Patrick Musonda

Background: Hypertension is a leading contributor to cardiovascular morbidity in sub-Saharan Africa. BMI, waist circumference (WC), waist-to-height ratio (WHtR), and waist-to-hip ratio (WHpR) are widely used for screening, yet their optimal thresholds and predictive utility vary by population. This study evaluated these indices for hypertension prediction in Zambian adults using traditional and machine learning approaches. Methods: This cross-sectional study assessed hypertension in 4,302 adults aged 18–69 years from the 2017 Zambia WHO STEPS survey. ROC analysis with Youden's Index identified optimal sex-stratified cut-offs for BMI, WC, WHtR, and WHpR. Five machine learning classifiers — logistic regression, linear SVM, random forest, gradient boosting, and XGBoost — were then trained on anthropometric, sociodemographic, and behavioural covariates using stratified 5-fold cross-validation, with recall prioritised as the primary metric given the imbalanced outcome. Results: Hypertension prevalence was 23.7% (n=1,021), higher in men (25.4% vs. 22.7%; p=0.046) and urban than rural residents (27.0% vs. 21.7%; p<0.001). WC and WHtR showed the highest discrimination (men: AUC 0.63/0.62; women: 0.66/0.66); locally derived cut-offs were below WHO thresholds (WC: 80.35/81.85 cm; WHtR: 0.48/0.53; BMI: 21.97/24.12 kg/m²). Logistic regression and linear SVM achieved the best ROC-AUC (0.74; recall 0.69–0.70; F1=0.51); ensemble models showed higher accuracy (0.78–0.79) but lower recall (0.22–0.26). Conclusion: Central adiposity measures (WC and WHtR) outperformed BMI for hypertension identification in Zambian adults, with locally derived cut-offs consistently below WHO thresholds. Logistic regression and linear SVM demonstrated the best predictive performance, supporting their utility in risk stratification using routinely collected data.

Background: A lot of unborn and newly born babies continue to die during pregnancy, childbirth and in the first week of life. Over 5,1 million stillbirths and neonatal deaths occur globally each year with 98% of these in low-middle-income countries. Since the Millennium Development Goals (MDG) era, global perinatal mortality reduced from 5.7 million in 2000 to 4.1 million by 2015. The global stillbirth rate which averaged 24.7 in 2000 declined to 18.4 per 1000 live births by 2015. Neonatal mortality rate equally showed a downward trend from 37 in 1990 to 19 deaths per 1000 live births by 2016. In spite of this, sub-Saharan Africa (SSA) has continued to record the highest rate as 2% of the 45% data collected is in high income countries. The average perinatal mortality rate is at 30.3 per 1000 live births with Ethiopia reportedly highest at 49. The main aim of this study is to determine the magnitude of perinatal mortality in SSA since 2000. Methods: The databases searched are EMBASE, Scopus, and CINAHIL the followed by grey literature sources including ProQuest Global Thesis, and Dissertation, and Google Scholar. The articles for the publications made from January 2000 to December, 2023. Quality of the articles check was with Joanna Briggs Institute quality assessment tool. Three reviewers screened all retrieved articles, extracted, and then critically appraised all identified studies. Stata software version 17 will be used. Discussion: The review will provide a detailed summary of the magnitude of perinatal mortality in SSA since 2000. The review will synthesis available observational studies on perinatal mortality to identify pooled estimates. Systematic review registration: PROSPERO CRD42023437432

