150 research outputs found
Supplemental_Material – Supplemental material for The effect of a home exercise intervention on persons with lower limb amputations: a randomized controlled trial
No full textSupplemental material, Supplemental_Material for The effect of a home exercise intervention on persons with lower limb amputations: a randomized controlled trial by Lonwabo Godlwana, Aimee Stewart and Eustasius Musenge in Clinical Rehabilitation</p
Incidence estimation and calibration from cross-sectional data of acute infection HIV-1 seroconvertors
No full textABSTRACT
Incidence estimation and calibration from cross-sectional data of acute infection
HIV-1 seroconvertors.
May 2007
Eustasius Musenge
Masters in Medicine in the Field of Biostatistics and Epidemiology
Supervised by: Mr E Marinda and Dr A Welte
Background: The HIV-1 incidence (a very important measure used as a proxy for
disease burden) can be estimated from a cross-sectional study. This incidence estimate
has the advantage of reducing on costs and time, thus enabling more timely
intervention; it is also ideal for developing nations. A common procedure used in
making this estimate utilizes two antibody tests (Sensitive/Less sensitive tests). Due to
the long window period of such tests (at least three months), persons classified as
recently infected would have been infected more than three months prior to the test
date. Detecting acute HIV-1 infection is very important since this is the most infectious
stage of the disease. This research report explores a method of estimating incidence
using an antibody test and a virological test, Polymerase Chain Reaction Ribonucleic
Acid (PCR-RNA).The cross-sectional data used are from the Centre for the AIDS
Programme of Research in South Africa (CAPRISA).
Methods: Actual follow-up cohort data from CAPRISA acute infection cohort (AIC),
comprised of 245 sex workers, were used to estimate the incidence of HIV-1 using a
PCR-RNA ,virology test based, incidence formula. The result obtained was compared to
the incidence estimate obtained by the classical method of estimating incidence
the AIDS
Programme of Research in South Africa (CAPRISA).
Methods: Actual follow-up cohort data from CAPRISA acute infection cohort (AIC),
comprised of 245 sex workers, were used to estimate the incidence of HIV-1 using a
PCR-RNA ,virology test based, incidence formula. The result obtained was compared to
the incidence estimate obtained by the classical method of estimating incidence
(prospective cohort follow-up). As a measure to reduce costs inherent in virological
tests (PCR-RNA), multistage pooling was discussed and several pooling strategies
simulations were proposed with their uncertainties. Point estimates and interval
estimates of the window period, window period prevalence and incidence from crosssectional
study of the AIC cohort were computed.
Findings: The mean window period was 6.6 days 95% CI: (2.7 – 13.0). The monthly
window period prevalence was 0.09423 percent 95 % CI: (0.0193 – 0.1865)%. The
incidence from the prospective cohort follow-up was 5.43 percent 95% CI: (3.9 – 9.2)
%. The incidence estimate from cross-sectional formulae was 5.21 percent 95% CI:
(4.1– 4.6). It was also shown by use of simulations that an optimum pool sample size is
obtained when at least half the samples are removed on every run.
