Evidano home
Commentary on News

AI Qualitative Analysis: BEP Research Lessons

Evidano9 min read

Maternal health and nutrition teams often have rich interview data but limited time to compare stakeholder views, identify implementation barriers, and connect findings to policy decisions. The November 2026 PLOS One study on balanced energy protein supplementation shows why structured qualitative analysis matters: researchers examined 13 stakeholder interviews and antenatal care guidelines from six countries to understand how BEP could be introduced through routine services. This article explains how AI qualitative analysis can support similar health research workflows while preserving human interpretation, methodological transparency, and ethical review. Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and research documents. The platform can help teams move from transcripts to themes, cross-segment comparisons, and evidence-backed questions without treating automated output as a substitute for research judgment.

Key Takeaways

According to the PLOS One study published on November 4, 2026, routine antenatal care is a potentially viable platform for balanced energy protein supplementation, but unclear definitions, weak provider capacity, supply problems, demand barriers, and limited monitoring could impede implementation.

AI qualitative analysis can help maternal health researchers organize large interview datasets, compare stakeholder groups, surface recurring implementation themes, and preserve an auditable path from evidence to recommendation.

  • The 2026 PLOS One landscape analysis combined 13 key informant interviews conducted from August to December 2023 with a review of antenatal care guidelines from six countries.
  • The 2026 PLOS One study found that Ethiopia, Malawi, and Mozambique had revised national antenatal care guidelines to create an enabling environment for BEP introduction.
  • The 2026 PLOS One article reported that a 2021 systematic review covering 15 randomized controlled and quasi-experimental trials in nine low- and middle-income countries associated BEP supplementation with reductions of 29% in small-for-gestational-age births, 40% in low birthweight, and 61% in stillbirth.
  • The 2026 PLOS One study reported that 64% of pregnant women globally attended at least four antenatal care visits, supporting antenatal care as a possible delivery platform.
  • The 2026 PLOS One researchers analyzed interviews thematically in Dedoose, demonstrating why codebook design, transcript quality, consensus procedures, and human review remain important even when AI tools are added.

What the BEP Antenatal Care Study Shows

The 2026 PLOS One study found that introducing BEP through routine antenatal care requires more than distributing a nutritional product because implementation depends on policy, provider readiness, logistics, social norms, and monitoring.

The researchers conducted 13 semi-structured key informant interviews between August and December 2023 with global and country stakeholders, and they reviewed national antenatal care guidelines from Colombia, Ethiopia, Ghana, Malawi, Mozambique, and Nepal. The researchers recorded and transcribed the interviews, translated Portuguese content into English with back translation for quality assurance, and analyzed themes and subthemes using Dedoose version 10.0.

The study defined BEP as a food or supplement product in which protein contributes less than 25% of total energy. The 2026 PLOS One article described an expert-consultation composition range of 250 to 500 kilocalories, 14 to 18 grams of protein, and 3 to 33 grams of fat.

The study identified confusion between BEP, ready-to-use therapeutic food, and small-quantity lipid-based nutrient supplements. A multilateral organization stakeholder said, "If you look at the [WHO] guideline, it’s not clearly defined in terms of what the exact composition of BEP is."

The researchers also found that weak provider training, stockouts, household sharing, resale, limited demand generation, and absent routine BEP indicators could undermine implementation. A global donor organization stakeholder summarized the training requirement as follows: "It is important to ensure that there are capacity building activities as well as proper training to provide or support quality counseling [for BEP]."

BEP Research Findings Snapshot

Date or periodMetricValueImplication for qualitative researchers
August to December 2023Key informant interviews13 interviewsInterview synthesis needs a transparent codebook and consistent comparison across stakeholder roles.
2023 study periodCountries whose guidelines enabled BEP introduction3 of 6 countries, Ethiopia, Malawi, and MozambiqueCross-country analysis can distinguish policy readiness from operational readiness.
2026 publicationCountries included in guideline review6 countriesDocument analysis can be combined with interviews to test whether formal policy reflects field experience.
2026 article reporting 2021 review evidenceReduction associated with BEP supplementation29% for small-for-gestational-age birthsQualitative findings should explain implementation conditions behind quantitative outcomes.
2026 article reporting 2021 review evidenceReduction associated with BEP supplementation40% for low birthweightResearchers can connect outcome evidence with provider, supply, and adherence themes.
2026 article reporting 2021 review evidenceReduction associated with BEP supplementation61% for stillbirthHigh-stakes health findings require careful source tracking and human validation.
2026 articlePregnant women attending at least four antenatal care visits64%Antenatal care may provide a broad service platform, but qualitative research must examine access and acceptability.

Implications for AI Qualitative Analysis in Maternal Health

How can researchers analyze implementation barriers more systematically?

Researchers can analyze implementation barriers more systematically by combining deductive codes from a program framework with inductive codes that emerge from interviews. The 2026 PLOS One study used predetermined themes covering the enabling environment, service readiness, products and logistics, and monitoring, then refined its codebook as new themes appeared.

An AI-assisted workflow can apply an initial code structure across transcripts, identify passages that may represent new subthemes, and group related excerpts for human review. Researchers should retain the original excerpts, code definitions, and decisions that changed the final interpretation.

