A Bayesian model for predicting monthly fire frequency in Kenya.

This study presents a comprehensive analysis of historical fire and climatic data to estimate the monthly frequency of vegetation fires in Kenya. This work introduces a statistical model that captures the behavior of fire count data, incorporating temporal explanatory factors and emphasizing the predictive significance of maximum temperature and rainfall. By employing Bayesian approaches, the paper integrates literature information, simulation studies, and real-world data to enhance model performance and generate more precise prediction intervals that encompass actual fire counts. To forecast monthly fire occurrences aggregated from the Moderate Resolution Imaging Spectroradiometer (MODIS) data in Kenya (2000-2018), the study utilizes maximum temperature and rainfall values derived from global GeoTiff (.tif) files sourced from the WorldClim database. The evaluation of the widely used Negative Binomial (NB) model and the proposed Bayesian Negative Binomial (BNB) model reveals the superiority of the latter in accounting for seasonal patterns and long-term trends. The simulation results demonstrate that the BNB model outperforms the NB model in terms of Root Mean Square Error (RMSE), and Mean Absolute Scaled Error (MASE) on both training and testing datasets. Furthermore, when applied to real data, the Bayesian Negative Binomial model exhibits better performance on the test dataset, showcasing lower RMSE (163.22 vs. 166.67), lower MASE (1.12 vs. 1.15), and reduced bias (-2.52% vs. -2.62%) compared to the NB model. The Bayesian model also offers prediction intervals that closely align with actual predictions, indicating its flexibility in forecasting the frequency of monthly fires. These findings underscore the importance of leveraging past data to forecast the future behavior of the fire regime, thus providing valuable insights for fire control strategies in Kenya. By integrating climatic factors and employing Bayesian modeling techniques, the study contributes to the understanding and prediction of vegetation fires, ultimately supporting proactive measures in mitigating their impact.

Factors associated with viral load non-suppression among treatment-experienced pre-teenage children living with HIV in Kenya: a nationwide population-based cohort study, 2015-2021.

Background: Viral load non-suppression (VLNS) in children is a major public health concern because of attendant HIV disease progression and risk of morbidity and mortality. Based on a nationally representative database we present estimates of the prevalence, trends and factors associated with VLNS in Kenyan pre-teenage children between 2015 and 2021. Methods: Kenya National AIDS & STI Control Program's (NASCOP) maintains an early infant diagnosis and viral load (EID/VL) database for all persons living with HIV who are enrolled in the country's primary care clinics for purposes of monitoring progress towards achievement of the 95% viral suppression goals. Participants were eligible if they were children living with HIV (CLHIV), on combination ART (cART) treatment, and ≤12 years old. The modified Mann-Kendall trend test for serially correlated data was used to identify VLNS trends. Generalized estimating equations (GEE) with a logit link was used to assess the effects of covariates on the odds of VLNS (VL ≥1,000 copies/mL) over repeated points in time, allowing for the correlation among the repeated measures. Findings: Between January 2015 and December 2021, 508,743 viral load tests were performed on samples collected from 109,682 pre-teenage children. The prevalence of VLNS decreased from 22.9% (95% CI 22.4-23.3) to 12.5% (95% CI 12.1-12.9), p < 0.0001, and mean age increased from 3.1 (4.2) to 8.0 (3.2) years in 2015 and 2021 respectively. A modified Mann-Kendall trend test for serially correlated data denotes a statistically significant decreasing trend (τ = -0.300, p < 0.0001) over the study period. In the multivariable GEE analysis adjusted for covariates, the odds of VLNS decreased by 11% per year during the study period, (GEE-aOR 0.89, 95% CI 0.88-0.90; p < 0.0001). Factors positively associated with VLNS were EFV/NVP-based first-line cART regimen (GEE-aOR 1.74, 95% CI 1.65-1.84, p < 0.0001), PI-based cART regimen (GEE-aOR 1.82, 95% CI 1.72-1.92, p < 0.0001), and children aged 1-3 years (toddlers) (GEE-aOR: 1.84, 95% CI 1.79-1.90, p < 0.0001). On the contrary, DTG-based cART regimen, were negatively associated with VLNS (GEE-aOR 0.70, 95% CI 0.65-0.75, p < 0.0001). Interpretation: There is a strong evidence of decreasing viremia between 2015 and 2021. To sustain the decreasing trend, accelerating the switch from the suboptimal EVP/NVP first-line regimen to optimised DTG regimen is warranted. Funding: U.S. President's Emergency Plan for AIDS Relief (PEPFAR) and Clinton Health Access Initiative (CHAI).

