The Impact of Local and Oceanic Climate Variability on the Incidence of Childhood Pneumonia in Papua New Guinea

摘要

Background: Childhood pneumonia is common in tropical regions, and a number of studies have associated pneumonia risk with climate and other factors. However, the findings have often been contradictory and unconvincing. The aim of this study was to evaluate quantitatively the association between geographic and climate factors on the incidence of childhood pneumonia in Papua New Guinea (PNG). Aims: To quantitatively evaluate the association between geographic and local and regional climate factors on the incidence of childhood pneumonia in Papua New Guinea (PNG). Methods: Based on the National Health Information System and the National Weather Service data in PNG, pneumonia incidence under 5-year-olds and the meteorological factors were obtained from six areas, including monthly rainfall (mm), monthly mean of the daily maximum temperatures (°C) of the period of 1997-2006. As regional climate factors, the southern oscillation index (SOI) and the dipole mode index (DMI) were used to explicate the oceanic oscillations. After testing linearity of effects of the climate variables using a generalized additive model, a generalized linear model was constructed to measure the effect size of each climate factor. Pooled effects of the climate variables among 6 sub-areas of PNG were estimated using meta-analysis. Results: Pooled risk per every 10 mm increase of rainfall is 0.24% and that of risk per every 1°C increase of the monthly mean of the maximum daily temperatures is 4.88%. SOI and DMI showed overall negative effect on childhood pneumonia incidence (-0.57%, -4.3%, respectively) and the dry season had a higher risk of pneumonia compared to the rainy season (Pooled: 12.08%). Conclusions: We verified the relationships between local and regional climate factors and risk of childhood pneumonia in PNG, and also found that the effect size can be different according to geographic features and local climate variables. Evaluation of health effect based on local and regional climate factors reflect the impact of local health determinants.