Predicting the drop out from the maternal, newborn and child healthcare continuum in three East African Community countries: application of machine learning models

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.

Development of Continuum of Care for McArdle disease: A practical tool for clinicians and patients

McArdle disease (glycogen storage disease type V; GSDV) is a rare genetic disease caused by the inability to break down glycogen in skeletal muscle due to a deficiency in myophosphorylase. Glycolysis is only partially blocked in GSDV, as muscle fibres can take up circulating glucose and convert it to glucose-6-phosphate downstream of the metabolic block. Because skeletal muscle predominantly relies on anaerobic energy during the first few minutes of transition from rest to activity, and throughout more intense activities, individuals with GSDV experience muscle fatigue/pain, tachypnea, and tachycardia during these activities. If warning signs are not heeded, a muscle contracture may rapidly occur, and if significant, may lead to acute rhabdomyolysis. Without a cure or treatment, individuals with GSDV must be consistent in employing proper management techniques; however, this can be challenging due to the nuances inherent in this metabolic myopathy. The International Association for Muscle Glycogen Storage Disease collaborated with an international team of five expert clinicians to identify areas of learning to achieve an optimal state. A Continuum of Care model was developed that outlines five pivotal steps (diagnosis; understanding; acceptance; learning and exercise) to streamline assessments and more succinctly assist clinicians in determining patient-specific learning needs. This model serves as a translational tool to help optimize care for this patient population.

Impact of COVID-19 pandemic in the care of people living with HIV and cancer at an oncologic center in Mexico

BACKGROUND

People living with HIV(PLWH) and cancer are among the most vulnerable patients and require constant access to medical services. We compared the characteristics of PLWH and cancer in Mexico, before and during the COVID-19 pandemic.

METHODS

Patients admitted 1 year before (pre-pandemic) and 1 year after the start of the pandemic (pandemic) were included. Clinical characteristics, HIV-related variables, and 90-day mortality were compared. Data are described a proportions (N,%) and central tendency measures. A multiple regression model for variables associated with 90-day mortality was performed.

RESULTS

Seventy-nine patients were seen in the pre-pandemic period; 92 during the pandemic. Main diagnoses were Kaposi Sarcoma and lymphoma. CD4⁺ cell count at diagnosis was lower during the pandemic: 81 cells/mm³ vs. 128 cells/mm³, p = .035. CD4+<100 cells/mm³ at first consultation increased from 41% to 58% during the pandemic (p = .041). Only BMI <20 kg/m² was associated to death (aOR 8.27, 95%CI 1.74-39.25) (p = .008). The pandemic period was not associated with a higher 90-day mortality.

CONCLUSIONS

PLWH and cancer presented to care with advanced disease overall. This was more pronounced during the pandemic period. Mortality was associated with AIDS-related variables regardless of study period. This underscores the need for strategies to maintain in-person access to health-care services for PLWH.

Homelessness, Politics, and Policy: Predicting Spatial Variation in COVID-19 Cases and Deaths

When COVID-19 began to spread in the United States, the first public health orders were to hunker down at home. But for the vulnerable people experiencing homelessness, especially those sleeping outdoors, retreating to a private dwelling was not possible. This suggests that places with greater homelessness would also have elevated COVID-19 infections. This paper examines how spatial variation in unsheltered homelessness was related to the cumulative number of cases and deaths from COVID-19. Although Continuums of Care (CoCs) with more households receiving welfare, without internet service, and more disabled residents had a higher rate of COVID-19-related cases and deaths, CoCs with more unsheltered homelessness had fewer COVID-19-related deaths. More research is needed to explain this counterintuitive result, but it may reflect the bicoastal pattern of homelessness which is higher where government intervention, community sentiment, and compliance with rules to promote the common welfare are greater. In fact, local politics and policies mattered. CoCs with more volunteering and a higher share of votes for the 2020 Democratic presidential candidate also had fewer COVID-19 cases and deaths. Yet, other policies did not matter. Having more homeless shelter beds, publicly assisted housing units, residents in group quarters, or greater use of public transportation had no independent associations with pandemic outcomes.