Interpretation and recommendations: The PCR-RNA test is very sensitive at
detecting acute HIV-1 infected persons. The incidence estimate from the crosssectional
study formulae was very similar to that obtained from a follow-up study. The
number of tests needed can be reduced and a good estimate of the incidence can still be
obtained. The calibration was not accurate since the samples used were small and the
window period duration was too short, hence, it was difficult to extrapolate to the whole
population. Further work still needs to be done on the calibration of the proposed
incidence formulae as it could be a very useful public health tool
Evaluating the accuracy of the CKD-EPI equations in estimating the glomerular filtration rate among adult Africans in Malawi, Uganda, and South Africa
No full textA research report submitted in fulfillment of the requirements for the Master of Science in Field Epidemiology, in the Faculty of Health Sciences, School of Public Health, University of the Witwatersrand, Johannesburg, 2024Background Chronic kidney disease (CKD) is a significant global health concern with a growing burden, affecting millions of individuals worldwide. CKD was ranked 18th amongst the highest non- communicable diseases cause of death worldwide. CKD is a global health issue, which reduces kidney function and increases cardiovascular risk. Accurate glomerular filtration rate (GFR) estimation is crucial for CKD management. The commonly used CKD-EPI (Chronic Kidney Disease-Epidemiology Collaboration) equations were primarily developed for Caucasian and African American populations. This study aimed to evaluate existing equations and develop new ones specific to the African population using data from Malawi, Uganda, and South Africa. Methods This study was a secondary analysis of data collected in three African countries collectively referred to as the ARK (African Research on Kidney Disease) Consortium. The study used data from 2433 participants, with plasma iohexol clearance as an indication of GFR (mGFR). Estimating equations were developed using serum cystatin C and or in combination with serum creatinine, mirroring the CKD-EPI equations by adopting a non-linear modelling approach. The study developed a predictive model using supervised machine learning techniques. Bland-Altman plots were used to assess linearity and agreement of the eGFR methods. Accuracy within 10% and 30% of mGFR, bias, and precision were assessed overall and by CKD stage. Results iv Analysis of 2433 participants from the three African countries revealed significant differences in mean measured glomerular filtration rate (mGFR) by country and sex. New serum cystatin C and creatinine-cystatin C-based equations for estimating GFR were developed, showing high accuracy ranging between (94-95%) and (93-95%), respectively for GFR ≥90 ml/min/1.73m2. The equations however had lower precision (2.07 – 2.12) compared to existing ARKM (African Research on Kidney Disease Model) equations (2.36 – 2.37). Six machine learning (ML) classification models were evaluated, with Random Forest emerging as the top performer, followed by Logistic Regression. ML approaches demonstrated higher F1 score measures (89%-100%) than eGFR equations, accuracies ranging between 75% and 95% and less bias. Overall, it was concluded for this study that ML techniques provide better performing models in comparison to the existing and developed eGFR models. This was evident as AUC measures for all ML models were higher (93% - 100%) than the accuracy measures of the eGFR equations (75% - 95%). Conclusion The results highlight the value of cystatin C as a biomarker for improving GFR estimation and underscore the importance of population-specific GFR estimation tools for African populations. While no single method is perfect across all GFR levels, the findings demonstrate the potential of both refined eGFR equations and machine learning models in enhancing GFR estimation accuracy. However, confirmation in broader populations is needed, and regular monitoring and adaptation of ML models will be required to maintain predictive performance over time. These findings can inform efforts to improve GFR estimation and CKD evaluation in Africa.MM202
Utilisation of maternal, newborn and child healthcare services in three sub-Saharan African countries (DRC, Kenya, and Tanzania) using Demographic Health Surveys data from 2007-2016: Application of Generalised Structural Equation and Machine Learning Models
No full textA research report submitted in fulfillment of the requirements for the Doctor of Philosophy, in the Faculty of Health Sciences, School of Public Health, University of the Witwatersrand, Johannesburg, 2024Background: The risk of child deaths within the first month of life is elevated than the later stages of childhood. Globally, Sub-Saharan Africa (SSA) has the highest neonatal mortality. Majority of the countries in SSA including the DRC, Kenya and Tanzania are struggling to meet Sustainable Development Goal (SDG) 3.2 of reducing the neonatal mortality rate to 12 deaths per 1,000 live births by 2030 (2). Most causes of neonatal deaths are preventable and treatable. Universal coverage, timely and effective utilisation of maternal, newborn, and child healthcare (MNCH) services during pregnancy, delivery, and postpartum has the potential to save many lives of newborns in high-burden countries. vii Antenatal care (ANC) is the first service offered to pregnant women in MNCH. The timing and frequency of ANC visits is critical for the mother and her unborn child. The WHO recommends that women initiate ANC