The role of national nutrition programs on stunting reduction in Rwanda using machine learning classifiers: a retrospective study.

BACKGROUND: In Rwanda, the prevalence of childhood stunting has slightly decreased over the past five years, from 38% in 2015 to about 33% in 2020. It is evident whether Rwanda's multi-sectorial approach to reducing child stunting is consistent with the available scientific knowledge. The study was to examine the benefits of national nutrition programs on stunting reduction under two years in Rwanda using machine learning classifiers. METHODS: Data from the Rwanda DHS 2015-2020, MEIS and LODA household survey were used. By evaluating the best method for predicting the stunting reduction status of children under two years old, the five machine learning algorithms were modelled: Support Vector Machine, Logistic Regression, K-Near Neighbor, Random Forest, and Decision Tree. The study estimated the hazard ratio for the Cox Proportional Hazard Model and drew the Kaplan-Meier curve to compare the survivor risk of being stunted between program beneficiaries and non-beneficiaries. Logistic regression was used to identify the nutrition programs related to stunting reduction. Precision, recall, F1 score, accuracy, and Area under the Curve (AUC) are the metrics that were used to evaluate each classifier's performance to find the best one. RESULTS: Based on the provided data, the study revealed that the early childhood development (ECD) program (p-value = 0.041), nutrition sensitive direct support (NSDS) program (p-value = 0.03), ubudehe category (p-value = 0.000), toilet facility (p-value = 0.000), antenatal care (ANC) 4 visits (p-value = 0.002), fortified blended food (FBF) program (p-value = 0.038) and vaccination (p-value = 0.04) were found to be significant predictors of stunting reduction among under two children in Rwanda. Additionally, beneficiaries of early childhood development (p < .0001), nutrition sensitive direct support (p = 0.0055), antenatal care (p = 0.0343), Fortified Blended Food (p = 0.0136) and vaccination (p = 0.0355) had a lower risk of stunting than non-beneficiaries. Finally, Random Forest performed better than other classifiers, with precision scores of 83.7%, recall scores of 90.7%, F1 scores of 87.1%, accuracy scores of 83.9%, and AUC scores of 82.4%. CONCLUSION: The early childhood development (ECD) program, receiving the nutrition sensitive direct support (NSDS) program, focusing on households with the lowest wealth quintile (ubudehe category), sanitation facilities, visiting health care providers four times, receiving fortified blended food (FBF), and receiving all necessary vaccines are what determine the stunting reduction under two among the 17 districts of Rwanda. Finally, when compared to other models, Random Forest was shown to be the best machine learning (ML) classifier. Random forest is the best classifier for predicting the stunting reduction status of children under two years old.

Risky sexual behaviors and their associated factors within high school students from Collège Saint André in Kigali, Rwanda: An institution-based cross-sectional study.

Background: Risky sexual behaviors (RSBs) remain public health concerns in adolescents from sub-Saharan Africa (SSA), and these practices may increase vulnerability to reproductive health problems if no early healthcare strategies are implemented. While previous studies reported that adolescents are engaged in these RSBs due to diverse influences such as the teenage stage, urbanization, and change in the environment they experience, there is a shortage of studies on RSB among adolescents in SSA. This study assessed the magnitude of RSBs and the RSB-associated factors among in-school adolescents. Methods: School-based cross-sectional study was conducted among 263 Saint Andre school students in Kigali, Rwanda, from July 3, 2020, to September 30, 2020. Systematic random sampling techniques were employed. All data were entered into Epi-Data and analyzed using SPSS version 25. Chi-square tests and multivariable logistic regression analyses were applied to determine factors associated with risky sexual behaviors. Confidence intervals (CIs) of 95% and 5% for statistical significance were maintained. Results: Of 263 participants, 109 (41%) experienced RSB in their lifetime. Among them, 66 respondents (60.55%) utilized contraceptive methods to prevent sexual and reproductive problems that can be caused by unprotected sexual intercourse. The students who experienced domestic violence had increased odds of experiencing RSB [odds ratio (OR) = 4.22; 95% CI: 1.6-11.23] than their counterparts. Those in grade 11 (OR = 2.68; 95% CI: 1.06-6.78) and grade 12 (OR = 4.39; 95% CI: 1.82-10.56) were more likely to practice RSB than those in grade 10. Alcohol users were almost more likely to experience RSB (OR = 3.9; 95% CI: 1.97-5.5) than their counterparts. Those who lived away from their biological parents had higher likelihood of experiencing RSB (OR = 2.5; 95% CI: 1.14-4.42) than those who lived with one or both parents. Students who experienced peer pressure were more likely to engage in RSB (OR = 3.9; 95% CI: 2.01-7.51) than their counterparts. Conclusion: Promoting specific intervention programs built upon the factors associated with RSB among high school students needs to be prioritized.

The use of Vsee videoconferencing for live telepathology in Rwanda, a potential solution for resource-limited area.

