Comparing Malaria Trends in the Comoros Islands: ARIMA Modeling and Retrospective Analysis
Dennis Baidoo, Emmanuel Kubuafor, Okeke Onyedikachi Joshua, Samuel Frimpong Osarfo, Frank Amo Agyei-Owusu, Justice Akuoko-Frimpong, Nicholas Appiah, Agnes Duah, Robert Amevor, Francis Effah Boateng, Fuseini Aboagye, Bright Benyi · 2025
Read Comparing Malaria Trends in the Comoros Islands: ARIMA Modeling and Retrospective Analysis by Dennis Baidoo, Emmanuel Kubuafor, Okeke Onyedikachi Joshua, Samuel Frimpong Osarfo, Frank Amo Agyei-Owusu, Justice Akuoko-Frimpong, Nicholas Appiah, Agnes Duah, Robert Amevor, Francis Effah Boateng, Fuseini Aboagye, Bright Benyi (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-16018.txt.
Introduction
Dennis Baidoo and co-authors chart, in a 2025 study, malaria trends in the Comoros Islands and forecast deaths with a time model. Malaria stays a grave health load there in spite of years of control work. Past studies showed case drops from nets, sprays, and mass drug rounds. Few, though, forecast deaths or weighed the long run of those acts. The team read death data from 1990 to 2019 and fit an ARIMA model, a tool that projects a trend from its own past. Checks proved the model sound for that use.
Case rates fell in a steep way in Grand Comoro, from 235 to about 5 per 1,000 people. The model says deaths stay low if control holds at its current pace. Drug and spray proof germs and thin funds are the main risks to that path.
Catalog record
| CREATOR | Dennis Baidoo, Emmanuel Kubuafor, Okeke Onyedikachi Joshua, Samuel Frimpong Osarfo, Frank Amo Agyei-Owusu, Justice Akuoko-Frimpong, Nicholas Appiah, Agnes Duah, Robert Amevor, Francis Effah Boateng, Fuseini Aboagye, Bright Benyi |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 18 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-16018
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Comparing Malaria Trends in the Comoros Islands: ARIMA Modeling and Retrospective Analysis Dennis Baidoo, Emmanuel Kubuafor, Okeke Onyedikachi Joshua, Samuel Frimpong Osarfo, Frank Amo Agyei-Owusu, Justice Akuoko-Frimpong, Nicholas Appiah, Agnes Duah, Robert Amevor, Francis Effah Boateng, Fuseini Aboagye, Bright Benyi Abstract Abstract Context : Malaria remains a serious health challenge in the Comoros Islands, despite ongoing control efforts. Past studies have shown reductions in cases due to prevention and treatment measures, but little work has been done to forecast future malaria deaths and assess the long-term impact of these measures. Methods : Malaria mortality data from 1990 to 2019 were analyzed using an ARIMA(1,0,0) model. The model was validated through diagnostic tests, ensuring reliability for forecasting trends. Results : The study confirmed significant reductions in malaria cases, such as in Grand Comoro, where cases fell from 235.36 to 5.47 per 1,000 people. The ARIMA model predicted that fatalities will remain low if current control measures, including bed nets, indoor spraying, and mass drug administration, are sustained. Interpretation : The findings highlight the success of these interventions in reducing malaria mortality. However, challenges like drug and insecticide resistance and financial limitations pose risks to further progress. Continued support and adaptation of strategies are essential to address these challenges and sustain low malaria mortality rates. Conclusion : The study demonstrates the effectiveness of malaria control efforts in the Comoros and underscores the importance of maintaining these measures to achieve malaria elimination and improve public health outcomes. Keywords : Malaria fatalities, ARIMA modeling, Retrospective analysis, Time series analysis, Public health, Forecasting. 1 Introduction Malaria continues to pose a significant public health issue in most sub-Saharan African countries, including Comoros, located in the Indian Ocean, with tremendously varying trajectories across the archipelago, despite the United Nations’ Sustainable Development Goal 3 in ensuring healthy lives and promoting well-being for all walks of life [ 14 , 15 , 16 ] . However, current studies have shown a substantial reduction in malaria morbidity and mortality between 2010 and 2014, associated with an effective combination of preventive and control treatment measures [ 17 ] . Efforts to reduce malaria fatalities have included various interventions such as chloroquine (CQ), artemisinin-based combination therapies (ACT), insecticide-treated bed nets (ITNs), mass drug administration (MDA) as part of fast elimination of malaria by source eradication (FEMSE), and indoor residual spraying (IRS). These approaches have significantly contributed to lowering mortality rates associated with malaria [ 17 , 18 ] . About 82% of heads of households in the villages in Comoros knew that malaria is a transmissible disease, which made the MDA a success [ 19 ] . Although significant progress has been made, challenges persist, including ACT-resistant malaria strains, mosquito pesticide resistance, and the need for continued financial support to achieve malaria elimination in the Comoros [ 20 ] . Recent studies on the Plasmodium