Widening geographic range of Rift Valley fever disease clusters associated with climate change in East Africa.

Background: Recent epidemiology of Rift Valley fever (RVF) disease in Africa suggests growing frequency and expanding geographic range of small disease clusters in regions that previously had not reported the disease. We investigated factors associated with the phenomenon by characterizing recent RVF disease events in East Africa. Methods: Data on 100 disease events (2008 - 2022) from Kenya, Uganda, and Tanzania were obtained from public databases and institutions, and modeled against possible geo-ecological risk factors of occurrence including altitude, soil type, rainfall/precipitation, temperature, normalized difference vegetation index (NDVI), livestock production system, land-use change, and long-term climatic variations. Decadal climatic variations between 1980-2022 were evaluated for association with the changing disease pattern. Results: Of 100 events, 91% were small RVF clusters with a median of one human (IQR, 1-3) and 3 livestock cases (IQR, 2-7). These clusters exhibited minimal human mortality (IQR 0-1), and occurred primarily in highlands (67%), with 35% reported in areas that had never reported RVF disease. Multivariate regression analysis of geo-ecological variables showed a positive correlation between occurrence and increasing temperature and rainfall. A 1oC increase in temperature and 1-unit increase in NDVI, 1-3 months prior were associated with increased RVF incidence rate ratios (IRR) of 1.20 (95% CI 1.1,1.2) and 9.88 (95% CI 0.85, 119.52), respectively. Long-term climatic trends showed significant decadal increase in annual mean temperature (0.12 to 0.3oC/decade, P<0.05), associated with decreasing rainfall in arid and semi-arid lowlands but increasing rainfall trends in highlands (P<0.05). These hotter and wetter highlands showed increasing frequency of RVF clusters, accounting for 76% and 43% in Uganda and Kenya, respectively. Conclusion: These findings demonstrate the changing epidemiology of RVF disease. The widening geographic range of disease is associated with climatic variations, with the likely impact of wider dispersal of virus to new areas of endemicity and future epidemics. Key questions: What is already known on this topic?: Rift Valley fever is recognized for its association with heavy rainfall, flooding, and El Niño rains in the East African region. A growing body of recent studies has highlighted a shifting landscape of the disease, marked by an expanding geographic range and an increasing number of small RVF clusters.What this study adds: This study challenges previous beliefs about RVF, revealing that it predominantly occurs in small clusters rather than large outbreaks, and its association with El Niño is not as pronounced as previously thought. Over 65% of these clusters are concentrated in the highlands of Kenya and Uganda, with 35% occurring in previously unaffected regions, accompanied by an increase in temperature and total rainfall between 1980 and 2022, along with a rise in the annual number of rainy days. Notably, the observed rainfall increases are particularly significant during the short-rains season (October-December), aligning with a secondary peak in RVF incidence. In contrast, the lowlands of East Africa, where typical RVF epidemics occur, display smaller and more varied trends in annual rainfall.How this study might affect research, practice, or policy: The worldwide consequence of the expanding RVF cluster is the broader dispersion of the virus, leading to the establishment of new regions with virus endemicity. This escalation heightens the risk of more extensive extreme-weather-associated RVF epidemics in the future. Global public health institutions must persist in developing preparedness and response strategies for such scenarios. This involves the creation and approval of human RVF vaccines and therapeutics, coupled with a rapid distribution plan through regional banks.

Widening geographic range of Rift Valley fever disease clusters associated with climate change in East Africa.