within 16 weeks of pregnancy and attend a minimum of four ANC visits for timely and optimum care before delivery (3, 4). The WHO also recommends that pregnant women receive assistance from a skilled worker during delivery and get postnatal checks with their newborns within 6 weeks of delivery (5, 6). Furthermore, utilising the Continuum of Care (CoC) for MNCH could significantly reduce maternal and newborn deaths in SSA. In the context of MNCH, the CoC is an approach that ensures continuous care from the period of pregnancy, through to childbirth, postnatal period, infancy, and the childhood period (7). Despite the recognition of the use of vital services in MNCH, timely and adequate uptake of MNCH services remains poor and the coverage of MNCH is far from universal in SSA. Most pregnant women initiate ANC after 16 weeks and hence fail to receive timely ANC interventions (8). Uptake of ANC visits, skilled birth attendance (SBA) and postnatal care (PNC) is suboptimal (8-11). Studies in SSA have explored various factors associated with MNCH services utilisation, however, our understanding of MNCH services utilisation in SSA is still limited. Trends in utilisation of MNCH services over time such as late ANC uptake have not been thoroughly assessed. Late uptake of ANC is still a common problem in SSA. Tracking women’s progress in the timing of ANC will ascertain if they are any changes in women’s late uptake of ANC and the contributing factors. This information will guide future policies and programmes which focus on improving the timely uptake of ANC in the SGD era. There is also a dearth of empirical evidence on the factors associated with the utilisation of ANC, skilled delivery and postnatal care in the CoC using nationally representative data. The CoC views both the mother and child as a collective rather than as separate/ individual entities. Understanding factors that viii contribute to the full utilisation of drop out from the CoC is essential for the formulation of interventions than enhance the CoC. Furthermore, studies which investigated either the individual utilisation of MNCH services such as timing of ANC, ANC visits, SBA and PNC services or the CoC have tended to use more of the traditional analysis methods such as the logistic regression. The application of more versatile analysis methods such as Machine Learning is not common. Machine Learning methods are capable of extracting information that commonly used methods (logistic regression) fail to do by uncovering hidden patterns and relationships, particularly in large data sets (12). The application of Machine Learning methods can offer opportunities of enhancing existing methods (conventional regression methods) for predicting and classifying MNCH utilisation leading to more effective interventions to improve MNCH utilisation. There is also a limited understanding on the interrelationships between MNCH services utilisation and neonatal outcomes. The associations between MNCH services utilisation and newborn outcomes such as neonatal mortality are commonly assessed using traditional approaches that assume direct associations. Specific analytical methods, such as Generalised Structural Equation Modeling (GSEM) can be used to model complex relationships such as interrelated links between utilisation of different MNCH services and neonatal outcomes. GSEM gives a clear understanding of how different services of MNCH are related to one another with neonatal outcomes by estimating both direct and indirect paths associations for more effective targeted interventions. Given the critical role of MNCH in ending preventable neonatal mortality, the overarching aim of this study was to describe the utilisation of MNCH services and their associations with neonatal mortality using GSEM and Machine Learning models in three sub-Saharan African countries: the DRC, Kenya, and Tanzania. ix Methods: The study utilised cross-sectional secondary data of reproductive-age women from the Democratic Republic of Congo (DRC) (2007-2013/14), Kenya (2008-2014) and Tanzania (2010-2015/16) Demographic Health Surveys. Firstly, the multivariate logistic regression analysed factors associated with late ANC initiation accounting for clusters, survey weights and stratification for the different rounds of the Demographic Health Surveys. Trends in late initiation of ANC over time in each country were assessed by comparing the earlier and later surveys using differences in prediction scores (prediction probabilities generated after running the multivariate logistic regression models). Secondly, the study assessed the main predictors of non-utilisation of PNC using the Decision Tree. The model performance of the Decision Tree was compared to the Logistic Regression using Accuracy, Sensitivity, Specificity and area under the Receiver Operating Characteristics. Thirdly, factors associated with the drop out from the MNCH continuum, defined as not fully utilising either ANC, SBA, or PNC services, were analysed using multivariate logistic regression accounting for clusters, survey weights and stratification. Machine Learning analysis was used to predict the drop out from the MNCH continuum using features (predictors) that were found significant in the multivariate logistic regression. Five classification Machine Learning models were built and developed including the Artificial Neural Network, Decision Tree, Logistic Regression, Random Forest and Support Vector Machine to predict the drop out from the MNCH continuum. The prediction accuracies of the models were then compared using parameters including