Abstract
Influenza is responsible for a large number of deaths which can only be estimated using modelling methods. Such methods have rarely been applied to describe the major socio-demographic characteristics of this disease burden.
We used quasi Poisson regression models with weekly counts of deaths and isolates of influenza A, B and respiratory syncytial virus for the period 1994 to 2008.
The estimated average mortality rate was 13.5 per 100,000 people which was 1.8% of all deaths in New Zealand. Influenza mortality differed markedly by age, sex, ethnicity and socioeconomic position. Relatively vulnerable groups were males aged 65–79 years (Rate ratio (RR) = 1.9, 95% CI: 1.9, 1.9 compared with females), Māori (RR = 3.6, 95% CI: 3.6, 3.7 compared with European/Others aged 65–79 years), Pacific (RR = 2.4, 95% CI: 2.4, 2.4 compared with European/Others aged 65–79 years) and those living in the most deprived areas (RR = 1.8, 95% CI: 1.3, 2.4) for New Zealand Deprivation (NZDep) 9&10 (the most deprived) compared with NZDep 1&2 (the least deprived).
These results support targeting influenza vaccination and other interventions to the most vulnerable groups, in particular Māori and Pacific people and men aged 65–79 years and those living in the most deprived areas.
•Total influenza-associated mortality estimated by modelling was about 23 times higher than measured influenza mortality.•Circulatory illness caused the largest proportion of influenza deaths, unlike hospitalisations where respiratory predominate.•This is the first reported use of modelling to estimate influenza mortality disparities by ethnicity and socioeconomic status.•Influenza mortality was associated with increasing age, male gender, Māori & Pacific ethnicity, and neighbourhood deprivation.•Future research should investigate ways of reducing these marked health inequities in influenza deaths.