Emerging but limited evidence suggests that HIV infection does not affect in-hospital COVID-19 mortality, regardless of the prevalence of HIV infection in most parts of sub-Saharan Africa, especially the southern Africa region, and Zambia, Lusaka District in particular, is not an exception. Therefore, this study aimed to determine the effect of HIV infection, demographics, and clinical factors on mortality among hospitalized COVID-19 patients at Levy Mwanawasa University Teaching Hospital (LMUTH). A cross-sectional study was conducted with a sample size of 698 adults admitted for COVID-19 at LMUTH from 18 March 2020 to 31 December 2021. For all statistical analysis of data, STATA statistical software, version 15 MP (College Station, TX 77845, USA) was used—ensuring that appropriate statistical techniques were applied to the data. Unadjusted and adjusted logistic regressions were conducted to model COVID-19 mortality among COVID-19 patients based on their HIV status while controlling for five predictor variables. Based on the results, the best predictors of in-hospital COVID-19 mortality were HIV status, number of comorbidities, age in years, smoking, and alcohol intake. The results suggest that COVID-19 mortality among those with HIV and those without HIV infection was different. People living with HIV infection had increased odds of COVID-19 mortality compared to those without HIV. The results further suggested that a unit increase in age was associated with increased odds of COVID-19 mortality. Furthermore, drinking alcohol and having two or more comorbidities increased the odds of COVID-19 mortality compared to not drinking alcohol, having no comorbidity, or having a comorbidity. This study, therefore, concludes that HIV infection has a significant effect on COVID-19 mortality among patients hospitalized at LMUTH and that the proportion of COVID-19 mortality in the HIV-infected group is relatively higher than in the uninfected group. Therefore, there is a need for close monitoring of COVID-19 patients with HIV infection.

Background: In Zambia, 3.8% of young women and men aged 15–24 are HIV positive. However, like in most developing nations, HIV prevalence is higher among young women than young men (5.6% versus 1.8%). Despite the recognition of the rights of young people to sexual reproductive health (SRH) information and services, adolescent and young people (AYP) still face challenges in accessing healthcare in public health institutions including access to comprehensive knowledge on HIV/AIDs, HIV testing and contraceptives. The overall objective of the study was to collect baseline HIV, SRH and gender based violence (GBV) data at district level to inform the design of interventions targeting adolescent girls and young women (AGYW) aged 10–24 years in 20 districts of Zambia.Methods: A cross-sectional, mixed-methods study was conducted in 20 districts of Zambia with the highest incidence of HIV. Data was collected between August and October 2022 with a total response rate of 92% (12,813/13960), constituting 5979 (46.7%) in-school and 6834 (53.3%) out-of-school participants.Results: Overall, Mwinilunga, Chinsali, Chisamba and Chembe districts had the highest number of respondents, while Sinazongwe and Mungwi districts contributed the least. The overall age distribution was such that 12.6% (n = 1617) of those interviewed were aged 10 to 14 years, 35.4% (n = 4536) were aged 15–19 years, and 52.0% (n = 6660) were aged 20–24 years. The overall mean age at first sex among AGYW interviewed was 16.6 years which was broken down as follows: 16.2 years for in-school and 16.8 years for out of school. Overall, most of the respondents had first time sex with either their boyfriend (80.4%) or husband (15.6%), with 2.4% of the in-school participants reporting to have had their sexual debut in marriage compared to 21.0% among out-of-school AGYW. Prevalence of HIV was higher in the out-of-school compared to the in-school participants (5.5% vs 2.0%), Similarly, the prevalence of syphilis was higher in the out-of-school than the in-school participants (4.1% vs 1.5%).Conclusion: The study focused on assessing the prevalence and vulnerability of HIV, syphilis, GBV, and SRH services uptake among adolescent girls and young women, and exploring factors affecting girls’ stay-in-school and re-engagement. The study found that HIV and syphilis are still significant public health problems among adolescent girls and young women in Zambia, emphasizing the need for increased efforts to prevent and manage these infections.