Continuum of care for maternal, newborn, and child health in 17 sub-Saharan African countries

BACKGROUND

Given the intricate relationship between mothers and their children with the continuum of care, it is quintessential for their healthcare and interventions to be enhanced through a continuum of care approach. We examined the factors associated with the continuum of care for maternal, newborn, and child health in sub-Saharan Africa.

METHODS

Data for the study were pooled from the Demographic and Health Surveys of 17 countries in sub-Saharan Africa. We extracted the data from the women's files in all 17 countries. We included 15,359 married and cohabiting women with the most recent children aged 12-23 months at the time of the survey in the study. Multivariable multilevel logistic regression analysis was performed to examine the factors associated with continuum of care. Results were presented as adjusted odds ratios (aORs) with their 95% confidence interval.

RESULTS

Women whose partners had secondary or higher level of education [aOR = 1.52; 95%CI = 1.07, 2.16], those with health insurance [aOR = 1.64; 95%CI = 1.18, 2.30], those who decide alone on their healthcare [aOR = 1.38; 95%CI = 1.01, 1.89], those with joint healthcare decision [aOR = 1.33; 95%CI = 1.02, 1.74], those exposed to radio [aOR = 1.38; 95%CI = 1.06, 1.79], those who started antenatal care early [aOR = 1.88; 95%CI = 1.50, 2.36] and those in southern Africa [aOR = 7.02; 95%CI = 3.23, 15.27] had higher odds of completing the continuum of care.

CONCLUSION

We found that only 3.4% of women across the 17 sub-Saharan African countries included in this study completed all of the 11 maternal, newborn, and child health care interventions along the continuum of care. The factors that are associated with the maternal, newborn, and child health continuum of care include women's autonomy in decision-making, partners' level of education, health insurance coverage, early initiation of antenatal care, exposure to radio, and residing in Southern Africa. Problem with the distance to the facility lowered women's odds of completing the continuum of care. Governments and departments of health services across sub-Saharan African countries must leverage the radio to disseminate critical messages about the need to complete the continuum of care. Much commitment must be made toward advancing the autonomy of women. Health insurance coverage would have to be extended to reach all women to facilitate the completion of the continuum of care.

Risk factors of dropout from institutional delivery among HIV positive antenatal care booked mothers within one year postpartum in Ethiopia: a case-control study

Background

The uptake of maternal healthcare services remains suboptimal in Ethiopia. Significant proportions of antenatal care attendees give birth at home in the context of HIV. However, in Ethiopia, evidence is scarce on the predictors of dropout from maternity continuum of care among HIV-positive mothers. Therefore, this study aimed to supply valuable information on risk factors regarding dropout of HIV-positive mothers for institutional delivery services in northwest Ethiopia.

Methods

A multicenter case–control study was conducted at governmental health facilities in Gondar City from May one to June 30/2018. A total of 222 HIV-positive women were included in the study. Data were collected using structured questionnaires and checklists through face-to-face interview and chart review; entered into EPI INFO version seven, and then exported to SPSS version 25. Both descriptive and analytical procedures were performed. Binary logistic regression analysis was undertaken. A significant association was declared based on the adjusted odds ratio (AOR) with its 95% CI and p-value of ≤ 0.05.

Results

This study illustrates that maternal age of ≥ 35 years (AOR = 2.37; 95%CI: 1.13,5.13), unmarried marital relation (AOR = 3.28; 95%CI: 1.51, 7.13), unemployed spousal occupation (AOR = 3.91; 95%CI: 1.54, 9.91), family monthly income of ≤ 36 US dollar (AOR = 4.87; 95%CI: 2.08, 11.42) and no obstetric complication in the index pregnancy (AOR = 13.89; 95%CI: 2.73, 27.71) were positively associated with dropout from institutional delivery among HIV positive antenatal care booked mothers.

Conclusion

In this study, the risk factors of dropout from institutional delivery in the context of HIV-positive women were connected to social determinants of health such as advanced maternal age, unmarried marital status, unemployed husband occupation, and low family income. Therefore, interacting with the health system by focusing on these women in lower socio-economic strata and unmarried HIV-positive ANC attendees, and increasing access to information on obstetric complications during the antenatal care visit would retain clients in the continuum of maternity services.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13690-022-00819-0.