Objective: The shortage of pathologists is a worldwide problem that is more severe in Africa. One of the solutions is the use of telepathology (TP); however, most of the TP systems are expensive and unaffordable in many developing countries. At the University Teaching Hospital of Kigali, Rwanda, we assessed the possibility of combining commonly available laboratory tools into a system that can be used for diagnostic TP using Vsee videoconferencing. Methodology: Using an Olympus microscope (with a camera) operated by a laboratory technologist, histologic images were transmitted to a computer whose screen was shared, using Vsee, with a remotely located pathologist who made the diagnoses. Sixty consecutive small biopsies (≤6 glass slides) from different tissues were examined to make a diagnosis using live Vsee-based videoconferencing TP. Vsee-based diagnoses were compared to pre-existing light microscopy-based diagnoses. Percent agreement and unweighted Cohen's kappa coefficient of the agreement were calculated. Results: For agreement between conventional microscopy-based and Vsee-based diagnoses, we found an unweighted Cohen's kappa of 0.77 ± 0.07SE with a 95% CI of 0.62-0.91. The perfect percent agreement was 76.6% (46 of 60). Agreement with minor discrepancy was 15% (9 of 60). There were 2 cases of major discrepancy (3.30%). We were unable to make a diagnosis in 3 cases (5%) because of poor image quality related to the instantaneous internet connectivity problems. Conclusion: This system provided promising results. However, additional studies to assess other parameters which can affect its performance are needed before this system can be considered an alternative method of providing TP services in resource-limited settings.

"A comparative analysis of risk factors of malaria" case study Gisagara and Bugesera District of Rwanda. RDHS 2014/2015. A retrospective study.

BACKGROUND: Malaria is a public health concern worldwide. A figure of 3.2 billion people is at risk of malaria a report of World Health Organization in 2013. A proportion of 89 and 91 cases of malaria reported during 2015 were respectively attributed to malaria cases and malaria deaths in Sub-Saharan Africa. Rwanda is among the Sub-Saharan Africa located in East Africa. The several reports indicate that from 2001 to 2011, malaria cases increased considerably especially in Eastern and Southern Province with five million cases. The affected districts included Bugesera in the Eastern and Gisagara in the Southern Province of Rwanda with a share of 41% of the country prevalence in 2014 and during 2017-2018 a figure of 11 deaths was attributed to malaria and both Gisagara and Bugesera Districts were the high burdened. METHODOLOGY: The RDHS 2014-2015 data was used for the study and a cross-sectional survey was used in which two clusters were considered both Gisagara and Bugesera Districts in the Southern and Eastern Province of Rwanda. Bivariate analysis was used to determine the significant predictors with malaria and reduced logistic regression model was used. RESULTS: The results of the study show that not having mosquito bed nets for sleeping is 0.264 times less likely of having malaria than those who have mosquito bed nets in Gisagara District. In Bugesera District, living in low altitude is 2.768 times more likely associated with the risk of getting malaria than living in high altitude. CONCLUSION: The results of the study concluded that environmental and geographical factor such as low altitude is the risk factor associated with malaria than the high altitude in Bugesera District. While not having mosquito bed nets for sleeping is the protective factor for malaria than those who have it in Gisagara District. On the other hand, socio-economic and demographic characteristics do not have any effect with malaria on the results of the study.

Knowledge, Barriers and Motivators to Cervical Cancer Screening in Rwanda: A Qualitative Study.

Background: Cervical cancer is a global public health problem with marked geographical disparity. High morbidity and mortality rates in developing countries are associated with low screening rates. In 2020, in Rwanda, 3.7 million women aged 15-59 years were at risk of developing cervical cancer, the most commonly diagnosed female cancer in Rwanda. Despite Rwanda being the first African country to vaccinate against human papilloma virus with a three-dose regimen vaccination coverage of nearly 93% in the target population of girls aged <15 years, and having established cervical cancer screening program, recent studies have found low screening rates. Our study sought to determine knowledge, motivators and barriers of cervical cancer screening. Methods: We conducted a qualitative descriptive study; using focus group interview in an urban health facility (Muhima district hospital) and a rural health center (Nyagasambu health center) offering cervical screening services in Rwanda. Participants were women seeking these services and other women attending the health facility for any reason, and female staff working in these facilities. Interviews were recorded and transcribed, and data were analyzed using content analysis. Results: Thirty women participated in focus group interview, with an average age of 39 years. Many of women showed knowledge about cervical cancer existence and prevention methods. However, fear for pain, lack of knowledge about screening, how and where the screening was done, and concern for privacy were recurring subthemes. Some participants also mentioned lack of health insurance as a barrier for cervical cancer screening. Conclusion: Barriers to uptake cervical cancer screening services in Rwanda are related to poor information about cervical cancer and the importance of screening as well as non-adherence to medical insurance. Population sensitization through campaign and community outreach activities could have a positive impact on increasing the usage of cervical cancer screening in Rwanda.