Significance of the ARIMA epidemiological modeling to predict the rate of HIV and AIDS in the Kumba Health District of Cameroon

Emmanuel Besong Arrey ¹ Odette Dzemo Kibu ^1,2,3 Elvis Asangbeng Tanue ^1,2,3 Agbor Obinkem ¹ Ginyu Innocentia Kwalar ^1,2,3 Fabrice Chethkwo ¹ Valentine Ndze Ngum ¹ Maurice Marcel Sandeu ^2,3,4 Patrick Jolly Ngono ^2,3,4 Nkweteyim Denis ^2,3,5 Onduo Moise ^2,3,4 Ayana Gelan ^2,3,6 Jude Kong ^2,3 Dickson Shey Nsagha ^1*

• ¹ Faculty of Health Sciences, Department of Public Health and Hygiene, University of Buea, Buea, Cameroon
• ² Global South Artificial Intelligence for Pandemic and Epidemic Preparedness and Response Network, Toronto, Ontario, Canada
• ³ Africa-Canada Artificial Intelligence and Data Innovation Consortium, Department of Mathematics and Statistics, Faculty of Science, York University, Canada, Ontario, Canada
• ⁴ School of Veterinary Medicine and Science, University of Ngaoundéré, Ngaoundéré, Adamawa, Cameroon
• ⁵ Faculty of Science, Department of Computer Sciences, University of Buea, Buea, Cameroon
• ⁶ School of Biomedical Engineering, Jimma University, Jimma, Oromia, Ethiopia

Background: AIDS is a severe medical condition caused by the human immunodeficiency virus (HIV) that primarily attacks the immune system, specifically CD4+ T lymphocytes (a type of white blood cell crucial for immune response), monocyte macrophages, and dendritic cells. This disease has significant health and socio-economic implications and is one of the primary causes of illness and death globally (UNAIDS, 2022). It presents significant challenges for public health and population well-being, both in developed and developing countries. By conducting a time series analysis, this research seeks to identify any significant changes in HIV rates over the next 4 years in the Kumba District Hospital and provide valuable insights to inform evidence-based decision-making and strategies for preventing and controlling HIV within the Kumba Health District. A hospital-based retrospective study on HIV/AIDS recorded cases was conducted at the Kumba District Hospital. Using data extraction form, hospital records from 2012 to 2022 were reviewed and data extracted and used to make predictions on the number of future incidence cases. Time series analysis using the Auto-Regressive Integrated Moving Average (ARIMA) model was done using Statistical Package for the Social Sciences (SPSS) Version 26Results: According to the ARIMA parameter (p,d,q), the results for the Partial Autocorrelation Factor (p) was 1, differencing (d) was 0 and Autocorrelation Factor (q) was 0. Putting these values together, we had the ARIMA model (1,0,0) which predicted an overall increase in HIV incidence cases at the Kumba District Hospital for the upcoming Years (2023 -2026). Interpretation: The ARIMA model was found to be independent of errors and a perfect fit, with a high R-squared value of 0.764 and a p-value of 0.410, indicating that the model's predictions aligned well with the observed data. The model predicted an increase in the number of HIV incidence cases over the coming years (2023 -2026), potentially suggesting a worsening situation. However, it's important to interpret these predictions with caution and consider other factors that may influence the incidence of HIV in reality.