Cost-effectiveness of a radio intervention to stimulate early childhood development: protocol for an economic evaluation of the SUNRISE trial in Burkina Faso.

INTRODUCTION: Approximately 250 million children under 5 years of age are at risk of poor development in low-income and middle-income countries. However, existing early childhood development (ECD) interventions can be expensive, labour intensive and challenging to deliver at scale. Mass media may offer an alternative approach to ECD intervention. This protocol describes the planned economic evaluation of a cluster-randomised controlled trial of a radio campaign promoting responsive caregiving and opportunities for early learning during the first 3 years of life in rural Burkina Faso (SUNRISE trial). METHODS AND ANALYSIS: The economic evaluation of the SUNRISE trial will be conducted as a within-trial analysis from the provider's perspective. Incremental costs and health outcomes of the radio campaign will be compared with standard broadcasting (ie, 'do nothing' comparator). All costs associated with creating and broadcasting the radio campaign during intervention start-up and implementation will be captured. The cost per child under 3 years old reached by the intervention will be calculated. Incremental cost-effectiveness ratios will be calculated for the trial's primary outcome (ie, incremental cost per SD of cognitive gain). A cost-consequence analysis will also be presented, whereby all relevant costs and outcomes are tabulated. Finally, an analysis will be conducted to assess the equity impact of the intervention. ETHICS AND DISSEMINATION: The SUNRISE trial has ethical approval from the ethics committees of the Ministry of Health, Burkina Faso, University College London and the London School of Hygiene and Tropical Medicine. The results of the economic evaluation will be disseminated in a peer-reviewed journal and presented at a relevant international conference. TRIAL REGISTRATION NUMBER: The SUNRISE trial was registered with ClinicalTrials.gov on 19 April 2019 (identifier: NCT05335395).

Early to mid-pregnancy HbA1c levels and its association with adverse pregnancy outcomes in three low middle-income countries in Asia and Sub-Saharan Africa.

BACKGROUND: Hyperglycemia during pregnancy leads to adverse maternal and fetal outcomes. Thus, strict monitoring of blood glucose levels is warranted. This study aims to determine the association of early to mid-pregnancy HbA1c levels with the development of pregnancy complications in women from three countries in South Asia and Sub-Saharan Africa. METHODS: We performed a secondary analysis of the AMANHI (Alliance for Maternal and Newborn Health Improvement) cohort, which enrolled 10,001 pregnant women between May 2014 and June 2018 across Sylhet-Bangladesh, Karachi-Pakistan, and Pemba Island-Tanzania. HbA1c assays were performed at enrollment (8 to < 20 gestational weeks), and epidemiological data were collected during 2-3 monthly household visits. The women were followed-up till the postpartum period to determine the pregnancy outcomes. Multivariable logistic regression models assessed the association between elevated HbA1c levels and adverse events while controlling for potential confounders. RESULTS: A total of 9,510 pregnant women were included in the analysis. The mean HbA1c level at enrollment was found to be the highest in Bangladesh (5.31 ± 0.37), followed by Tanzania (5.22 ± 0.49) and then Pakistan (5.07 ± 0.58). We report 339 stillbirths and 9,039 live births. Among the live births were 892 preterm births, 892 deliveries via cesarean section, and 532 LGA babies. In the multivariate pooled analysis, maternal HbA1c levels of ≥ 6.5 were associated with increased risks of stillbirths (aRR = 6.3, 95% CI = 3.4,11.6); preterm births (aRR = 3.5, 95% CI = 1.8-6.7); and Large for Gestational Age (aRR = 5.5, 95% CI = 2.9-10.6). CONCLUSION: Maternal HbA1c level is an independent risk factor for predicting adverse pregnancy outcomes such as stillbirth, preterm birth, and LGA among women in South Asia and Sub-Saharan Africa. These groups may benefit from early interventional strategies.

Implementing maternal and newborn health quality of care standards in healthcare facilities to improve the adoption of respectful maternity care in Bangladesh, Ghana and Tanzania: a controlled before and after study.

