Data Analysis
Two one-year periods, October 2017 to September 2018 (year 1) and
October 2018 to September 2019 (year 2), were studied to assess seasonal
differences in population-based incidence and clinical burden for the
respiratory viruses detected. To calculate population-based incidence
per 100,000 persons for each virus, the number of adults with a
laboratory-confirmed respiratory virus was divided by the adjusted 2010
U.S. Census population estimate for the catchment area. Population-based
incidence and 95% confidence intervals were calculated overall and for
four age strata: 18-49, 50-64, 65-79, and > 80 years
of age [9]. If more than one virus was detected in a single
specimen, each virus contributed to the respective incidence
calculation. The population estimate was adjusted by the hospitals’
percent market share for zip codes as determined by the New York
Statewide Planning and Research Cooperative System (SPARCS) [10]. To
improve the incidence estimate’s reliability, the catchment area was
defined as the eight zip codes in which the hospitals had> 60% market share (10032, 10033, 10034, 10040,
10452, 10453, 10463, and 10471).
To assess the clinical burden of specific viruses, the following
outcomes associated with each virus were determined: the median hospital
LOS, the proportion of patients who had an ICU admission, the ICU LOS,
and in-hospital all-cause mortality. Time spent in the emergency
department prior to being admitted to an inpatient unit was included in
LOS calculations. If more than one virus was detected in a single
specimen, each virus contributed to the respective median LOS, ICU
admission and ICU LOS, and in-hospital mortality which was determined
from recorded death notes in the EHR. Two sample tests of proportions
were used to compare outcomes (ICU admissions and mortality) between
each virus and all others (e.g., adenovirus vs. all other viruses).
Median hospital LOS and median ICU LOS were compared using the
Mann-Whitney test for each virus vs. all others.
The population-based incidence and clinical burden of influenza viruses
(combined A H1, A H3, and B) were compared with those of non-influenza
viruses (combined CoV229E, CoVHKU1, CoVNL63, CoVOC43, PIV types 1-4,
RSV, hMPV, and AV) using the two sample test of proportions. RV/EV were
excluded from the comparative analyses of the burden of influenza vs.
non-influenza viruses. Patients with co-detections of influenza and
non-influenza viruses were excluded from these comparative analyses.