Accuracy, Precision, Recall, Specificity, F1 score and area under the Receiver Operating Characteristics. Fourthly, the Generalised Structural Equation Modeling (GSEM) was used to assess the mediatory role of MNCH services utilisation on neonatal mortality. The endogenous variables x were ANC attendance, SBA and PNC attendance, low birth weight and neonatal mortality. The GSEM analysis also accounted for survey weights and considered cluster random effects. Results: The findings showed a reduction in late ANC initiation (67.8%-60.5%) between 2008-2014 in Kenya as well as in Tanzania (60.9%-49.8%) between 2010-2016, but an increase was observed in the DRC (56.8%-61.0%) between 2007-2014. In the DRC, higher birth order was associated with ANC initiation delays from 2007-2014, whilst rural residency, lower maternal education and household income was linked to ANC initiation delays in 2014. In Kenya, lower maternal education and household income was associated with ANC initiation delays from 2008-2014, whilst rural residency and increased birth order were linked to ANC initiation delays in 2014. In Tanzania, higher birth order and larger households were linked to ANC delays from 2010-2016, whilst ANC initiation delays were associated with lower maternal education in 2010 and lower-income households in 2016. The results also showed that the Decision Tree models had higher prediction accuracy of non- utilisation of PNC than the Logistic Regression models. Using the Decision Tree, low quality of ANC, home deliveries and unemployment were associated with the highest probability of not utilising PNC (92.0%) in the DRC. In Kenya, home deliveries, unemployment and lack of access to mass media were associated the highest likelihood of not utilising PNC (87.0%). In Tanzania, home deliveries, low quality of ANC and unwanted pregnancies exhibited the highest likelihood of not utilising PNC (100.0%). The results also revealed very high rates of dropping out from the MNCH continuum in the DRC (91.0%), in Kenya (72.3%) and Tanzania (93.7%). Rural residence, lower maternal education and non-exposure to mass media were common predictors of dropping out from the MNCH continuum across the three countries. Further, the influence of factors such as xi household wealth, household size, access to money for medication, travel distance to health facilities, and parity and maternal age varied by country. Results from the Machine Learning analysis showed that the Logistic Regression had the least prediction accuracy, while the Random Forest exhibited the highest prediction accuracy. Using the Random Forest, the study further ranked the most important predictors of the drop out from the MNCH continuum. Household wealth, place of residence, maternal education and exposure to mass media were the top four most important predictors. The results also showed direct and indirect associations between MNCH services utilisation and neonatal mortality. ANC attendance mediated the total effects of PNC attendance on neonatal mortality by 8.8% in Kenya and 5.5% in Tanzania. ANC attendance and SBA also sequentially mediated the total effects of PNC attendance on neonatal mortality by 1.9% in Kenya and 1.0% in Tanzania. The results in Tanzania also showed ANC attendance mediated 2.8% of the total effects of LBW on neonatal mortality. No presence of mediation was observed in the DRC; however, ANC attendance moderated the relationship between parity and neonatal mortality. Conclusions: The study found that late uptake of ANC decreased between the two survey rounds in Kenya and Tanzania but increased in the DRC. Women from various geographic, educational, parity, and economic groups showed varying levels of late ANC uptake. Increasing women’s access to information platforms and strengthening initiatives that enhance female education, household incomes, and localise services may enhance early ANC uptake. The Decision Tree models showed higher prediction accuracy of non-utilisation of PNC than the Logistic Regression models in the DRC, Kenya and Tanzania. Using the Decision Tree, women who had poor quality of ANC, home deliveries, unemployment, unplanned pregnancies, and no mass media access were identified as high-risk subpopulations of non- xii utilisation of PNC. Improving access and quality of care, incorporation of TBAs into the formal health systems, government health financing, increasing access to mass media and integrating maternal healthcare services with family planning services should be considered as top priority interventions to improving the utilisation of PNC. Most women and children drop out of the MNCH continuum in the DRC, Kenya and Tanzania. Rural residence, lower maternal education and non-exposure to mass media were common factors linked to the high dropout in the MNCH continuum. The use of Machine Learning can help support evidence-based decisions in MNCH interventions. Rapid response mechanisms such as web-based applications can also be developed through the use Machine Learning whereby a pregnant woman’s future utilisation of the services in CoC is assessed and monitored in real-time. The GSEM findings showed interconnections between MNCH services utilisation such as timing of ANC, ANC visits, SBA, PNC and neonatal mortality. This suggests that more than direct and indirect factors are accountable for the associations between MNCH services utilisation and neonatal mortality. The mediation role of MNCH services on neonatal mortality indicates critical areas for targeted interventions to reduce neonatal mortality. Overall, the study aimed to describe the utilisation of the MNCH services