INTRODUCTION: Many women worldwide cannot access respectful maternity care (RMC). We assessed the effect of implementing maternal and newborn health (MNH) quality of care standards on RMC measures. METHODS: We used a facility-based controlled before and after design in 43 healthcare facilities in Bangladesh, Ghana and Tanzania. Interviews with women and health workers and observations of labour and childbirth were used for data collection. We estimated difference-in-differences to compare changes in RMC measures over time between groups. RESULTS: 1827 women and 818 health workers were interviewed, and 1512 observations were performed. In Bangladesh, MNH quality of care standards reduced physical abuse (DiD -5.2;-9.0 to -1.4). The standards increased RMC training (DiD 59.0; 33.4 to 84.6) and the availability of policies and procedures for both addressing patient concerns (DiD 46.0; 4.7 to 87.4) and identifying/reporting abuse (DiD 45.9; 19.9 to 71.8). The control facilities showed greater improvements in communicating the delivery plan (DiD -33.8; -62.9 to -4.6). Other measures improved in both groups, except for satisfaction with hygiene. In Ghana, the intervention improved women's experiences. Providers allowed women to ask questions and express concerns (DiD 37.5; 5.9 to 69.0), considered concerns (DiD 14.9; 4.9 to 24.9), reduced verbal abuse (DiD -8.0; -12.1 to -3.8) and physical abuse (DiD -5.2; -11.4 to -0.9). More women reported they would choose the facility for another delivery (DiD 17.5; 5.5 to 29.4). In Tanzania, women in the intervention facilities reported improvements in privacy (DiD 24.2; 0.2 to 48.3). No other significant differences were observed due to improvements in both groups. CONCLUSION: Institutionalising care standards and creating an enabling environment for quality MNH care is feasible in low and middle-income countries and may facilitate the adoption of RMC.

Malaria vaccine-related adverse events among children under 5 in sub-Saharan Africa: systematic review and meta-analysis protocol.

INTRODUCTION: The RTS,S vaccine has been approved for use in children under 5 living in moderate to high malaria transmission areas. However, clinically important adverse events have been reported in countries in sub-Saharan Africa. This systematic review aims to assess the frequency, severity and clinical importance of vaccine-related adverse events. METHODS AND ANALYSIS: This systematic review protocol has been prepared following robust methods and reported in line with the Preferred Reporting Items for Systematic reviews and Meta-Analyses for protocols guidelines. We will search PubMed, CINAHL, LILACS, Google Scholar, SCOPUS, WEB OF SCIENCE, Cochrane library, HINARI, African Journals Online, Trip Pro and TOXNET from 2000 to 30 September 2023, without language restrictions. We will also search conference proceedings, dissertations, World Bank Open Knowledge Repository, and WHO, PATH, UNICEF, Food and Drugs Authorities and European Medicines Agency databases, preprint repositories and reference lists of relevant studies for additional studies. Experts in the field will be contacted for unpublished or published studies missed by our searches. At least two reviewers will independently select studies and extract data using pretested tools and assess risk of bias in the included studies using the Cochrane risk of bias tool. Any disagreements will be resolved through discussion between the reviewers. Heterogeneity will be explored graphically, and statistically using the I2 statistic. We will conduct random-effects meta-analysis when heterogeneity is appreciable, and express dichotomous outcomes (serious adverse events, cerebral malaria and febrile convulsion) as risk ratio (RR) with their 95% CI. We will perform subgroup analysis to assess the impact of heterogeneity and sensitivity analyses to test the robustness of the effect estimates. The overall level of evidence will be assessed using Grading of Recommendations Assessment, Development and Evaluation. ETHICS AND DISSEMINATION: Ethical approval is not required for a systematic review. The findings of this study will be disseminated through stakeholder forums, conferences and peer-review publications. PROSPERO REGISTRATION NUMBER: CRD42021275155.

Design and implementation of a community-based mother-to-mother peer support programme for the follow-up of low birthweight infants in rural western Kenya.

