Waliullah Md, Hossain Md Jamal, Hasan Md Raqibul, Hannan Abdul, Rahman Mohammad Mafizur
Department of Applied Mathematics, Noakhali Science and Technology University, Noakhali, 3814, Bangladesh.
School of Business, University of Southern Queensland, Queensland (QLD), 4350, Australia.
BMC Public Health. 2025 Jan 24;25(1):318. doi: 10.1186/s12889-024-20920-z.
Diarrheal infections continue to be a major public health concern in Bangladesh, especially in urban areas where population density and environmental variables increase dissemination risks. Identifying the intricate connections between weather variables and diarrhea epidemics is critical for developing effective public health remedies.
We deploy the novel approach of Wavelet-Autoregressive Integrated Moving Average with Exogenous Variable (WARIMAX) and the traditional Autoregressive Integrated Moving Average with Exogenous Variable (ARIMAX) technique to forecast the incidence of diarrhea by analyzing the influence of climate factors.
Higher temperatures are associated with greater diarrheal occurrences, demonstrating the vulnerability of diarrheal epidemics to weather fluctuations. The Wavelet-ARIMAX method, which uses wavelet analysis within the ARIMAX structure, is better at forecasting performance and model fit than the standard ARIMAX model. Based on climatic variables, Wavelet-ARIMAX can accurately predict diarrheal occurrence, as indicated by the mean absolute error (MAE), root mean squared error (RMSE), and root mean squared logarithmic error (RMSLE). The outcomes highlight the necessity of employing advanced time-series modeling tools such as Wavelet-ARIMAX to better understand and anticipate climate-health interactions. Wavelet-ARIMAX uses wavelet analysis to identify time-varying patterns in climate-disease interactions, providing useful insights for public health initiatives.
The results of this research have implications for climate-adaptive health planning, emphasizing the need for focused actions to reduce the impact of climate change on diarrheal illness burdens in towns and cities.
腹泻感染仍是孟加拉国的一个主要公共卫生问题，特别是在城市地区，那里的人口密度和环境变量增加了传播风险。识别天气变量与腹泻流行之间的复杂联系对于制定有效的公共卫生补救措施至关重要。
我们采用带外生变量的小波自回归积分移动平均（WARIMAX）这一新颖方法以及传统的带外生变量的自回归积分移动平均（ARIMAX）技术，通过分析气候因素的影响来预测腹泻发病率。
较高温度与更多腹泻病例相关，表明腹泻流行易受天气波动影响。在ARIMAX结构内使用小波分析的小波 - ARIMAX方法在预测性能和模型拟合方面优于标准ARIMAX模型。基于气候变量，小波 - ARIMAX能够准确预测腹泻发生情况，平均绝对误差（MAE）、均方根误差（RMSE）和均方根对数误差（RMSLE）表明了这一点。结果凸显了采用小波 - ARIMAX等先进时间序列建模工具以更好理解和预测气候与健康相互作用的必要性。小波 - ARIMAX利用小波分析识别气候 - 疾病相互作用中的时变模式，为公共卫生举措提供了有用见解。
本研究结果对气候适应性健康规划具有启示意义，强调需要采取针对性行动以减少气候变化对城镇腹泻疾病负担的影响。