Determinants of the continuum of maternal health care in Cambodia: an analysis of the Cambodia demographic health survey 2014

Introduction

Cambodia has achieved significant progress in maternal health, yet remains in the group of countries with the highest maternal mortality ratio in South-East Asia. Extra efforts are needed to improve maternal health through assessing the coverage of maternal health services as a continuum of care (CoC) and identifying the gaps. Our study aims to explore the coverage level of the Optimal CoC by (1) measuring the continuity of optimal antenatal care (ANC), skilled birth attendance (SBA) and optimal post-natal care (PNC), (2) identifying the determinants of dropping out from one service to another and (3) of not achieving the complete CoC.

Method

The study employed data from the Cambodia Demographic Health Survey 2014. We restricted our analysis to married women who had a live birth in the five years preceding the survey (n = 5678). Bi-variate and multivariate logistic regression were performed using STATA version 14.

Results

Almost 50% of women had achieved the complete optimal CoC, while the remaining have used only one or two of the services. The result shows that the level of women’s education was positively associated with the use of optimal ANC, the continuation to using optimal PNC and achieving the complete CoC. More power of women in household decision making was also positively associated with receiving the complete CoC. The birth order was negatively associated with achieving the complete CoC, while exposure to the mass media and having health insurance increased the odds of achieving the complete CoC. Household wealth consequently emerged as an influential predictor of dropping out and not achieving the complete CoC. Receiving all different elements of ANC care improved the continuity of care from optimal ANC to SBA and from SBA to optimal ANC.

Conclusion

The findings urge policy makers to approach maternal health care as a continuum of care with different determinants at each step. Household wealth was found to be the most influential factor, yet the study discovered also other barriers to optimal maternal health care which need to be addressed: future intervention should thus not only aim to increase wealth or health insurance coverage but also stimulate the education of women and empower women to claim power in household decision-making.

Integrated Information System for Early Detection of Maternal Risk Factors Based on Continuum of Care Approach of Mother and Toddler Cohorts

Objectives

The aim of this study is to demonstrate how an integrated information system of mother and toddler cohorts can be developed as a basis of the continuum of care approach that subsequently can be used as the basis of early detection of risk factors of maternal mortality.

Methods

This research was carried out qualitatively. The data was collected through three techniques: in-depth interviews, focus group discussion, and document studies at six public health centers located in four sub-districts of Surabaya, Indonesia. This research was conducted from 2016 to 2018.

Results

The data collected from this research has become a basis input data requirement analysis for an integrated mother and toddler cohort information system. The system accommodates all the variables in each period of pre-marriage, pregnancy, labor, infancy and toddlerhood. The system facilitates healthcare workers to retrieve data and information related to mother and toddler health status.

Conclusions

The availability of various pieces of information enables the health status of mothers and toddlers to be monitored thoroughly throughout their long-life cycle. This continuum of care approach is beneficial in the early detection and management of risk factors of maternal mortality, such as pregnancy complications as well as childbirth and postpartum complications.

A Rapid Review of eHealth Interventions Addressing the Continuum of HIV Care (2007-2017)

National HIV prevention goals call for interventions that address Continuum of HIV Care (CoC) for persons living with HIV. Electronic health (eHealth) can leverage technology to rapidly develop and disseminate such interventions. We conducted a qualitative review to synthesize (a) technology types, (b) CoC outcomes, (c) theoretical frameworks, and (d) behavior change mechanisms. This rapid review of eHealth, HIV-related articles (2007-2017) focused on technology-based interventions that reported CoC-related outcomes. Forty-five studies met inclusion criteria. Mobile texting was the most commonly reported technology (44.4%, k = 20). About 75% (k = 34) of studies showed proven or preliminary efficacy for improving CoC-related outcomes. Most studies (60%, k = 27) focused on medication adherence; 20% (k = 9) measured virologic suppression. Many eHealth interventions with preliminary or proven efficacy relied on mobile technology and integrated knowledge/cognition as behavior change mechanisms. This review identified gaps in development and application of eHealth interventions regarding CoC.