and associations with neonatal mortality in the DRC, Kenya and Tanzania. The study showed declines in late ANC uptake in two countries, however, early uptake of ANC is far is still not universal. The study also showed very low levels of retention in the CoC, and most women and children drop out in the CoC at postpartum period. The findings also showed the existence of social, health system and individual inequalities in MNCH and their impact on early childhood survival. Women who are vulnerable to unequal and poor MNCH services utilisation are characterised by poverty, rural residence, long travel distances to health facilities, unaffordable medical expenses, home deliveries, low quality of xiii care, low education, high parity, younger age, unemployment, limited exposure to mass media, and unplanned pregnancies. Context-specific intervention programs such as female education, government health financing, MNCH promotion programs through mass media and improved accessibility and quality of care in health facilities, particularly for the most vulnerable groups of the populations such as women of low socioeconomic status and women from underserved rural areas are essential to improve the overall health of mothers and children and meeting the SDG-3 goals. Modern biostatistical models like Machine Learning provide essential tools to understand public health problems. These techniques should be applied to complement the conventional statistical methods, particularly the tree-based models like the Decision Tree and Random Forest for predicting and classifying the utilisation of MNCH services. The GSEM established interconnections between timing of ANC, ANC visits, SBA and PNC and neonatal mortality. The timing of the first ANC contact is an important starting point to a continuation through the COC. It makes women better informed about pregnancy and the subsequent use of MNCH services. All stakeholders should work more on promoting early uptake of ANC by setting up initiatives that increase women’s access to information platforms, enhance female education, improve household incomes, and bring services closer to communities.MM202
HIV-exposure as a risk factor for mortality among neonates with culture- confirmed bloodstream infection and meningitis in South Africa, 2019- 2020
No full textA research report submitted in fulfillment of the requirements for the Master of Science in Field Epidemiology, in the Faculty of Health Sciences, School of Public Health, University of the Witwatersrand, Johannesburg, 2024Background: HIV-exposed but uninfected neonates (HEU) are a growing population. Exposure to HIV has been associated with increased mortality and morbidity. We aimed to determine the effect of HIV-exposure as a risk factor for mortality in neonates admitted with bloodstream infections (BSIs) and/or meningitis at non-academic hospitals in South Africa. Methods: We conducted a retrospective cohort study using data from the Baby GERMS-SA surveillance project of hospitalised neonates with culture-confirmed BSI and meningitis at six non-academic hospitals in South Africa from October 2019 to September 2020. A multivariable Cox proportional hazards regression was used to determine the effect of HIV- exposure regardless of HIV-status as a risk factor for mortality. We further examined the effect of HIV-exposure using a multivariable logistics regression. Results: Of 697 neonates with a known maternal HIV status and in-hospital outcome, 34% (239/697) were exposed to HIV and 1% (4/239) were HIV PCR-positive. The HEU neonates had significant low gestational age (77% (184/239) vs. 66% (296/458), p=0.001) and very low birth weight (48% (115/239) vs. 40% (184/458), p=0.016) compared to HIV-unexposed uninfected (HUU) neonates. Exclusive breastfeeding was more common in HUU neonates (44% (202/458) vs. 32% (77/239)). We did not observe significant differences in age at the time of infection (median age 6 vs. 6 days p=0.14), and duration of hospitalisation (median length of 17 vs. 15 days p=0.12) between the HEU and HUU neonates. The crude in-hospital mortality among HIV-exposed neonates and HUU neonates was 26% (63/239) and 23% (104/458), respectively. After adjusting for relevant confounders such as birth weight, timing of infection, use of invasive devices, breastfeeding, and maternal age, there was no difference in the risk for mortality between HEU neonates and those who were HUU (HR 1.12, 95% CI: 0.76-1.67, p=0.549) at 28-days. The odds of mortality were 1.21 (95% CI 0.72–2.05, v p=0.467) times more among HEU neonates than among HUU neonates in the exploratory analysis. Conclusions: We did not find a difference in mortality between HEU and HUU neonates with culture-confirmed invasive infections in our study. Regardless of their HIV exposure status, approximately a quarter of these neonates died in hospital.MM202
The association between intermittent preventive treatment uptake and anaemia amongst pregnant women in Zambia in 2018: a spatial analysis
No full textA research report submitted in partial fulfillment of the requirements for the degree of Master of Science in Epidemiology (Implementation Science) to the Faculty of Health Sciences, School of Public Health, University of the Witwatersrand, Johannesburg 2024Background: This study investigated the association between Intermittent Preventive Treatment with Sulphadoxine Pyrimethamine (IPTp-SP) uptake and anaemia among Zambian pregnant women aged 15-49 in 2018. Despite WHO's endorsement of IPTp-SP to combat malaria-related anaemia, its prevalence continued to rise, significantly impacting maternal health.