Background: Globally, low birthweight (LBW) infants (<2,500 g) have the highest risk of mortality during the first year of life. Those who survive often have adverse health outcomes. Post-discharge outcomes of LBW infants in impoverished communities in Africa are largely unknown. This paper describes the design and implementation of a mother-to-mother peer training and mentoring programme for the follow-up of LBW infants in rural Kenya. Methods: Key informant interviews were conducted with 10 mothers of neonates (infants <28 days) from two rural communities in western Kenya. These data informed the identification of key characteristics required for mother-to-mother peer supporters (peer mothers) following up LBW infants post discharge. Forty potential peer mothers were invited to attend a 5-day training programme that focused on three main themes: supportive care using appropriate communication, identification of severe illness, and recommended care strategies for LBW infants. Sixteen peer mothers were mentored to conduct seven community follow-up visits to each mother-LBW infant pair with fifteen completing all the visits over a 6-month period. A mixed methods approach was used to evaluate the implementation of the programme. Quantitative data of peer mother socio-demographic characteristics, recruitment, and retention was collected. Two post-training focus group discussions were conducted with the peer mothers to explore their experiences of the programme. Descriptive statistics were generated from the quantitative data and the qualitative data was analysed using a thematic framework. Results: The median age of the peer mothers was 26 years (range 21-43). From March-August 2019, each peer mother conducted a median of 28 visits (range 7-77) with fourteen (88%) completing all their assigned follow-up visits. Post training, our interviews suggest that peer mothers felt empowered to promote appropriate infant feeding practices. They gave multiple examples of improved health seeking behaviours as a result of the peer mother training programme. Conclusion: Our peer mother training programme equipped peer mothers with the knowledge and skills for the post-discharge follow-up of LBW infants in this rural community in Kenya. Community-based interventions for LBW infants, delivered by appropriately trained peer mothers, have the potential to address the current gaps in post-discharge care for these infants.

Population cause of death estimation using verbal autopsy methods in large-scale field trials of maternal and child health: lessons learned from a 20-year research collaboration in Central Ghana.

Low and middle-income countries continue to use Verbal autopsies (VAs) as a World Health Organisation-recommended method to ascertain causes of death in settings where coverage of vital registration systems is not yet comprehensive. Whilst the adoption of VA has resulted in major improvements in estimating cause-specific mortality in many settings, well documented limitations have been identified relating to the standardisation of the processes involved. The WHO has invested significant resources into addressing concerns in some of these areas; there however remains enduring challenges particularly in operationalising VA surveys for deaths amongst women and children, challenges which have measurable impacts on the quality of data collected and on the accuracy of determining the final cause of death. In this paper we describe some of our key experiences and recommendations in conducting VAs from over two decades of evaluating seminal trials of maternal and child health interventions in rural Ghana. We focus on challenges along the entire VA pathway that can impact on the success rates of ascertaining the final cause of death, and lessons we have learned to optimise the procedures. We highlight our experiences of the value of the open history narratives in VAs and the training and skills required to optimise the quality of the information collected. We describe key issues in methods for ascertaining cause of death and argue that both automated and physician-based methods can be valid depending on the setting. We further summarise how increasingly popular information technology methods may be used to facilitate the processes described. Verbal autopsy is a vital means of increasing the coverage of accurate mortality statistics in low- and middle-income settings, however operationalisation remains problematic. The lessons we share here in conducting VAs within a long-term surveillance system in Ghana will be applicable to researchers and policymakers in many similar settings.

Implementation of distance learning IMCI training in rural districts of Tanzania.

BACKGROUND: The standard face-to-face training for the integrated management of childhood illness (IMCI) continues to be plagued by concerns of low coverage of trainees, the prolonged absence of trainees from the health facility to attend training and the high cost of training. Consequently, the distance learning IMCI training model is increasingly being promoted to address some of these challenges in resource-limited settings. This paper examines participants' accounts of the paper-based IMCI distance learning training programme in three district councils in Mbeya region, Tanzania. METHODS: A cross-sectional qualitative descriptive design was employed as part of an endline evaluation study of the management of possible serious bacterial infection in Busokelo, Kyela and Mbarali district councils of Mbeya Region in Tanzania. Key informant interviews were conducted with purposefully selected policymakers, partners, programme managers and healthcare workers, including beneficiaries and training facilitators. RESULTS: About 60 key informant interviews were conducted, of which 53% of participants were healthcare workers, including nurses, clinicians and pharmacists, and 22% were healthcare administrators, including district medical officers, reproductive and child health coordinators and programme officers. The findings indicate that the distance learning IMCI training model (DIMCI) was designed to address concerns about the standard IMCI model by enhancing efficiency, increasing outputs and reducing training costs. DIMCI included a mix of brief face-to-face orientation sessions, several weeks of self-directed learning, group discussions and brief face-to-face review sessions with facilitators. The DIMCI course covered topics related to management of sick newborns, referral decisions and reporting with nurses and clinicians as the main beneficiaries of the training. The problems with DIMCI included technological challenges related to limited access to proper learning technology (e.g., computers) and unfriendly learning materials. Personal challenges included work-study-family demands, and design and coordination challenges, including low financial incentives, which contributed to participants defaulting, and limited mentorship and follow-up due to limited funding and transport. CONCLUSION: DIMCI was implemented successfully in rural Tanzania. It facilitated the training of many healthcare workers at low cost and resulted in improved knowledge, competence and confidence among healthcare workers in managing sick newborns. However, technological, personal, and design and coordination challenges continue to face learners in rural areas; these will need to be addressed to maximize the success of DIMCI.

