Search

Background: Hypertensive disorders of pregnancy (HDP) are significant drivers of maternal and neonatal morbidity and mortality. Current management strategies include early identification and initiation of risk mitigating interventions facilitated by a rules-based checklist. Advanced analytic techniques, such as machine learning, can potentially offer improved and refined predictive capabilities. Objective: To develop and internally validate a machine learning prediction model for hypertensive disorders of pregnancy (HDP) when initiating prenatal care. Study Design: We developed a prediction model using data from the prospective multisite cohort Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be (nuMoM2b) among low-risk individuals without a prior history of aspirin utilization for preeclampsia prevention. The primary outcome was the development of HDP. Random forest modeling was utilized to develop predictive models. Recursive feature elimination (RFE) was employed to create a reduced model for each outcome. Area under the curve (AUC), 95% confidence intervals (CI), and calibration curves were utilized to assess discrimination and accuracy. Sensitivity analyses were conducted to compare the sensitivity and specificity of the reduced model compared to existing risk factor-based algorithms. Results: Of 9,124 assessed low risk nulliparous individuals, 21% (n=1,927) developed HDP. The prediction model for HDP had satisfactory discrimination with an AUC of 0.73 (95% CI: 0.70, 0.75). After RFE, a parsimonious reduced model with 30 features was created with an AUC of 0.71 (95% CI: 0.68, 0.74). Variables included in the model after RFE included body mass index at the first study visit, pre-pregnancy weight, first trimester complete blood count results, and maximum systolic blood pressure at the first visit. Calibration curves for all models revealed relatively stable agreement between predicted and observed probabilities. Sensitivity analysis noted superior sensitivity (AUC 0.80 vs 0.65) and specificity (0.65 vs 0.53) of the model compared to traditional risk factor-based algorithms. Conclusion: In cohort of low-risk nulliparous pregnant individuals, a prediction model may accurately predict HDP diagnosis at the time of initiating prenatal care and aid employment of close interval monitoring and prophylactic measures earlier in pregnancy.

Abstract Background Although community health worker (CHW) programs focus on improving access to healthcare, some individuals may not receive the intended quality or quantity of an intervention. The objective of this research was to examine if certain populations of pregnant women differentially experience the implementation of a community health worker-led maternal health intervention in Zanzibar. Methods We included pregnant women enrolled in the Safer Deliveries (Uzazi Salama) program, which operated in 10 of 11 districts in Zanzibar, Tanzania between January 1, 2017, and June 19, 2019 (N = 33,914). The outcomes of interest were receipt of the entire postpartum intervention (three CHW visits) and time to first postpartum CHW visit (days). Visits by CHWs were done at the women’s home, however, a telehealth option existed for women who were unable to be reached in-person. We conducted statistical tests to investigate the bivariate associations between our outcomes and each demographic and health characteristic. We used multivariate logistic regression to estimate the relationships between covariates and the outcomes and multivariate linear regression to estimate the association between covariates and the average time until first postpartum visit. Results Higher parity (OR = 0.85; P = 0.014; 95%CI: 0.75–0.97), unknown or unreported HIV status (OR = 0.64; p

AbstractUser engagement with remote blood pressure monitoring during pregnancy is critical to optimize the associated benefits of blood pressure control and early detection of hypertensive disorders of pregnancy. In our study population of pregnant individuals, we found that connectedunconnected

Background The COVID-19 pandemic has disrupted health services worldwide, which may have led to increased mortality and secondary disease outbreaks. Disruptions vary by patient population, geographic area, and service. While many reasons have been put forward to explain disruptions, few studies have empirically investigated their causes. Objective We quantify disruptions to outpatient services, facility-based deliveries, and family planning in seven low- and middle-income countries during the COVID-19 pandemic and quantify relationships between disruptions and the intensity of national pandemic responses. Methods We leveraged routine data from 104 Partners In Health-supported facilities from January 2016 to December 2021. We first quantified COVID-19-related disruptions in each country by month using negative binomial time series models. We then modelled the relationship between disruptions and the intensity of national pandemic responses, as measured by the stringency index from the Oxford COVID-19 Government Response Tracker. Results For all the studied countries, we observed at least one month with a significant decline in outpatient visits during the COVID-19 pandemic. We also observed significant cumulative drops in outpatient visits across all months in Lesotho, Liberia, Malawi, Rwanda, and Sierra Leone. A significant cumulative decrease in facility-based deliveries was observed in Haiti, Lesotho, Mexico, and Sierra Leone. No country had significant cumulative drops in family planning visits. For a 10-unit increase in the average monthly stringency index, the proportion deviation in monthly facility outpatient visits compared to expected fell by 3.9% (95% CI: −5.1%, −1.6%). No relationship between stringency of pandemic responses and utilisation was observed for facility-based deliveries or family planning. Conclusions Context-specific strategies show the ability of health systems to sustain essential health services during the pandemic. The link between pandemic responses and healthcare utilisation can inform purposeful strategies to ensure communities have access to care and provide lessons for promoting the utilisation of health services elsewhere.