BACKGROUND: Recent epidemiology of Rift Valley fever (RVF) disease in Africa suggests growing frequency and expanding geographic range of small disease clusters in regions that previously had not reported the disease. We investigated factors associated with the phenomenon by characterising recent RVF disease events in East Africa. METHODS: Data on 100 disease events (2008-2022) from Kenya, Uganda and Tanzania were obtained from public databases and institutions, and modelled against possible geoecological risk factors of occurrence including altitude, soil type, rainfall/precipitation, temperature, normalised difference vegetation index (NDVI), livestock production system, land-use change and long-term climatic variations. Decadal climatic variations between 1980 and 2022 were evaluated for association with the changing disease pattern. RESULTS: Of 100 events, 91% were small RVF clusters with a median of one human (IQR, 1-3) and three livestock cases (IQR, 2-7). These clusters exhibited minimal human mortality (IQR, 0-1), and occurred primarily in highlands (67%), with 35% reported in areas that had never reported RVF disease. Multivariate regression analysis of geoecological variables showed a positive correlation between occurrence and increasing temperature and rainfall. A 1°C increase in temperature and a 1-unit increase in NDVI, one months prior were associated with increased RVF incidence rate ratios of 1.20 (95% CI 1.1, 1.2) and 1.93 (95% CI 1.01, 3.71), respectively. Long-term climatic trends showed a significant decadal increase in annual mean temperature (0.12-0.3°C/decade, p<0.05), associated with decreasing rainfall in arid and semi-arid lowlands but increasing rainfall trends in highlands (p<0.05). These hotter and wetter highlands showed increasing frequency of RVF clusters, accounting for 76% and 43% in Uganda and Kenya, respectively. CONCLUSION: These findings demonstrate the changing epidemiology of RVF disease. The widening geographic range of disease is associated with climatic variations, with the likely impact of wider dispersal of virus to new areas of endemicity and future epidemics.

Improving performance of hurdle models using rare-event weighted logistic regression: an application to maternal mortality data.

In this paper, performance of hurdle models in rare events data is improved by modifying their binary component. The rare-event weighted logistic regression model is adopted in place of logistic regression to deal with class imbalance due to rare events. Poisson Hurdle Rare Event Weighted Logistic Regression (REWLR) and Negative Binomial Hurdle (NBH) REWLR are developed as two-part models which use the REWLR model to estimate the probability of a positive count and a Poisson or NB zero-truncated count model to estimate non-zero counts. This research aimed to develop and assess the performance of the Poisson and Negative Binomial (NB) Hurdle Rare Event Weighted Logistic Regression (REWLR) models, applied to simulated data with various degrees of zero inflation and to Nairobi county's maternal mortality data. The study data on maternal mortality were pulled from JPHES. The data contain the number of maternal deaths, which is the outcome variable, and other obstetric and demographic factors recorded in MNCH facilities in Nairobi between October 2021 and January 2022. The models were also fit and evaluated based on simulated data with varying degrees of zero inflation. The obtained results are numerically validated and then discussed from both the mathematical and the maternal mortality perspective. Numerical simulations are also presented to give a more complete representation of the model dynamics. Results obtained suggest that NB Hurdle REWLR is the best performing model for zero inflated count data due to rare events.

Assessing predictive performance of supervised machine learning algorithms for a diamond pricing model.

This study conducted a comprehensive analysis of multiple supervised machine learning models, regressors and classifiers, to accurately predict diamond prices. Diamond pricing is a complex task due to the non-linear relationships between key features such as carat, cut, clarity, table, and depth. The analysis aimed to develop an accurate predictive model by utilizing both regression and classification approaches. To preprocess the data, the study employed various techniques. The work addressed outliers, standardized the predictors, performed median imputation of missing values, and resolved multicollinearity issues. Equal-width binning on the cut variable was performed to handle class imbalance. Correlation-based feature selection was utilized to eliminate highly correlated variables, ensuring that only relevant features were included in the models. Outliers were handled using the inter-quartile range method, and numerical features were normalized through standardization. Missing values in numerical features were imputed using the median, preserving the integrity of the dataset. Among the models evaluated, the RF regressor exhibited exceptional performance. It achieved the lowest root mean squared error (RMSE) of 523.50, indicating superior accuracy compared to the other models. The RF regressor also obtained a high R-squared ([Formula: see text]) score of 0.985, suggesting it explained a significant portion of the variance in diamond prices. Furthermore, the area under the curve with RF classifier for the test set was 1.00 [Formula: see text], indicating perfect classification performance. These results solidify the RF's position as the best-performing model in terms of accuracy and predictive power, both in regression and classification. The MLP regressor showed promising results with an RMSE of 563.74 and an [Formula: see text] score of 0.980, demonstrating its ability to capture the complex relationships in the data. Although it achieved slightly higher errors than the RF regressor, further analysis is needed to determine its suitability and potential advantages compared to the RF regressor. The XGBoost Regressor achieved an RMSE of 612.88 and an [Formula: see text] score of 0.972, indicating its effectiveness in predicting diamond prices but with slightly higher errors compared to the RF regressor. The Boosted Decision Tree Regressor had an RMSE of 711.31 and an [Formula: see text] score of 0.968, demonstrating its ability to capture some of the underlying patterns but with higher errors than the RF and XGBoost models. In contrast, the KNN regressor yielded a higher RMSE of 1346.65 and a lower [Formula: see text] score of 0.887, indicating its inferior performance in accurately predicting diamond prices compared to the other models. Similarly, the Linear Regression model performed similarly to the KNN regressor, with an RMSE of 1395.41 and an [Formula: see text] score of 0.876. The Support Vector Regression model showed the highest RMSE of 3044.49 and the lowest [Formula: see text] score of 0.421, indicating its limited effectiveness in capturing the complex relationships in the data. Overall, the study demonstrates that the RF outperforms the other models in terms of accuracy and predictive power, as evidenced by its lowest RMSE, highest [Formula: see text] score, and perfect classification performance. This highlights its suitability for accurately predicting diamond prices. The study not only provides an effective tool for the diamond industry but also emphasizes the importance of considering both regression and classification approaches in developing accurate predictive models. The findings contribute valuable insights for pricing strategies, market trends, and decision-making processes in the diamond industry and related fields.