Prevalence and risk factors of Severe Acute Respiratory Syndrome Coronavirus 2 infection in women and children in peri-urban communities in Pakistan: A prospective cohort study.

Background: Population-based seroepidemiological surveys provide accurate estimates of disease burden. We compare the COVID-19 prevalence estimates from two serial serological surveys and the associated risk factors among women and children in a peri-urban area of Karachi, Pakistan. Methods: The AMANHI-COVID-19 study enrolled women and children between November 2020 and March 2021. Blood samples were collected from March to June 2021 (baseline) and September to December 2021 (follow-up) to test for anti-SARS-CoV-2 antibodies using ROCHE Elecsys®. Participants were visited or called weekly during the study for recording symptoms of COVID-19. We report the proportion of participants with anti-SARS-CoV-2 antibodies and symptoms in each survey and describe infection risk factors using step-wise binomial regression analysis. Results: The adjusted seroprevalence among women was 45.3% (95% confidence interval (CI) = 42.6-47.9) and 82.3% (95% CI = 79.9-84.4) at baseline and follow-up survey, respectively. Among children, it was 18.4% (95% CI = 16.1-20.7) and 57.4% (95% CI = 54.3-60.3) at baseline and follow-up, respectively. Of the women who were previously seronegative, 404 (74.4%) tested positive at the follow-up survey, as did 365 (50.4%) previously seronegative children. There was a high proportion of asymptomatic infection. At baseline, being poorest and lacking access to safe drinking water lowered the risk of infection for both women (risk ratio (RR) = 0.8, 95% CI = 0.7-0.9 and RR = 1.2, 95% CI = 1.1-1.4, respectively) and children (RR = 0.7, 95% CI = 0.5-1.0 and RR = 1.4, 95% CI = 1.0-1.8, respectively). At the follow-up survey, the risk of infection was lower for underweight women and children (RR = 0.4, 95% CI = 0.3-0.7 and RR = 0.7, 95% CI = 0.5-0.8, respectively) and for women in the 30-39 years age group and children who were 24-36 months of age (RR = 0.6, 95% CI = 0.4-0.9 and RR = 0.7, 95% CI = 0.5-0.9, respectively). In both surveys, paternal employment was an important predictor of seropositivity among children (RR = 0.7, 95% CI = 0.6-0.9 and RR = 0.8, 95% CI = 0.7-1.0, respectively). Conclusion: There was a high rate of seroconversion among women and children. Infection was generally mild. Parental education plays an important role in protection of children from COVID-19.

Evidence of High Frequencies of Insecticide Resistance Mutations in Aedes aegypti (Culicidae) Mosquitoes in Urban Accra, Ghana: Implications for Insecticide-based Vector Control of Aedes-borne Arboviral Diseases.