BackgroundWhile high-quality primary health care services can meet 80%-90% of health needs over a person’s lifetime, this potential is severely hindered in many low-resource countries by a constrained health care system. There is a growing consensus that effectively designed, resourced, and managed community health worker programs are a critical component of a well-functioning primary health system, and digital technology is recognized as an important enabler of health systems transformation. ObjectiveIn this implementation report, we describe the design and rollout of Zanzibar’s national, digitally enabled community health program–Jamii ni Afya. MethodsSince 2010, D-tree International has partnered with the Ministry of Health Zanzibar to pilot and generate evidence for a digitally enabled community health program, which was formally adopted and scaled nationally by the government in 2018. Community health workers use a mobile app that guides service delivery and data collection for home-based health services, resulting in comprehensive service delivery, access to real-time data, efficient management of resources, and continuous quality improvement. ResultsThe Zanzibar government has documented increases in the delivery of health facilities among pregnant women and reductions in stunting among children younger than 5 years since the community health program has scaled. Key success factors included starting with the health challenge and local context rather than the technology, usage of data for decision-making, and extensive collaboration with local and global partners and funders. Lessons learned include the significant time it takes to scale and institutionalize a digital health systems innovation due to the time to generate evidence, change opinions, and build capacity. ConclusionsJamii ni Afya represents one of the world’s first examples of a nationally scaled digitally enabled community health program. This implementation report outlines key successes and lessons learned, which may have applicability to other governments and partners working to sustainably strengthen primary health systems.

Abstract Background Maternal and neonatal mortality remain elevated in low and middle income countries, and progress is slower than needed to achieve the Sustainable Development Goals. Existing strategies appear to be insufficient. One proposed alternative strategy, Service Delivery Redesign for Maternal and Neonatal Health (SDR), centers on strengthening higher level health facilities to provide rapid, definitive care in case of delivery and post-natal complications, and then promoting delivery in these hospitals, rather than in primary care facilities. However to date, SDR has not been piloted or evaluated. Methods We will use a prospective, non-randomized stepped-wedge design to evaluate the effectiveness and implementation of Service Delivery Redesign for Maternal and Neonatal Health in Kakamega County, Kenya. Discussion This protocol describes a hybrid effectiveness/implementation evaluation study with an adaptive design. The impact evaluation (“effectiveness”) study focuses on maternal and newborn health outcomes, and will be accompanied by an implementation evaluation focused on program reach, adoption, and fidelity.

Abstract Background Many children in low- and middle-income countries fail to reach their cognitive potential, with experiences before age 3 critical in shaping long-term development. Zanzibar’s Jamii ni Afya program is the first national, digitally enabled community health volunteer (CHV) program promoting early childhood development (ECD) following the Nurturing Care Framework within an integrated maternal and child healthcare package. Using program baseline data, we explored home environment, caregivers’ parenting, health and nutrition knowledge and practices, and ECD outcomes in Zanzibar. Methods We conducted a national household survey among 499 children aged 18-29 months using two-stage cluster sampling in February 2019. The primary outcome was child development score measured using the Caregiver Reported Early Developmental Index (CREDI), with higher scores representing higher levels of child development. We analyzed CREDI scores, along with MICS questions on parenting knowledge, practices, and characteristics of the home environment. We developed multivariate regression models to assess associations between caregiver-child interactions, knowledge of dietary diversity, and ECD. Results Ten percent of children had overall CREDI z-scores 2 standard deviations [SD] or more below the global reference population mean, with 28% of children at risk of developmental delay with z-scores 1 SD or more below the mean. Cognitive and language domains were of highest concern (10.2 and 12.7% with z-score