Methodology: Using Zambia Demographic Health Survey 2018 data, 665 pregnant women receiving IPTp-SP were analysed for haemoglobin levels, determining anaemia through blood tests. Statistical methods included survey-adjusted proportions, means, bivariate analysis, multiple linear regression, and multi-level ordinal logistic models with spatial random effects. Spatial analyses used ArcMap for coverage analysis, ordinary least squares, and geographically weighted regression maps (GWR) techniques in R and Stata.
Results: Optimal IPTp-SP doses resulted in 36.98% anaemia prevalence (124/369), and suboptimal doses led to 42.85% (112/296). Factors associated with anaemia included household size, rich wealth index, high parity, and employment during pregnancy. Associations between IPTp-SP uptake and anaemia were identified: household size (four to six: AOR= 0.53; 95% CI 0.34 to 0.80; seven or more: AOR=0.57; 95% CI 0.35 to 0.91), adequate antenatal visits (AOR=0.68; 95% CI 0.48 to 0.97), and rich wealth index (AOR= 0.68; 95% CI 0.34 to 0.98). Spatial analysis revealed anaemia hotspots in Southern, Luapula, and Eastern provinces, with iron
supplements and household size identified as influential factors.
Conclusion: Despite IPTp-SP use, overall anaemia prevalence was 40%, with the highest rates in Southern, Luapula, and Western provinces. Targeted strategies focusing on improving iron tablet access, antenatal care attendance, and utilising spatial maps are crucial for mitigating adverse anaemia outcomes in these regionMM202
Modelling spatiotemporal patterns of childhood HIV/TB related mortality and malnutrition: applications to Agincourt data in rural South Africa
No full textBackground: South Africa accounts for more than a seventh of the global population living
with HIV/AIDS and TB, and ranks highest in HIV/TB co-infection worldwide. Consequent
high child mortality is exacerbated by child malnutrition, which is an important indicator of
health status and is associated with morbidity as well as mortality. Rural areas usually present with the greatest burden of morbidity and mortality, yet the extent of geographical disparities in child mortality, malnutrition and HIV/TB has hardly been explored. This is a reservoir of information useful for effective public health interventions. In this thesis we investigated the factors associated with childhood HIV/TB mortality and malnutrition, how they interrelate and their spatial distribution in the rural Agincourt sub-district located in north-east South Africa close to the border with Mozambique.
Rationale: Africa at large lacks data that are routinely and reliably collected then validated, to guide policy and intervention programmes. Causes of deaths and even death counts are often misclassified and underestimated respectively, especially for children. To bridge this gap, a health and socio-demographic surveillance systems located in the rural Agincourt sub-district hosts which annually collects and collates data on vital events including fertility, mortality and migration. These data have been collected since 1992 to-date and now cover 80,000 people living in more than 16,000 households situated in 27 villages; all households are fully
geo-coded. These hierarchical data allow us to address several epidemiological questions on how person, place (spatial) and time (temporality) have impacted on mortality and
malnutrition patterns in children living in the rural Agincourt sub-district.
Objectives: The aims of this thesis were both methodological and applied:
Methodological
(1) To investigate the presence of spatial autocorrelation in the Agincourt sub-district
and model this using geographical and geo-statistical procedures
(2) To model large spatial random effects accurately and efficiently
(3) To model hierarchical data with zero inflated outcomes
Applied
(1) To investigate childhood HIV/TB mortality determinants and their geographical
distribution using retrospective and cross-sectional data
(2) To determine factors associated with malnutrition outcomes adjusting for their
multivariate spatial random effects and selection bias for children under five years
(3) To model how the associated factors were interrelated as either underlying or
proximate factors of child mortality or malnutrition using pathway analysis.