The most widespread arboviral diseases such as Dengue, Chikungunya, and Zika are transmitted mainly by Aedes mosquitoes. Due to the lack of effective therapeutics for most of these diseases, vector control remains the most effective preventative and control measure. This study investigated and compared the species composition, insecticide susceptibility, and resistance mechanisms in Aedes mosquito populations from a forest reserve converted to an eco-park and a peri-domestic sites in urban Accra, Ghana. Immature Aedes were sampled from the study sites, raised to adults, and exposed to deltamethrin, permethrin, DDT, fenitrothion, bendiocarb, permethrin + PBO, and deltamethrin + PBO using WHO tube assays. Melting curve analyses were performed for F1536C, V1016I, and V410L genetic mutations in surviving and dead mosquitoes following exposure to deltamethrin and permethrin. Microplate assay was used to access enzyme activity levels in adult mosquitoes from both populations. Aedes aegypti was found to be the dominant species from both study populations. The susceptibility test results revealed a high frequency of resistance to all the insecticides except fenitrothion. F1534C mutations were observed in 100% and 97% of mosquitoes from the peri-domestic and forest population, respectively but were associated with pyrethroid resistance only in the forest population (P < 0.0001). For the first time in Aedes mosquitoes in Ghana, we report the existence V410L mutations, mostly under selection only in the forest population (HWE P < 0.0001) and conclude that Aedes vectors in urban Accra have developed resistance to many commonly used insecticides. This information is important for the formulation of vector control strategies for Aedes control in Ghana.

Institutionalising maternal and newborn quality-of-care standards in Bangladesh, Ghana and Tanzania: a quasi-experimental study.

INTRODUCTION: Facility interventions to improve quality of care around childbirth are known but need to be packaged, tested and institutionalised within health systems to impact on maternal and newborn outcomes. METHODS: We conducted cross-sectional assessments at baseline (2016) and after 18 months of provider-led implementation of UNICEF/WHO's Every Mother Every Newborn Quality Improvement (EMEN-QI) standards (preceding the WHO Standards for improving quality of maternal and newborn care in health facilities). 19 hospitals and health centres (2.8M catchment population) in Bangladesh, Ghana and Tanzania were involved and 24 from adjoining districts served for 'comparison'. We interviewed 43 facility managers and 818 providers, observed 1516 client-provider interactions, reviewed 12 020 records and exit-interviewed 1826 newly delivered women. We computed a 39-criteria institutionalisation score combining clinical, patient rights and cross-cutting domains from EMEN-QI and used routine/District Health Information System V.2 data to assess the impact on perinatal and maternal mortality. RESULTS: EMEN-QI standards institutionalisation score increased from 61% to 80% during EMEN-QI implementation, exceeding 75% target. All mortality indicators showed a downward trajectory though not all reached statistical significance. Newborn case-fatality rate fell significantly by 25% in Bangladesh (RR=0·75 (95% CI=0·59 to 0·96), p=0·017) and 85% in Tanzania (RR=0.15 (95% CI=0.08 to 0.29), p<0.001), but not in Ghana. Similarly, stillbirth (RR=0.64 (95% CI=0.45 to 0.92), p<0.01) and perinatal mortality in Tanzania reduced significantly (RR=0.59 (95% CI=0.40 to 0.87), p=0.007). Institutional maternal mortality ratios generally reduced but were only significant in Ghana: 362/100 000 to 207/100 000 livebirths (RR=0.57 (95% CI=0.33 to 0.99), p=0.046). Routine mortality data from comparison facilities were limited and scarce. Systematic death audits and clinical mentorship drove these achievements but challenges still remain around human resource management and equipment maintenance systems. CONCLUSION: Institutionalisation of the UNICEF/WHO EMEN-QI standards as a package is feasible within existing health systems and may reduce mortality around childbirth. Critical gaps around sustainability must be fundamental considerations for scale-up.

Sero-prevalence and risk factors for Severe Acute Respiratory Syndrome Coronavirus 2 infection in women and children in a rural district of Bangladesh: A cohort study.

