Faster Evidence: AI Thematic Analysis for Public Health

According to the November 4, 2026 PLOS One study, routine antenatal care is a potentially viable platform for balanced energy and protein supplementation, but implementation depends on clear definitions, trained providers, reliable supplies, demand generation, and routine monitoring.

AI thematic analysis for public health can help researchers synthesize interview transcripts and policy documents faster while preserving a transparent link between themes, quotations, countries, and implementation decisions.

• The PLOS One study analyzed 13 stakeholder interviews conducted from August to December 2023 and reviewed antenatal care guidelines from 6 countries.
• The PLOS One study found that Ethiopia, Malawi, and Mozambique had revised national antenatal care guidance to create an enabling environment for balanced energy and protein supplementation.
• The PLOS One study reported that 3 countries had revised relevant guidance, while stakeholders still described confusion about the definition of balanced energy and protein supplementation.
• A 2021 systematic review cited by the PLOS One study found reductions of 29% in small-for-gestational-age births, 40% in low birthweight, and 61% in stillbirth across 9 low- and middle-income countries.
• The PLOS One study concluded that standardized definitions, operational guidance, and monitoring indicators are needed for quality country implementation.

The PLOS One research team used a mixed-methods landscape analysis to connect stakeholder perspectives with formal policy evidence. From August through December 2023, the team conducted 13 semi-structured key informant interviews with global and country stakeholders working in maternal health and nutrition programs.

The PLOS One study also reviewed national antenatal care guidelines from Colombia, Ethiopia, Ghana, Malawi, Mozambique, and Nepal. Interviews lasted 1.5 to 2 hours, were recorded and transcribed, and included English and Portuguese content that was translated and back-translated for quality assurance.

The researchers developed a deductive codebook from a conceptual framework, refined the codebook with emergent themes, and analyzed themes and subthemes in Dedoose version 10.0. Three research team members independently analyzed transcripts, then resolved initial coding discrepancies through consensus.

The study identified recurring implementation themes involving government commitment, provider readiness, supply chains, product acceptability, household sharing or resale, demand generation, and monitoring. One global donor stakeholder summarized the operational need directly: "It is important to ensure that there are capacity building activities as well as proper training to provide or support quality counseling [for BEP]."

For AI-enabled qualitative research, the study offers a practical model: combine structured coding with emergent themes, preserve quotations as evidence, compare findings across countries, and use human review to resolve ambiguity.

Public health researchers should treat qualitative evidence as an implementation asset, because the PLOS One study shows that policy readiness alone does not guarantee service readiness. Ethiopia, Malawi, and Mozambique had enabling policy developments, yet stakeholders still reported weak provider capacity, supply constraints, and uncertainty about how to integrate supplementation into routine services.

• Researchers should separate policy availability from operational readiness by coding guidelines, interviews, supply-chain evidence, and provider experiences as distinct evidence streams.
• Researchers should analyze acceptability by country and stakeholder group because the PLOS One study linked taste, texture, cultural norms, household food insecurity, sharing, and resale to adherence.
• Researchers should preserve negative cases and conflicting definitions because the PLOS One study found that stakeholders confused balanced energy and protein supplements with other nutrition products.
• Program teams should convert qualitative themes into measurable indicators because the PLOS One study reported that no examined country routinely collected balanced energy and protein indicators through existing health systems.
• Evaluation teams should document the limits of the evidence because the PLOS One analysis used purposive sampling, focused on selected USAID-supported programs, and did not directly interview frontline providers or community health workers.

A public health team can use AI qualitative research analysis to create a first-pass code structure, retrieve supporting quotations, compare stakeholder groups, and identify themes that require expert review. Human researchers still need to validate codes, assess context, protect participant confidentiality, and make final interpretations.

The November 4, 2026 PLOS One study shows that balanced energy and protein supplementation through routine antenatal care requires more than a policy recommendation: implementation depends on provider capacity, reliable supplies, culturally appropriate demand generation, and usable monitoring systems.

The study also illustrates how rigorous qualitative research connects stakeholder quotations and cross-country comparisons to practical program decisions. AI thematic analysis can accelerate those steps when researchers preserve source evidence, review model outputs, and apply clear ethical and methodological controls.

Evidano helps public health and implementation teams analyze interviews, documents, and survey responses in a structured workflow. Try Evidano for free to turn qualitative evidence into clearer, reviewable insights.