falciparum populations on Grande Comoros Island, one species of parasite that causes malaria in humans and is spread by the bites of female Anopheles mosquitoes , have revealed a significant decrease in parasitic genetic diversity and various infections from 2006-2007 to 2013-2016, suggesting a tremendous decrease in malaria transmission intensity [ 21 , 22 ] . As a result, it is crucial to consider the overall mortality cases from 1990 to 2019 to assess the long-term effectiveness of the interventions implemented in eradicating malaria in Comoros. Out of all the reviewed literature, there has not been a single paper forecasting the future burdens of malaria fatalities on the effectiveness of these treatment and preventive measures in this critical region. By analyzing the post-implementation of tools like IRS, ITNs, and the FEMSE strategy, this study will provide critical insights into the long-term impact of these interventions, aiding future efforts toward malaria elimination in the region. Building on previous research, this study evaluates the overall effectiveness of malaria control interventions in the Comoros by forecasting national-level malaria mortality trends, rather than examining trends for individual islands. 2 Materials and Methods 2.1 Data Collection and Source The data used for the study are the global annual causes of death numbers for different countries [10]. Fatalities of malaria in Comoros were extracted from 1990 to 2019. The dataset was obtained from a reliable global source, ensuring accuracy and consistency in malaria fatality records for the Comoros Islands. The data analysis was conducted using RStudio version 2024.4.2.764 (Posit Software, PBC, Boston, MA). The Autoregressive Integrated Moving Average (ARIMA) time series model was employed to evaluate and predict trends in malaria fatalities across all islands. 2.2 Statistical Analysis and Model Selection A special case of the ARIMA model is the autoregressive with order p, AR(p), and it is formulated below as; x_{t}=\phi_{1}x_{t-1}+\phi_{2}x_{t-2}+\cdots+\phi_{p}x_{t-p}+w_{t}, (1) where x_{t} is stationary (the current malaria fatalities at time t ), x_{t-i} ’s are the previous malaria fatalities for i=1,2,\ldots,p lags, w_{t}\sim wn(0,\sigma_{w}^{2}) is a white noise and \phi_{1},\phi_{2},\ldots,\phi_{p} are constants (lags) with \phi_{p}\neq 0 . If the mean, \mu , of x_{t} is non-zero, we replace x_{t} by x_{t}-\mu in (1) and get x_{t}-\mu=\phi_{1}(x_{t-1}-\mu)+\phi_{2}(x_{t-2}-\mu)+\cdots+\phi_{p}(x_{t-p}-\mu)+w_{t} (2) x_{t}=\alpha+\phi_{1}x_{t-1}+\phi_{2}x_{t-2}+\cdots+\phi_{p}x_{t-p}+w_{t} (3) where \alpha=\mu(1-\phi_{1}-\cdots-\phi_{p}). The ARIMA(p,0,0) model is equivalent to an autoregressive with order p, AR(p) [ 23 ] . To assess model performance, we conducted several diagnostic checks. Residuals were tested for normality using the Shapiro-Wilk test, and heteroscedasticity was evaluated by plotting residuals against fitted values. The Ljung-Box test was used to check for autocorrelation, while the corrected Akaike Information Criterion (AICc) guided model selection. These diagnostics helped validate assumptions and ensure model robustness. ARIMA was chosen for its robustness in modeling and forecasting time series data with strong temporal dependencies. Since the Comoros malaria mortality data do not exhibit clear seasonal patterns, using a seasonality-based model is unnecessary. While machine learning models could capture complex nonlinear patterns, ARIMA provides a more interpretable and reliable approach for short-to medium-term forecasting. Its effectiveness in assessing long-term intervention impacts further supports its suitability for this study. 3 Results of Analysis 3.1 Access to Malaria Prevention Measures and Malaria Case Distribution In 2019, access to malaria prevention measures, specifically LLINs, IRS, ACT, and FEMSE, varied across the studied African countries (Fig. 1, left). Botswana and Eswatini had the highest coverage, with a large proportion of their populations protected, whereas Eritrea, Comoros, Namibia, and South Africa had relatively lower coverage levels. The distribution of estimated malaria cases also varied significantly among these countries (Fig. 1, right). Eritrea and Comoros, with lower coverage rates, had a higher proportion of cases within their high-risk populations, while Botswana and South Africa showed lower malaria case shares, potentially reflecting the impact of higher prevention coverage. These findings underscore the relationship between malaria prevention access and case distribution across different at-risk groups. Figure 1 : Percentage of population with access to prevention measures, 2019 (left) and share of estimated malaria cases, 2019 (right) (WHO, 2020) Malaria cases by country in the WHO African Region for 2019 (Fig. 2, left), and reported indigenous malaria cases in countries with national elimination activities, comparing 2015 to 2019 (Fig. 2, right) (WHO, 2020). The bar chart on the left shows that malaria cases are highly concentrated in certain countries, with the highest burden observed in Nigeria, with Comoros among the lowest number of malaria