Background: Bangladesh reported its first COVID-19 case on March 8, 2020. Despite lockdowns and promoting behavioural interventions, as of December 31, 2021, Bangladesh reported 1.5 million confirmed cases and 27 904 COVID-19-related deaths. To understand the course of the pandemic and identify risk factors for SARs-Cov-2 infection, we conducted a cohort study from November 2020 to December 2021 in rural Bangladesh. Methods: After obtaining informed consent and collecting baseline data on COVID-19 knowledge, comorbidities, socioeconomic status, and lifestyle, we collected data on COVID-like illness and care-seeking weekly for 54 weeks for women (n = 2683) and their children (n = 2433). Between March and July 2021, we tested all participants for SARS-CoV-2 antibodies using ROCHE's Elecsys® test kit. We calculated seropositivity rates and 95% confidence intervals (95% CI) separately for women and children. In addition, we calculated unadjusted and adjusted relative risk (RR) and 95% CI of seropositivity for different age and risk groups using log-binomial regression models. Results: Overall, about one-third of women (35.8%, 95% CI = 33.7-37.9) and one-fifth of children (21.3%, 95% CI = 19.2-23.6) were seropositive for SARS-CoV-2 antibodies. The seroprevalence rate doubled for women and tripled for children between March 2021 and July 2021. Compared to women and children with the highest household wealth (HHW) tertile, both women and children from poorer households had a lower risk of infection (RR, 95% CI for lowest HHW tertile women (0.83 (0.71-0.97)) and children (0.75 (0.57-0.98)). Most infections were asymptomatic or mild. In addition, the risk of infection among women was higher if she reported chewing tobacco (RR = 1.19,95% CI = 1.03-1.38) and if her husband had an occupation requiring him to work indoors (RR = 1.16, 95% CI = 1.02-1.32). The risk of infection was higher among children if paternal education was >5 years (RR = 1.37, 95% CI = 1.10-1.71) than in children with a paternal education of ≤5 years. Conclusions: We provided prospectively collected population-based data, which could contribute to designing feasible strategies against COVID-19 tailored to high-risk groups. The most feasible strategy may be promoting preventive care practices; however, collecting data on reported practices is inadequate. More in-depth understanding of the factors related to adoption and adherence to the practices is essential.

Prediction of gestational age using urinary metabolites in term and preterm pregnancies.

Assessment of gestational age (GA) is key to provide optimal care during pregnancy. However, its accurate determination remains challenging in low- and middle-income countries, where access to obstetric ultrasound is limited. Hence, there is an urgent need to develop clinical approaches that allow accurate and inexpensive estimations of GA. We investigated the ability of urinary metabolites to predict GA at time of collection in a diverse multi-site cohort of healthy and pathological pregnancies (n = 99) using a broad-spectrum liquid chromatography coupled with mass spectrometry (LC-MS) platform. Our approach detected a myriad of steroid hormones and their derivatives including estrogens, progesterones, corticosteroids, and androgens which were associated with pregnancy progression. We developed a restricted model that predicted GA with high accuracy using three metabolites (rho = 0.87, RMSE = 1.58 weeks) that was validated in an independent cohort (n = 20). The predictions were more robust in pregnancies that went to term in comparison to pregnancies that ended prematurely. Overall, we demonstrated the feasibility of implementing urine metabolomics analysis in large-scale multi-site studies and report a predictive model of GA with a potential clinical value.

Machine learning prediction of gestational age from metabolic screening markers resistant to ambient temperature transportation: Facilitating use of this technology in low resource settings of South Asia and East Africa.

Background: Knowledge of gestational age is critical for guiding preterm neonatal care. In the last decade, metabolic gestational dating approaches emerged in response to a global health need; because in most of the developing world, accurate antenatal gestational age estimates are not feasible. These methods initially developed in North America have now been externally validated in two studies in developing countries, however, require shipment of samples at sub-zero temperature. Methods: A subset of 330 pairs of heel prick dried blood spot samples were shipped on dry ice and in ambient temperature from field sites in Tanzania, Bangladesh and Pakistan to laboratory in Iowa (USA). We evaluated impact on recovery of analytes of shipment temperature, developed and evaluated models for predicting gestational age using a limited set of metabolic screening analytes after excluding 17 analytes that were impacted by shipment conditions of a total of 44 analytes. Results: With the machine learning model using all the analytes, samples shipped in dry ice yielded a Root Mean Square Error (RMSE) of 1.19 weeks compared to 1.58 weeks for samples shipped in ambient temperature. Out of the 44 screening analytes, recovery of 17 analytes was significantly different between the two shipment methods and these were excluded from further machine learning model development. The final model, restricted to stable analytes provided a RMSE of 1.24 (95% confidence interval (CI) = 1.10-1.37) weeks for samples shipped on dry ice and RMSE of 1.28 (95% CI = 1.15-1.39) for samples shipped at ambient temperature. Analysis for discriminating preterm births (gestational age <37 weeks), yielded an area under curve (AUC) of 0.76 (95% CI = 0.71-0.81) for samples shipped on dry ice and AUC of 0.73 (95% CI = 0.67-0.78) for samples shipped in ambient temperature. Conclusions: In this study, we demonstrate that machine learning algorithms developed using a sub-set of newborn screening analytes which are not sensitive to shipment at ambient temperature, can accurately provide estimates of gestational age comparable to those from published regression models from North America using all analytes. If validated in larger samples especially with more newborns <34 weeks, this technology could substantially facilitate implementation in LMICs.

