Purpose: The aim of this study is to measure the seasonal trends of adenovirus conjunctivitis.
Methods: A total of 926 positive cases of HAdV conjunctivitis were reviewed at the University of Pittsburgh Medical Center (UPMC) in Pittsburgh (Allegheny County, Pennsylvania, USA) from 11th March 1987 to 15th December 2016. To determine seasonal variation, daily counts were adjusted for yearly changes in Allegheny County population size, modelled as a function of long-term trend using Basis-spline (B-spline), and analysed for seasonal trend by quarter (1Q = January to March, 2Q = April to June, 3Q = July to September, 4Q = October to December).
Results: The number of adenovirus conjunctivitis cases in 3Q was 45% higher than that of 2Q (95% confidence interval [CI] 1.21-1.74, P = 0.0001). When 3Q is compared to 1Q and 4Q, it was not statistically significant (95% CI 0.96-1.42, P = 0.1209 and 95% CI 0.96-1.41, P = 0.1180, respectively).
Conclusion: To the best of our knowledge, this is the first report and the longest time-series evaluating the potential seasonality of adenovirus conjunctivitis in the USA. We observed the highest number of cases from July to September and the lowest number of cases from April to June in our study population. Knowledge of seasonality can be used to guide outbreak expectations, limit antibiotic over-prescription, and enhance disease prevention.
Keywords: Conjunctivitis; adenovirus; epidemic keratoconjunctivitis; observational study; seasonality.