Background: The burden of Malaria in Zambia remains a challenge, with the entire population at risk of contracting this infectious disease. Despite concerted efforts by African countries, including Zambia, to implement malaria policies and strategies aimed at reducing case incidence, the region faces significant hurdles, especially with emerging pandemics such as COVID-19. The efforts to control malaria were impacted by the constraints imposed to curb its transmission during the COVID-19 pandemic. The aim of the study was to assess the effect of the COVID-19 pandemic on malaria cases in Zambia and the factors associated by comparing the COVID-19 period and the pre-COVID-19 era. Methods: This was a cross-sectional panel study in which routinely collected programmatic data on malaria was used. The data were extracted from the Health Management Information System (HMIS) for the period January 2018 to January 2022. The period 2018 to 2022 was selected purely due to the availability of data and to avoid the problem of extrapolating too far away from the period of interest of the study. A summary of descriptive statistics was performed in which the number of cases were stratified by province, age group, and malaria cases. The association of these variables with the COVID-19 era was checked using the Wilcoxon rank-sum test and Kruskal‒Wallis test as applicable. In establishing the factors associated with the number of malaria cases, a mixed-effect multilevel model using the Poisson random intercept and random slope of the COVID-19 panel. The model was employed to deal with the possible correlation of the number of cases in the non-COVID-19 panel and the expected correlation of the number of cases in the COVID-19 panel. Results: A total of 18,216 records were extracted from HMIS from January 2018 to January 2022. Stratifying this by the COVID-19 period/era, it was established that 8,852 malaria cases were recorded in the non-COVID-19 period, whereas 9,364 cases were recorded in the COVID-19 era. Most of the people with malaria were above the age of 15 years. Furthermore, the study found a significant increase in the relative incidence of the COVID-19 panel period compared to the non-COVID-19 panel period of 1.32, 95% CI (1.18, 1.48, p < 0.0001). The observed numbers, as well as the incident rate ratio, align with the hypothesis of this study, indicating an elevated incidence rate ratio of malaria during the COVID-19 period. Conclusion: This study found that there was an increase in confirmed malaria cases during the COVID-19 period compared to the non-COVID-19 period. The study also found Age, Province, and COVID-19 period to be significantly associated with malaria cases.

Cases of teenage pregnancy remain high in Eastern Province of Zambia—contributing to health, economic, and social problems. This study sought to determine the effect of exposure to boarding schooling on pregnancy among school-going teenagers, taking into account individual and school characteristics. This was a retrospective case-control study involving 261 cases and 522 controls obtained from nine boarding and nine day secondary schools, between 2019 and 2021. STATA 16.1 MP was used for all statistical analyses at a 5% level of significance. Unadjusted and adjusted effects were obtained using logistic regression analysis—taking into account 18 school clusters. Cases had a mean age of 17.4 years (±1.14 years), controls were 16.1 years (±1.71 years) old on average, and 315 (40%) of the participants were exposed to boarding schooling. The unadjusted odds of pregnancy in the exposed and unexposed (day scholars) girls were 0.25 and 0.73, respectively (cOR = 0.34, CI: 0.24–0.48, p < 0.0001). Adjusting for other characteristics, teenage girls in boarding schools had 60% lower odds of pregnancy (aOR = 0.40, CI = 0.16–1.00, p = 0.049). In light of this evidence, enrollment of teenage girls in boarding schools is encouraged to help mitigate teen pregnancy in the province. Additionally, a multicenter prospective study is recommended.

