Temporal dynamics and predictive modelling of mpox outbreak using autoregressive integrated moving average: insights from the Nigeria Centre for Disease Control and Prevention, 2017–2025 data

Abstract Objectives We characterise the temporal dynamics and patterns of mpox in Nigeria from 2017 to August 2025, and predict future mpox case numbers and inform public health interventions Methods We conducted a retrospective, observational time-series analysis using weekly mpox surveillance data from the Nigeria Centre for Disease Control. We used an Autoregressive Integrated Moving Average (ARIMA) to forecast the number of confirmed and death cases. We fitted ARIMAX models with rainfall covariates as exogenous predictors to probe potential climatic drivers of mpox transmission. Results We identified three epidemic phases: low endemicity (2017–2021), a significant surge in 2022 (suspected 2,123; confirmed 762), and post-epidemic persistence (2023–2025). We found that mpox cases consistently increased during the rainy months and decreased in the dry season, with peaks most often observed between May and September. Our forecasts projected a continued low to moderate incidence and near-zero mortality. Rainfall was not a dominant driver of mpox incidence; the conditional mean of predicted cases remained relatively stable and close to recent observed levels. Conclusion Mpox has transitioned from sporadic outbreaks to an endemic pattern in Nigeria, with near-term incidence projected to remain low to moderate. By spanning pre-epidemic, epidemic, and post-epidemic phases, this analysis demonstrates how time-series forecasting can be operationalised into a practical, tiered early-warning framework to guide routine public health surveillance and proactive outbreak response.