Methods: We conducted a secondary data analysis based on retrospective and cross-sectional data collected from 1992 to 2010 from the Agincourt sub-district in rural northeast South Africa. During the period of our study 71,057 children aged 0 to 9 years from 15,703 households were observed. All the data in the thesis were for children aged 1 to under 5 except for the chapter 6 (last paper) who were aged from 0 to 9 years of age. Child HIV/TB death and malnutrition were the outcome measures; mortality was derived from physicianbased verbal autopsy. We investigated presence of spatial autocorrelation using Moran’s and Geary’s coefficients, semi-variograms and estimated the spatial parameters using Bayesianbased univariate and multivariate procedures. Regression modelling that adjusted for spatial random effects was done using linear regression and zero inflated variants for logistic, Poisson and Negative Binomial regression models. Structural equation models were used in modelling the complex relationships between multiple exposures and child HIV/TB mortality and malnutrition portrayed by conceptual frameworks. Risk maps were drawn based on spatial residuals (posteriors) with prediction (kriging) procedures used to estimate for households where no data were observed. Statistical inference on parameter estimation was done using both the frequentist; maximum likelihood estimation and Bayesian; Markov Chain Monte Carlo (MCMC) directly and sometimes aided with Metropolis Hastings or Integrated
Nested Laplace Approximations (INLA).
Results: The levels of child under-nutrition in this area were: 6.6% wasted, 17.3% stunted
and 9.9% underweight. Moran’s (I) and Geary’s (c) coefficients indicated that there was
global and local clustering respectively. Estimated severity of spatial variation using the
partial-sill-to-sill ratio yielded 12.1%, 4.7% and 16.5%, for weight-for-age, height-for-age and weight-for-height Z-scores measures respectively. Maternal death had the greatest negative impact on child HIV/TB mortality. Other determinants included being a male child and belonging to a household that had experienced multiple deaths. A protective effect was found in households with better socio-economic status and where older children were present. Pathway analyses of these factors showed that HIV had a significant mediator effect and the greatest worsening effect on malnutrition after controlling for low birth-weight selection bias Several spatial hot spots of mortality and malnutrition were observed, with these regions consistently emerging as areas of greater risk, which reinforces geographical differentials in these public health indicators.
Conclusion: Modelling that adjusts for spatial random effects, is a potentially useful
technique to disclose hidden patterns. These geographical differences are often ignored in
epidemiological regression modelling resulting in reporting of biased estimates. Proximate
and underlying determinants, notably socioeconomic status and maternal deaths, impacteddirectly and indirectly on child mortality and malnutrition. These factors are highly relevant locally and should be used to formulate interventions to reduce child mortality. Spatial prediction maps can guide policy on where to best target interventions. Child interventions can be more effective if there is a dual focus: treatment and care for those already HIV/TB infected, coupled with prevention in those geographical areas of greatest risk. Public health population-level interventions aimed at reducing child malnutrition are pivotal in lowering morbidity and mortality in remote areas.
Keywords: HIV/TB, Child mortality, Child malnutrition, Conceptual framework, Spatial
analysis, MCMC, Path analysis, South Afric
LATENT TUBERCULOSIS INFECTION PREVALENCE, SPATIAL CLUSTERING AND RISK FACTORS IN A SOUTH AFRICAN URBAN INFORMAL SETTLEMENT
No full textThis dissertation investigated the burden spatial clustering and risk factors of latent tuberculosis infection (LTBI), in a South African urban informal settlement. Using data from a large community-based household survey with random sampling and from the 2011 South African census as disseminated by Statistics South Africa (STATSSA), we estimated the prevalence of LTBI in the general population, the annual risk of infection (ARI) in children, and investigated individual-, household- and neighborhood-level factors associated with LTBI (paper 1). We assessed spatial heterogeneity of LTBI prevalence and the association between community-level factors and LTBI clusters (paper 2). In paper 1, we observed that the overall prevalence of LTBI was 34.3% (95% CI, 30% – 39%), the annual risk of infection among children age 0-14 years was 3.1% (95% CI: 2.1 - 5.2). In multivariable logistic regression analysis, LTBI was associated with age, male gender, marital status, and higher