Why does cross-segment analysis matter for BEP implementation?

Cross-segment analysis matters because government stakeholders, health providers, donors, implementing organizations, and community actors can describe the same intervention differently. The 2026 PLOS One study found that policy readiness in some countries coexisted with uncertainty about product definitions, provider roles, supply chains, and monitoring.

AI can help compare themes by country, stakeholder type, interview language, or implementation domain. Human researchers must determine whether differences reflect genuine segment variation, sampling limitations, translation effects, or differences in question prompts.

What should health researchers do with qualitative evidence about sharing and selling?

Health researchers should treat sharing and selling as implementation findings that require contextual interpretation rather than simple compliance failures. The 2026 PLOS One study linked these behaviors to household food insecurity, social norms, product taste, product bulk, monotony, and financial pressure.

AI-assisted thematic analysis can cluster these explanations and reveal how they vary across locations or respondent groups. Researchers should avoid using automated frequency counts alone because a less frequently mentioned barrier can still be ethically or operationally important.

How Evidano Helps With AI Qualitative Analysis

Problem: Interview transcripts take too long to synthesize

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interview transcripts, open-ended survey responses, papers, reports, and other documents.

Evidano can support thematic, content, and frequency analysis so a maternal health team can move from raw interview text to a structured view of policy barriers, provider capacity, supply challenges, demand generation, and monitoring gaps. Researchers can review the supporting passages before using findings in a report.

Problem: Country and stakeholder differences are difficult to compare

Evidano can support cross-segment analysis by organizing qualitative evidence across countries, stakeholder groups, study waves, or implementation themes. This workflow is relevant to the 2026 PLOS One study because the researchers compared stakeholder evidence with guideline documents across six countries.

Researchers can use hierarchical codes and subcodes to separate broad domains, such as service readiness, from specific issues, such as training, task shifting, or provider workload. Teams can explore the resulting patterns through visualizations including co-occurrence networks and code hierarchies.

Problem: Researchers need to interrogate a large evidence base

Evidano’s AI chat over documents and analyses can help researchers ask focused follow-up questions about coded evidence, such as which stakeholders discussed stockouts or how acceptability concerns differed by country. AI-generated answers should be checked against the underlying excerpts and study design.

For sensitive health research, Evidano states that data is encrypted and never used to train third-party models. Researchers can review the Evidano data security information when evaluating tools for interview and document analysis.

FAQ: AI Qualitative Analysis for Maternal Health Research

What is AI qualitative analysis?

AI qualitative analysis is the use of artificial intelligence to organize, code, summarize, compare, and question unstructured research data such as interview transcripts, open-ended survey responses, and documents.

AI qualitative analysis supports researcher judgment rather than replacing study design, interpretation, ethics review, or validation. The 2026 PLOS One study illustrates the value of combining structured coding with iterative refinement and consensus.

How was qualitative data analyzed in the BEP antenatal care study?

The BEP antenatal care study analyzed 13 key informant interview transcripts using themes and subthemes in Dedoose version 10.0. Three research team members developed a codebook, analyzed transcripts independently, resolved early discrepancies through consensus, and applied revised codes across the dataset.

The researchers also reviewed antenatal care guidelines from six countries, creating a mixed-methods comparison between stakeholder experience and formal policy.

Can AI qualitative analysis replace human coding in health research?

AI qualitative analysis should not replace human coding and interpretation in health research. Human researchers must assess context, ambiguity, sampling limitations, translation quality, ethical risk, and whether an excerpt supports a conclusion.

AI can reduce repetitive organization work and help researchers find potentially relevant evidence, while documented human review protects validity and accountability.

What qualitative themes emerged from the 2026 BEP study?

The 2026 PLOS One study identified unclear BEP definitions, low awareness of recommendations, weak provider capacity, supply interruptions, high logistical burdens, acceptability concerns, household sharing, resale, and weak routine monitoring as major themes.

The study concluded that successful introduction requires government commitment, provider training, demand generation, compliance strategies, supply planning, local production assessment, and routine indicators.

How can researchers use AI to analyze antenatal care interviews responsibly?

Researchers can use AI responsibly by documenting the codebook, preserving source excerpts, checking translations, reviewing AI-generated codes, recording disagreements, and separating descriptive patterns from causal claims.

The 2026 PLOS One study reported that interview participants gave verbal informed consent, identifiers were removed, and data was stored securely, providing a useful reminder that data governance must accompany analysis.

Conclusion and Next Steps

The 2026 PLOS One BEP study shows that qualitative evidence can reveal the practical conditions behind maternal health policy, including provider readiness, supply chains, cultural acceptability, and monitoring. AI qualitative analysis can help research teams organize these dimensions faster, compare stakeholder perspectives, and connect recommendations to supporting evidence, provided that human researchers validate the results. Evidano gives health and policy researchers a workspace for thematic, frequency, document, and cross-segment analysis across interviews and reports. Try Evidano for free to turn qualitative evidence into a clearer research workflow.

Topics

  • AI qualitative analysis
  • AI thematic analysis
  • qualitative research in maternal health
  • BEP antenatal care research
  • health research interview analysis

Keep reading

Browse all articles