Time-series analysis of seasonal patterns for pneumonia deaths in Peru, 2003-2017 period

Abstract

Introduction. It is generally accepted that respiratory infections are seasonal, but high-frequency periods are rarely identified at the local level in a country with diverse climates. Anticipating the pneumonia season locally can provide a better use of critical resources. Objective. The aim of the study was to examine seasonal variability in monthly deaths due to pneumonia in Peru. Methods. Observational retrospective study using time series analysis to identify periods of high pneumonia mortality in seven cities in Peru. We reviewed administrative reports from the Ministry of Health for pneumonia deaths during 2003-2017. Descriptive statistics were calculated and a time series analysis at a monthly scale was performed on the frequency of deaths due to pneumonia, cumulative rainfall, and maximum and minimum environmental temperatures. Results: 94.4% of pneumonia deaths (N = 166,844) were reported as pneumonia due to “unspecified organism”, and 75.6% were adults older than 65 years. In the cities of Tarma, Arequipa and Cusco, minimum ambient temperature is negatively correlated to pneumonia deaths in all age groups and at least one risk group. In Iquitos, minimum temperature is positively correlated with deaths among children under 5 years of age. The cities of Pucallpa and Cajamarca reported no statistically significant correlation. The climate in Lima is a peculiar case. The distribution of pneumonia deaths throughout the year suggests a north-south sequence, while the climate space-time analysis suggests a south-north pattern. Conclusion: Results show different seasonal patterns for pneumonia deaths in different cities and risk groups.