Introduction The COVID-19 pandemic has caused disruptions in access to routine healthcare services worldwide, with a particularly high impact on chronic care patients and low and middle-income countries. In this study, we used routinely collected electronic medical records data to assess the impact of the COVID-19 pandemic on access to cancer care at the Butaro Cancer Center of Excellence (BCCOE) in rural Rwanda.Methods We conducted a retrospective time-series study among all Rwandan patients who received cancer care at the BCCOE between 1 January 2016 and 31 July 2021. The primary outcomes of interest included a comparison of the number of patients who were predicted based on time-series models of pre-COVID-19 trends versus the actual number of patients who presented during the COVID-19 period (between March 2020 and July 2021) across four key indicators: the number of new patients, number of scheduled appointments, number of clinical visits attended and the proportion of scheduled appointments completed on time.Results In total, 8970 patients (7140 patients enrolled before COVID-19 and 1830 patients enrolled during COVID-19) were included in this study. During the COVID-19 period, enrolment of new patients dropped by 21.7% (95% prediction interval (PI): −31.3%, −11.7%) compared with the pre-COVID-19 period. Similarly, the number of clinical visits was 25.0% (95% PI: −31.1%, −19.1%) lower than expected and the proportion of scheduled visits completed on time was 27.9% (95% PI: −39.8%, −14.1%) lower than expected. However, the number of scheduled visits did not deviate significantly from expected.Conclusion Although scheduling procedures for visits continued as expected, our findings reveal that the COVID-19 pandemic interrupted patients’ ability to access cancer care and attend scheduled appointments at the BCCOE. This interruption in care suggests delayed diagnosis and loss to follow-up, potentially resulting in a higher rate of negative health outcomes among cancer patients in Rwanda.

Background: Surgical scrubbing, gowning, and gloving is challenging for medical trainees to learn in the operating room environment. Currently, there are few reliable or valid tools to evaluate a trainee’s ability to scrub, gown and glove. The objective of this study is to test the reliability and validity of a checklist that evaluates the technique of surgical scrubbing, gowning and gloving (SGG). Methods: This Institutional Review Board-approved study recruited medical students, residents, and fellows from an academic, tertiary care institution. Trainees were stratified based upon prior surgical experience as novices, intermediates, or experts. Participants were instructed to scrub, gown and glove in a staged operating room while being video-recorded. Two blinded raters scored the videos according to the SGG checklist. Reliability was assessed using the intraclass correlation coefficient for total scores and Cohen’s kappa for item completion. The internal consistency and discriminant validity of the SGG checklist were assessed using Cronbach alpha and the Wilcoxon rank sum test, respectively. Results: 56 participants were recruited (18 novices, 19 intermediates, 19 experts). The intraclass correlation coefficient demonstrated excellent inter-rater reliability for the overall checklist (0.990), and the Cohen’s kappa ranged from 0.598 to 1.00. The checklist also had excellent internal consistency (Cronbach’s alpha 0.950). A significant difference in scores was observed between all groups (p < 0.001). Conclusion: This checklist demonstrates a high inter-rater reliability, discriminant validity, and internal consistency. It has the potential to enhance medical education curricula.

Background: In the context of diminutive COVID-19 screening and testing, syndromic surveillance can be used to identify areas with higher-than-expected SARS-CoV-2 symptoms for targeted public health interventions. We used syndromic surveillance to monitor potential SARS-CoV-2 outbreaks in 14 health facilities in the Neno district of rural Malawi. Methods: We monitored three indicators identified as potential symptoms of SARS-CoV-2 infection: the proportion of outpatient visits for fast-breathing cases in children under 5 years (FBC