Socio-economic determinants of maternal health care utilization in Kailahun District, Sierra Leone, 2020.

BACKGROUND: Ascertaining the key determinants of maternal healthcare service utilization and their relative importance is critical to priority setting in policy development. Sierra Leone has one of the world's highest maternal death ratios in the context of a weak health system. The objectives of this study were to determine; the level of utilization of Antenatal Care (ANC), Skilled Delivery Attendants (SDA), Postnatal Care (PNC) services, and factors that influence the utilization of these services. METHODS: We conducted a community-based cross-sectional study involving 554 women of reproductive age (15-49 years) who had at least one delivery in the last 3 years and lived in the Kailahun District, Sierra Leone from November 2019 to October 2020. Data were analysed using analysed using bivariate, multivariate and multinomial logistic regression models. RESULTS: The median age of respondents was 25 years (Q1 = 17 years, Q3 = 30 years). Eighty-nine percent (89%) had 4 or more ANC visits. Only 35.9% of women were delivered by SDA. Women residing in urban areas had over six-fold increased odds of utilizing SDA as compared to women residing in rural areas (AOR = 6.20, 95% CI = 3.61-10.63). Women whose husbands had a primary level of education had 2.38 times increased odds of utilizing SDA than women whose husbands had no education (AOR = 2.38, 95% CI = 1.30-4.35). Women that walked longer distances (30-60 min) to seek healthcare had 2.98 times increased odds of utilizing SBA than those that walked shorter distances (< 30 min) (AOR = 2.98, 95% CI = 1.67-5.33). Women who had a secondary/vocational level of education had 2.35 times increased odds of utilizing the standard PNC category as compared to those with no education (OR = 2.35, 95% CI = 1.19-4.63). CONCLUSION: The majority of women had 4 or more ANC visits yet the use of skilled birth attendants was low. Urban residence and education were significantly associated with the use of the standard PNC category. To improve the utilization of maternal health care services, national healthcare policies should target the advancement of education, train skilled Maternal Healthcare (MHC) attendants, rural infrastructure, and the empowerment of women.

Implementing effective community-based surveillance in research studies of maternal, newborn and infant outcomes in low resource settings.

BACKGROUND: Globally adopted health and development milestones have not only encouraged improvements in the health and wellbeing of women and infants worldwide, but also a better understanding of the epidemiology of key outcomes and the development of effective interventions in these vulnerable groups. Monitoring of maternal and child health outcomes for milestone tracking requires the collection of good quality data over the long term, which can be particularly challenging in poorly-resourced settings. Despite the wealth of general advice on conducting field trials, there is a lack of specific guidance on designing and implementing studies on mothers and infants. Additional considerations are required when establishing surveillance systems to capture real-time information at scale on pregnancies, pregnancy outcomes, and maternal and infant health outcomes. MAIN BODY: Based on two decades of collaborative research experience between the Kintampo Health Research Centre in Ghana and the London School of Hygiene and Tropical Medicine, we propose a checklist of key items to consider when designing and implementing systems for pregnancy surveillance and the identification and classification of maternal and infant outcomes in research studies. These are summarised under four key headings: understanding your population; planning data collection cycles; enhancing routine surveillance with additional data collection methods; and designing data collection and management systems that are adaptable in real-time. CONCLUSION: High-quality population-based research studies in low resource communities are essential to ensure continued improvement in health metrics and a reduction in inequalities in maternal and infant outcomes. We hope that the lessons learnt described in this paper will help researchers when planning and implementing their studies.

Machine learning guided postnatal gestational age assessment using new-born screening metabolomic data in South Asia and sub-Saharan Africa.

