The aim of the study is to analyze the annual cycle of pediatric medically attended respiratory illnesses. Data on 141 million pediatric respiratory visits from the years 2010-2019 were obtained from the Polish National Healthcare Fund. To identify underlying patterns and trends within the aggregated data, techniques like seasonal-trend decomposition using LOESS (STL) and principal component analysis (PCA) were applied. A strongly recurring pattern was observed. Following the annual minimum in late summer, there was a sudden surge in upper respiratory infections in early September. Subsequently, overall visits declined gradually, while the share of lower respiratory infections increased, particularly during the influenza peaks from January to March. Afterwards, visits declined steadily, with an additional peak of tonsillopharyngitis noted in midsummer. Dimensionality reduction of diagnoses implied the existence of two major groups of co-occurring diagnoses, the proportions of which change over the year: one smaller but more severe, peaking during the influenza season, and the second dominating with lower severity. Age differences in diagnoses were observed, with babies showing upper respiratory infections likely diagnosed with the common cold rather than a more specific upper respiratory infection. While enhancing surveillance strategies is indeed a desirable long-term goal, it is worth noting that despite the variability observed in the onset of the influenza season, the infection cycles generally follow a relatively fixed pattern. This consistency provides a foundation for effective planning and underscores the potential for proactive measures to mitigate the impact of seasonal outbreaks.
Read full abstract