BackgroundOur objective was to assess differences in TB treatment outcomes between individuals who were HIV negative, HIV positive on anti-retroviral treatment (ART) and HIV positive not on ART, at TB treatment initiation at a rural district hospital in Eastern Cape, South Africa.MethodsThis was a retrospective cohort study of individuals diagnosed with TB between January 2017 and April 2020 at a district hospital. Adults 15 years and over with reported HIV status and treatment outcome were included (N = 711). A categorical outcome with three levels was considered: unfavorable, down referral, and success. We report descriptive statistics for the association between HIV and ART status and treatment outcome using Chi-square and Fisher's exact tests. A multinomial baseline logit model was used to estimate odds ratios for treatment outcomes.ResultsOverall, 59% of included patients were HIV positive with 75% on ART. Eighty-eight patients 12% had an unfavorable outcome. Half of all patients were down referred with an additional 37% having a successful outcome. Individuals without HIV were more likely to be down referred (versus unfavorable) compared to individuals with untreated HIV (2.90 OR, 1.36, 6.17 95% CI). There was a greater likelihood for individuals without HIV having a successful TB treatment outcome compared to individuals with untreated HIV (4.98 OR, 2.07, 11.25 95% CI).ConclusionThe majority of individuals had positive TB treatment outcomes (down referred or success). However, people without HIV had nearly five times greater odds of having successful outcomes than those with untreated HIV.

Abstract Background Most maternal health programs in low- and middle- income countries estimate gestational age to provide appropriate antenatal care at the correct times throughout the pregnancy. Although various gestational dating methods have been validated in research studies, the performance of these methods has not been evaluated on a larger scale, such as within health systems. The objective of this research was to investigate the magnitude and impact of errors in estimated delivery dates on health facility delivery among women enrolled in a maternal health program in Zanzibar. Methods This study included 4225 women who were enrolled in the Safer Deliveries program and delivered before May 31, 2017. The exposure of interest was error in estimated delivery date categorized as: severe overestimate, when estimated delivery date (EDD) was 36 days or more after the actual delivery date (ADD); moderate overestimate, when EDD was 15 to 35 days after ADD; accurate, when EDD was 6 days before to 14 days after ADD; and underestimate, when EDD was 7 days or more before ADD. We used Chi-squared tests to identify factors associated with errors in estimated delivery dates. We performed logistic regression to assess the impact of errors in estimated delivery dates on health facility delivery adjusting for age, district of residence, HIV status, and occurrence of past home delivery. Results In our data, 28% of the estimated delivery dates were a severe overestimate, 23% moderate overestimate, 41% accurate, and 8% underestimate. Compared to women with an accurate delivery date, women with a moderate or severe overestimate were significantly less likely to deliver in a health facility (OR = 0.71, 95% CI: [0.59, 0.86]; OR = 0.74, 95% CI: [0.61, 0.91]). When adjusting for multiple confounders, women with moderate overestimates were significantly less likely to deliver in a health facility (AOR = 0.76, 95% CI: [0.61, 0.93]); the result moved slightly towards null for women with severe overestimates (AOR = 0.84, 95% CI: [0.69, 1.03]). Conclusions The overestimation of women’s EDDs reduces the likelihood of health facility delivery. To address this, maternal health programs should improve estimation of EDD or attempt to curb the effect of these errors within their programs.

Vaccines are the most effective mechanism for ending the COVID-19 pandemic. However, reluctance to accept vaccines has hindered the efforts of health authorities to combat the virus. In Haiti, as of July 2021, less than 1% of the country's population has been fully vaccinated in part due to vaccine hesitancy. Our goal was to assess Haitian attitudes toward COVID-19 vaccination and investigate the primary reasons for Moderna vaccine hesitancy. We conducted a cross-sectional survey across three rural Haitian communities, in September 2021. The research team used electronic tablets to collect quantitative data from 1,071 respondents, selected randomly across the communities. We report descriptive statistics and identify variables associated with vaccine acceptance using logistic regression built using a backward stepwise approach. Among 1,071 respondents, the overall acceptance rate was 27.0% (n = 285). The most common reason for vaccine hesitancy was "concern about side effects" (n = 484, 67.1%) followed by "concern about contracting COVID-19 from the vaccine" (n = 472, 65.4%). Three-quarters of respondents (n = 817) identified their healthcare workers as their most trustworthy source for information related to the vaccine. In the bivariate analysis, male gender ( p  = .06) and no history of drinking alcohol ( p  < .001) were significantly associated with being more likely to take the vaccine. In the final reduced model, only those with a history of drinking alcohol were significantly more likely to take the vaccine (aOR = 1.47 (1.23, 1.87) p  < .001). The acceptance rate for the COVID-19 vaccine is low, and public health experts should design and strengthen vaccination campaigns to combat misinformation and public distrust.