BACKGROUND: Babies born early and/or small for gestational age in Low and Middle-income countries (LMICs) contribute substantially to global neonatal and infant mortality. Tracking this metric is critical at a population level for informed policy, advocacy, resources allocation and program evaluation and at an individual level for targeted care. Early prenatal ultrasound examination is not available in these settings, gestational age (GA) is estimated using new-born assessment, last menstrual period (LMP) recalls and birth weight, which are unreliable. Algorithms in developed settings, using metabolic screen data, provided GA estimates within 1-2 weeks of ultrasonography-based GA. We sought to leverage machine learning algorithms to improve accuracy and applicability of this approach to LMICs settings. METHODS: This study uses data from AMANHI-ACT, a prospective pregnancy cohorts in Asia and Africa where early pregnancy ultrasonography estimated GA and birth weight are available and metabolite screening data in a subset of 1318 new-borns were also available. We utilized this opportunity to develop machine learning (ML) algorithms. Random Forest Regressor was used where data was randomly split into model-building and model-testing dataset. Mean absolute error (MAE) and root mean square error (RMSE) were used to evaluate performance. Bootstrap procedures were used to estimate confidence intervals (CI) for RMSE and MAE. For pre-term birth identification ROC analysis with bootstrap and exact estimation of CI for area under curve (AUC) were performed. RESULTS: Overall model estimated GA had MAE of 5.2 days (95% CI 4.6-6.8), which was similar to performance in SGA, MAE 5.3 days (95% CI 4.6-6.2). GA was correctly estimated to within 1 week for 85.21% (95% CI 72.31-94.65). For preterm birth classification, AUC in ROC analysis was 98.1% (95% CI 96.0-99.0; p < 0.001). This model performed better than Iowa regression, AUC Difference 14.4% (95% CI 5-23.7; p = 0.002). CONCLUSIONS: Machine learning algorithms and models applied to metabolomic gestational age dating offer a ladder of opportunity for providing accurate population-level gestational age estimates in LMICs settings. These findings also point to an opportunity for investigation of region-specific models, more focused feasible analyte models, and broad untargeted metabolome investigation.

Using AMANHI-ACT cohorts for external validation of Iowa new-born metabolic profiles based models for postnatal gestational age estimation.

BACKGROUND: Globally, 15 million infants are born preterm and another 23.2 million infants are born small for gestational age (SGA). Determining burden of preterm and SGA births, is essential for effective planning, modification of health policies and targeting interventions for reducing these outcomes for which accurate estimation of gestational age (GA) is crucial. Early pregnancy ultrasound measurements, last menstrual period and post-natal neonatal examinations have proven to be not feasible or inaccurate. Proposed algorithms for GA estimation in western populations, based on routine new-born screening, though promising, lack validation in developing country settings. We evaluated the hypothesis that models developed in USA, also predicted GA in cohorts of South Asia (575) and Sub-Saharan Africa (736) with same precision. METHODS: Dried heel prick blood spots collected 24-72 hours after birth from 1311 new-borns, were analysed for standard metabolic screen. Regression algorithm based, GA estimates were computed from metabolic data and compared to first trimester ultrasound validated, GA estimates (gold standard). RESULTS: Overall Algorithm (metabolites + birthweight) estimated GA to within an average deviation of 1.5 weeks. The estimated GA was within the gold standard estimate by 1 and 2 weeks for 70.5% and 90.1% new-borns respectively. Inclusion of birthweight in the metabolites model improved discriminatory ability of this method, and showed promise in identifying preterm births. Receiver operating characteristic (ROC) curve analysis estimated an area under curve of 0.86 (conservative bootstrap 95% confidence interval (CI) = 0.83 to 0.89); P < 0.001) and Youden Index of 0.58 (95% CI = 0.51 to 0.64) with a corresponding sensitivity of 80.7% and specificity of 77.6%. CONCLUSION: Metabolic gestational age dating offers a novel means for accurate population-level gestational age estimates in LMIC settings and help preterm birth surveillance initiatives. Further research should focus on use of machine learning and newer analytic methods broader than conventional metabolic screen analytes, enabling incorporation of region-specific analytes and cord blood metabolic profiles models predicting gestational age accurately.