The coronavirus disease 2019 (COVID‐19) has wreaked havoc globally, resulting in millions of cases and deaths. The objective of this study was to predict mortality in hospitalized COVID‐19 patients in Zambia using machine learning (ML) methods based on factors that have been shown to be predictive of mortality and thereby improve pandemic preparedness. This research employed seven powerful ML models that included decision tree (DT), random forest (RF), support vector machines (SVM), logistic regression (LR), Naïve Bayes (NB), gradient boosting (GB), and XGBoost (XGB). These classifiers were trained on 1,433 hospitalized COVID‐19 patients from various health facilities in Zambia. The performances achieved by these models were checked using accuracy, recall, F 1‐Score, area under the receiver operating characteristic curve (ROC_AUC), area under the precision‐recall curve (PRC_AUC), and other metrics. The best‐performing model was the XGB which had an accuracy of 92.3%, recall of 94.2%, F 1‐Score of 92.4%, and ROC_AUC of 97.5%. The pairwise Mann–Whitney U‐test analysis showed that the second‐best model (GB) and the third‐best model (RF) did not perform significantly worse than the best model (XGB) and had the following: GB had an accuracy of 91.7%, recall of 94.2%, F 1‐Score of 91.9%, and ROC_AUC of 97.1%. RF had an accuracy of 90.8%, recall of 93.6%, F 1‐Score of 91.0%, and ROC_AUC of 96.8%. Other models showed similar results for the same metrics checked. The study successfully derived and validated the selected ML models and predicted mortality effectively with reasonably high performance in the stated metrics. The feature importance analysis found that knowledge of underlying health conditions about patients’ hospital length of stay (LOS), white blood cell count, age, and other factors can help healthcare providers offer lifesaving services on time, improve pandemic preparedness, and decongest health facilities in Zambia and other countries with similar settings.

The increase in health research in sub-Saharan Africa (SSA) has led to a high demand for biostatisticians to develop study designs, contribute and apply statistical methods in data analyses. Initiatives exist to address the dearth in statistical capacity and lack of local biostatisticians in SSA health projects. The Sub-Saharan African Consortium for Advanced Biostatistics (SSACAB) led by African institutions was initiated to improve biostatistical capacity according to the needs identified by African institutions, through collaborative masters and doctoral training in biostatistics. SACCAB has created a critical mass of biostatisticians and a network of institutions over the last five years and has strengthened biostatistics resources and capacity for health research studies in SSA. SSACAB comprises 11 universities and four research institutions which are supported by four European universities. In 2015, only four universities had established Masters programmes in biostatistics and SSACAB supported the remaining seven to develop Masters programmes. In 2019 the University of the Witwatersrand became the first African institution to gain Royal Statistical Society accreditation for a Biostatistics Masters programme. A total of 150 fellows have been awarded scholarships to date of which 123 are Masters fellowships (41 female) of whom 58 have already graduated. Graduates have been employed in African academic (19) and research (15) institutions and 10 have enrolled for PhD studies. A total of 27 (10 female) PhD fellowships have been awarded; 4 of them are due to graduate by 2020. To date, SSACAB Masters and PhD students have published 17 and 31 peer-reviewed articles, respectively. SSACAB has also facilitated well-attended conferences, face-to-face and online short courses. Pooling of limited biostatistics resources in SSA combined with co-funding from external partners has shown to be an effective strategy for the development and teaching of advanced biostatistics methods, supervision and mentoring of PhD candidates.

Accurate estimates of gestational age at birth are important for preterm birth surveillance but can be challenging to reliably acquire in low and middle income countries. Our objective was to develop machine learning models to accurately estimate gestational age shortly after birth using clinical and metabolic data. We derived and internally validated three models using ELASTIC NET multivariable linear regression in heel prick blood samples and clinical data from a retrospective cohort of newborns from Ontario, Canada. We conducted external model validation in heel prick and cord blood sample data collected from prospective birth cohorts in Lusaka, Zambia (N=311) and Matlab, Bangladesh (N=1176). The best-performing model accurately estimated gestational age within about 6 days of early pregnancy ultrasound estimates in both cohorts when applied to heel prick data (MAE (95% CI) = 0.79 weeks (0.69, 0.90) for Zambia; 0.81 weeks (0.75, 0.86) for Bangladesh), and within about 7 days when applied to cord blood data (1.02 weeks (0.90, 1.15) for Zambia; 0.95 weeks (0.90, 0.99) for Bangladesh). Algorithms developed in Canada provided accurate estimates of gestational age when applied to external cohorts from Zambia and Bangladesh. Model performance was superior in heel prick data as compared to cord blood data.