US benchmarks.