Nevirapine plasma concentration is associated with virologic failure and the emergence of drug-resistant mutations among HIV patients in Kenya: A cross sectional study.

This study aimed to determine the association between the plasma concentration of nevirapine (NVP) and clinical outcomes. In this cross-sectional study, sociodemographic and clinical data were collected from 233 HIV patients receiving NVP-based first-line antiretroviral therapy (ART) regimens in Nairobi, Kenya. The mean age was 41.2 (SD ±â€…11.9) years. Fifty-four (23.2%) patients had virological failure (>1000 copies/mL), whereas 23 (9.9%) were infected with drug-resistant HIV strains. Eleven patients had nucleoside reverse transcriptase inhibitor resistance mutations, including M184V and T215Y, whereas 22 had non-nucleoside reverse transcriptase inhibitor resistance mutations, including G190A, K103N, V106A, Y181C, A98G, and Y188L. The median NVP plasma concentration was 6180 ng/mL (IQR 4444-8843 ng/mL), with 38 (16.3%) patients having suboptimal NVP plasma levels of <3400 ng/mL. The majority 23 of the 38 (60.5%) patients with NVP Cmin < 3400 ng/mL were significantly infected with drug-resistant HIV virus (P = .001). In the multivariate analysis, the time taken to arrive at the ART clinic (ß -11.1, 95% CI -21.2 to -1.1; P = .031), higher HIV viral load (ß -2008, 95% CI -3370.7 to -645.3; P = .004), and the presence of HIV drug resistance mutation (ß 3559, 95% CI 2580.8-4537.2; P = .0001) were associated with NVP plasma concentration. A significant proportion of patients receiving the NVP-based regimen had supra- and sub-therapeutic plasma concentrations. Higher HIV viral load and the presence of HIV drug-resistant mutations are important factors associated with NVP plasma concentrations.

Application of optimal control to the onchocerciasis transmission model with treatment.

In this paper, we present a model for onchocerciasis that considers mass administration of ivermectin, contact prevention controls and vector elimination. The model equilibria are computed and stability analysis carried out in terms of the basic reproduction number R0. The model is found to exhibit a backward bifurcation so that for R0 less than unity is not sufficient to eradicate the disease from the population and the need is to lower R0 to below a certain threshold, R0c for effective disease control. The model is fitted to data on individuals with onchocerciasis in Ghana. A sensitivity analysis reveals that the parameters with the most control over the epidemic are the vector death rate and the effective contact rates between susceptible individuals and infected vector and susceptible vector with infected individuals. This suggests that programs aimed controlling vector will be significantly more effective in combating the disease. Optimal control theory is applied to investigate optimal control strategies for controlling onchocerciasis using insect repellent and both insecticide and larvicide as system control variables. We use Pontryagin's Maximum Principle to show the necessary conditions for the optimal control of onchocerciasis. Numerical simulations of the model show that restricted and proper use of control measures might considerably decrease the number of infections in the human population.