Predicting undernutrition among elementary schoolchildren in the Philippines using machine learning algorithms.

OBJECTIVES: This study aimed to compare the accuracy of four machine-learning (ML) algorithms, using two classification schemes, to predict undernutrition based on individual and household risk factors. METHODS: Data on public-school children were collected from a rural province (310 children) and a highly urbanized city (308 children) in the Philippines using 24-h dietary recalls and a household socioeconomic and demographic survey. Children's nutritional risk was classified based on acceptable macronutrient distribution ranges (AMDRs) developed by the National Academy of Medicine (NAM) and Philippine Dietary Reference Intakes (PDRIs). Four algorithms (random forest, support-vector machine, linear discriminant analysis, and logistic regression) predicted undernutrition in the sample, and their accuracy, sensitivity, and specificity were compared. Predictions were also compared with the national school feeding program's anthropometric classifications. RESULTS: The prevalence of undernutrition was greater under NAM AMDRs (82.67%) compared with PDRI AMDRs (78.71%). Random forest was the most accurate ML algorithm (78.55%), able to predict undernutrition based on household expenditures, child and household age, food insecurity, and dietary diversity. Compared with anthropometric classification (213 children), AMDRs classified more children as at risk for inadequate dietary intake (477 children). CONCLUSIONS: The random forest algorithm performed best in predicting undernutrition among Filipino elementary schoolchildren, although results could be improved with bootstrap aggregation. The AMDR classification shows potential for targeting feeding beneficiaries. However, local dietary culture should be considered in the development of nutrition interventions. Government use of big-data techniques such as ML must also address underrepresentation in health data collected from and accessible to poor populations or risk further marginalizing them.

National multisectoral governance challenges of implementing the Philippines' Reproductive Health Law.

In recognition of the role of reproductive health in individual and national development, the Responsible Parenthood and Reproductive Health (RPRH) Law of 2012 was passed in the Philippines after 30 years of opposition and debate. Seven years later, this article examined the cohesiveness of national multi-sectoral governance among state and non-state actors and identified challenges in coordination as part of the first comprehensive evaluation of the landmark policy. Using a qualitative intrinsic case study design and guided by the World Health Organization's systems checklist for governing health equity as our theoretical perspective, we conducted 20 semi-structured interviews with national implementers from health agencies (n = 11), non-health agencies (n = 6) and non-state actors (n = 3) that included civil society organizations (CSOs). Key themes identified through thematic analysis were supported with document reviews of policy issuances, accomplishment reports and meeting transcripts of the RPRH National Implementation Team (NIT). The study found that despite aspirations for vibrant multi-sectoral coordination, the implementation of the RPRH Law in the Philippines was incohesive. National leaders, particularly the health sector, were neither able to rally non-health sector actors around RPRH nor strategically harness the power of CSOs. Local resource limitations associated with decentralization were exacerbated by paternalistic financing, coordination, and monitoring. The absence of multi-agency plans fostered a culture of siloed opportunism, without consideration to integrated implementation. This case study shows that for neutral policies without conflicts in sector objectives, the interest and buy-in of non-health state actors, even with a national law, cannot be assumed. Moreover, possible conflicts in interests and perspectives between state and civil society actors must be managed in national governance bodies. Overall, there is need for participatory policymaking and health-sector advocacy to set health equity as an intersectoral goal, involving subnational leaders in developing concrete action plans, and strengthening NIT's formal accountability systems.

The Impact of COVID-19 on Hospital Admissions for Twelve High-Burden Diseases and Five Common Procedures in the Philippines: A National Health Insurance Database Study 2019-2020.

BACKGROUND: The Philippines has the highest cumulative COVID-19 cases and deaths in the Western-Pacific. To explore the broader health impacts of the pandemic, we assessed the magnitude and duration of changes in hospital admissions for 12 high-burden diseases and the utilization of five common procedures by lockdown stringency, hospital level, and equity in patient access. METHODS: Our analysis used Philippine social health insurance data filed by 1,295 hospitals in 2019 and 2020. We calculated three descriptive statistics of percent change comparing 2020 to the same periods in 2019: (1) year-on-year, (2) same-month-prior-year, and (3) lockdown periods. FINDINGS: Disease admissions declined (-54%) while procedures increased (13%) in 2020 versus 2019. The increase in procedures was caused by hemodialysis surpassing its 2019 utilization levels in 2020 by 25%, overshadowing declines for C-section (-5%) and vaginal delivery (-18%). Comparing months in 2020 to the same months in 2019, the declines in admissions and procedures occurred at pandemic onset (March-April 2020), with some recovery starting May, but were generally not reversed by the end of 2020. Non-urgent procedures and respiratory diseases faced the largest declines in April 2020 versus April 2019 (range: -60% to -70%), followed by diseases requiring regular follow-up (-50% to -56%), then urgent conditions (-4% to -40%). During the strictest (April-May 2020) and relaxed (May-December 2020) lockdown periods compared to the same periods in 2019, the declines among the poorest (-21%, -39%) were three-times greater than in direct contributors (-7%, -12%) and two-times more in the south (-16%, -32%) than the richer north (-8%, -10%). Year-on-year admission declines across the 12 diseases and procedures (except for hemodialysis) was highest for level three hospitals. Compared to public hospitals, private hospitals had smaller year-on-year declines for procedures, because of increases in utilization in lower level private hospitals. INTERPRETATION: COVID-19's prolonged impact on the utilization of hospital services in the Philippines suggests a looming public health crisis in countries with frail health systems. Through the periodic waves of COVID-19 and lockdowns, policymakers must employ a whole-of-health strategy considering all conditions, service delivery networks, and access for the most vulnerable. FUNDING: Open Philanthropy.