Temporal and spatial associations between influenza and asthma hospitalisations in New York City from 2002 to 2012: A longitudinal ecological study

Abstract

Objectives To determine whether asthma hospitalisations of children and adults in the five boroughs of New York City are correlated with influenza hospitalisations temporally and spatially. Design A longitudinal ecological study. Inclusion criteria We reviewed the Statewide Planning and Research Cooperative System's records of hospitalisations in Manhattan, Bronx, Queens, Brooklyn and Staten Island from 2002 to 2012. All hospitalisations with a primary diagnosis of either asthma or influenza were identified using the International Classification of Diseases, Ninth Revision, Clinical Modification diagnosis codes. Methods A time-series regression analysis was performed using aggregate monthly counts of influenza hospitalisations as predictors of asthma hospitalisations. Time-series regression models were also applied to different age groups and boroughs to examine the magnitude of influenza and asthma correlations across strata. The per cent excess risk was also calculated across age groups and boroughs. Results Time-series analysis of the overall population revealed a significant positive correlation between influenza and asthma hospitalisations (p=0.011). When stratifying by age, there was a significant positive correlation between asthma and influenza hospitalisations for individuals 18 and older (p<0.01), and no significant correlation found for age groups younger than 18. Percentages of excess risk of influenza-related asthma hospitalisations also increased with increasing age with adults 18-44, 45-64 and 65+ having excess risk percentages of 2.9%, 3.4% and 4%, respectively. Time-series analysis by location revealed positive significant correlations between asthma and influenza hospitalisations in Brooklyn (p=0.03) and Manhattan (p<0.01). Manhattan and Brooklyn had a 2.5% and 1.6%, respectively, percentage of excess risk of influenza-related asthma hospitalisations. Conclusion Influenza and asthma hospitalisations are significantly associated at the population level among adults. These associations vary by age and geographical location. Influenza prevention strategies targeting adult populations, particularly individuals living in Manhattan and Brooklyn, have the potential for meaningful reduction of influenza-related asthma hospitalisations.