socio-economic status. In paper 2, we investigated the spatial clustering and spatial heterogeneity of LTBI prevalence and predictive community-level factors. One statistically significant cluster of high LTBI prevalence was found using the spatial scan statistic. Higher socio-economic status (SES) was associated with higher LTBI prevalence in both a non-spatial regression model and a geographically weighted regression (GWR) model. However, only a small part of the spatial heterogeneity in LTBI prevalence was explained by variation in community-level SES, suggesting that further research is needed to better understand the determinants of LTBI in such settings. Overall, this dissertation suggests that spatial analysis of LTBI can identify clusters within a single community and that LTBI prevalence is not associated with HIV status but may be associated with higher SES, in contrast to the well-established association between TB disease, HIV, and poverty.Doctor of Philosoph
Space-time confounding adjusted determinants of child HIV/TB mortality for large zero-inflated data in rural South Africa
No full textAbstractSouth Africa is experiencing a major burden of HIV/TB. We used longitudinal data from the Agincourt sub-district in rural northeast South Africa over the years 2000 to 2005. A total of 187 HIV/TB deaths were observed among 16,844 children aged 1–5years coming from 8,863 households. In this paper we used Bayesian models to assess risk factors for child HIV/TB mortality taking into account the presence of spatial correlation. Bayesian zero inflated spatiotemporal models were able to detect hidden patterns within the data. Our main finding was that maternal orphans experienced a threefold greater risk of HIV/TB death compared to those with living mothers (AHR=2.93, 95% CI[1.29;6.93]). Risk factor analyses which adjust for person, place and time provide evidence for policy makers that includes a spatial distribution of risk. Child survival is dependent on the mother’s survival; hence programs that promote maternal survival are critical
Predicting the drop out from the maternal, newborn and child healthcare continuum in three East African Community countries: application of machine learning models
No full textAbstract Background For optimal health, the maternal, newborn, and child healthcare (MNCH) continuum necessitates that the mother/child receive the full package of antenatal, intrapartum, and postnatal care. In sub-Saharan Africa, dropping out from the MNCH continuum remains a challenge. Using machine learning, the study sought to forecast the MNCH continuum drop out and determine important predictors in three East African Community (EAC) countries. Methods The study utilised Demographic Health Surveys data from the Democratic Republic of Congo (DRC) (2013/14), Kenya (2014) and Tanzania (2015/16). STATA 17 was used to perform the multivariate logistic regression. Python 3.0 was used to build five machine learning classification models namely the Logistic Regression, Random Forest, Decision Tree, Support Vector Machine and Artificial Neural Network. Performance of the models was assessed using Accuracy, Precision, Recall, Specificity, F1 score and area under the Receiver Operating Characteristics (AUROC). Results The prevalence of the drop out from the MNCH continuum was 91.0% in the DRC, 72.4% in Kenya and 93.6% in Tanzania. Living in the rural areas significantly increased the odds of dropping out from the MNCH continuum in the DRC (AOR:1.76;95%CI:1.30–2.38), Kenya (AOR:1.23;95%CI:1.03–1.47) and Tanzania (AOR:1.41;95%CI:1.01–1.97). Lower maternal education also conferred a significant increase in the DRC (AOR:2.16;95%CI:1.67–2.79), Kenya (AOR:1.56;95%CI:1.30–1.84) and Tanzania (AOR:1.70;95%CI:1.24–2.34). Non exposure to mass media also conferred a significant positive influence in the DRC (AOR:1.49;95%CI:1.15–1.95), Kenya (AOR:1.46;95%CI:1.19–1.80) and Tanzania (AOR:1.65;95%CI:1.13–2.40). The Random Forest exhibited superior predictive accuracy (Accuracy = 75.7%, Precision = 79.1%, Recall = 92.1%, Specificity = 51.6%, F1 score = 85.1%, AUROC = 70%). The top four predictors with the greatest influence were household wealth, place of residence, maternal education and exposure to mass media. Conclusions The MNCH continuum dropout rate is very high in the EAC countries. Maternal education, place of residence, and mass media exposure were common contributing factors to the drop out from MNCH continuum. The Random Forest had the highest predictive accuracy. Household wealth, place of residence, maternal education and exposure to mass media were ranked among the top four features with significant influence. The findings of this study can be used to support evidence-based decisions in MNCH interventions and to develop web-based services to improve continuity of care retention
- …
