AI-enabled qualitative research: LARC uptake in Bangladesh

Primary keyword: AI-enabled qualitative research. Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. The PLOS One study by Shoma and Barden-O’Fallon (published August 25, 2026) provides a detailed, multi-level qualitative dataset on why long-acting reversible and permanent contraceptive methods (LARC/PM) remain underused in rural Cumilla, Bangladesh, and offers a practical case for demonstrating AI-enabled qualitative research workflows. The following post explains what the study measured, highlights concrete statistics from the paper, and shows how AI-enabled qualitative research accelerates transparent thematic analysis and rights-respecting program design.

Key Takeaways

According to the PLOS One study (Shoma & Barden-O’Fallon, published August 25, 2026), persistent individual, interpersonal, community, institutional, and health-system barriers keep LARC/PM uptake low in rural Cumilla, Bangladesh, and these multilevel barriers are well suited to reproducible AI-enabled qualitative research synthesis using coded interview data and cross-segment analysis.

28 in-depth interviews and 9 key informant interviews were conducted during 21 January–20 February, 2023, and reported in the PLOS One article published August 25, 2026.
The PLOS One analysis reports that LARC/PM share in Bangladesh declined from 12.1% in 1991 to 8% in 2022, and that Cumilla district recorded 13.6% LARC/PM use in 2022, underscoring the geographic concentration of low uptake.
The PLOS One authors found that fear of side effects, provider bias, male-dominated decision-making, religious stigma, and workforce shortages together limit voluntary, informed LARC/PM access in rural communities.

What happened: study design and measures

Answer: The PLOS One study used a socio-ecological qualitative design to probe multilevel barriers to LARC/PM uptake in rural Cumilla, Bangladesh.

The PLOS One study (Shoma & Barden-O’Fallon, published August 25, 2026) collected data via 28 in-depth interviews with married women and men and 9 key informant interviews with family planning providers and a religious leader, with data collection conducted between 21 January and 20 February, 2023.

The PLOS One team analyzed verbatim Bengali transcripts translated into English using Braun and Clarke’s thematic analysis framework, and organized findings across individual, interpersonal, community, institutional, and health-system levels.

The PLOS One authors reported concrete method-level findings: among 22 female IDI participants described in the results, 20 knew about implants, 9 knew about IUDs, and 8 were aware of vasectomy as an option, illustrating uneven familiarity across methods.

Findings snapshot

Date / Source	Metric	Value	Implication
Published Aug 25, 2026 (PLOS One)	Qualitative sample	28 IDIs + 9 KIIs	Rich multi-stakeholder narratives suitable for thematic coding and cross-segment analysis
Data collected Jan 21–Feb 20, 2023 (PLOS One)	Female IDI familiarity	Implant: 20/22 women; IUD: 9/22 women; Vasectomy knowledge: 8/22 women	Knowledge gaps concentrate on IUDs and male methods, guiding targeted education
National data cited in PLOS One	LARC/PM national trend	12.1% in 1991 → 8% in 2022	Programmatic decline despite overall contraceptive progress, requires systems-level response
DGFP data cited in PLOS One (2022)	Cumilla district LARC/PM use	13.6% in 2022	Local hotspots need tailored, community-sensitive interventions

Implications for qualitative researchers and program designers

Answer: The PLOS One findings imply that qualitative researchers and program designers must analyze LARC/PM uptake as a multilevel problem that mixes embodied experiences, rumor networks, religious meaning, and service quality.

The PLOS One study (Shoma & Barden-O’Fallon, published August 25, 2026) shows that fears often arise from lived or vicarious bodily experiences (for example, excessive bleeding after implant insertion), which qualitative analysis must code as embodied side-effect narratives rather than mere misinformation.

The PLOS One authors observed provider-level problems including brief counselling, privacy deficits, and coercive practices; researchers should therefore code and quantify occurrences of consent violations and counselling depth to feed rights-based program feedback loops.

Program designers should prioritize couple- and community-engagement, faith leader dialogue, and workforce supports, because the PLOS One analysis links male-dominated decision-making (15 of 22 women reported husband final say) and religious stigma to low uptake.

How Evidano helps translate PLOS One-style interviews into actionable evidence

Problem: Large transcript sets are slow to synthesize

Solution: Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents; Evidano automates transcription, thematic coding, and frequency counts to reduce time-to-insight.

Evidano’s thematic and content-frequency analyses let researchers reproduce the PLOS One workflow (inductive + deductive coding guided by the socio-ecological model) and generate code co-occurrence maps to spot links between, for example, "bleeding concerns" and "provider mistrust" across segments.

Problem: Missing cross-segment comparisons (e.g., women vs men, migrants vs non-migrants)

Solution: Evidano produces cross-segment analyses and visualizations so teams can compare code prevalence by gender, migration status, or service user group as the PLOS One study contrasted migrants and non-migrants.

Evidano users can upload interview transcripts and spreadsheets and run multigroup frequency and co-occurrence queries that reproduce the PLOS One-style disaggregation without manual counting.

Problem: Incomplete, inconsistent transcriptions and privacy risk

Solution: Evidano provides transcription with custom dictionaries, PII redaction, and translation support so researchers can preserve participant confidentiality and the study’s original meaning during English translation, mirroring the PLOS One translation checks.

Learn more on the Evidano features page about transcription, PII redaction, and AI-chat over documents.

FAQ: AI-enabled qualitative research

How does AI-enabled qualitative research speed thematic analysis of interviews?

Answer: AI-enabled qualitative research speeds analysis by automating transcription, initial code suggestions, and frequency counting while preserving researcher-led interpretation.

AI-assisted platforms like Evidano generate reproducible code suggestions and let researchers accept, edit, or reject AI codes, which shortens the mechanical workload described in traditional manual coding approaches such as Braun and Clarke’s thematic analysis used by the PLOS One authors.

Can AI reproduce the socio-ecological coding frame used in the PLOS One study?

Answer: Yes, AI-assisted tools can implement hybrid coding schemes that combine deductive SEM-based labels with inductive codes from participant narratives.

The PLOS One study (Shoma & Barden-O’Fallon, published August 25, 2026) used a hybrid inductive-deductive approach; Evidano supports uploading a deductive codebook and then running inductive discovery to surface emergent sub-themes.

Is AI qualitative analysis appropriate for sensitive topics like contraception?

Answer: AI-enabled workflows are appropriate when combined with strong ethics: privacy safeguards, human review, and rights-based consent practices.

The PLOS One research obtained ethical approval from the University of Dhaka and maintained de-identification in quotes; similarly, Evidano provides PII redaction and encryption to help teams meet ethical standards while analyzing sensitive interview data.

How do I preserve verbatim quotes and context when using AI tools?

Answer: Preserve verbatim quotes by keeping the original transcript layer intact, using AI for coding and summarization but not for replacing original textual evidence.

The PLOS One paper included verbatim participant quotes and cross-checked translations against Bengali audio; Evidano supports exportable quotations linked to source timestamps so analysts can verify context.

Conclusion & Next Steps

The PLOS One study (published August 25, 2026) demonstrates that low LARC/PM uptake in rural Bangladesh results from interacting socio-ecological barriers that demand transparent, multilevel qualitative analysis.

AI-enabled qualitative research accelerates reproducible synthesis of those barriers by automating transcripts, surfacing themes and code co-occurrences, and enabling rapid cross-segment queries to guide program and policy responses.

If you want to reproduce PLOS One-style coding and generate actionable, rights-respecting insights from interview data, Try Evidano for free.

Topics

ai-enabled qualitative research
AI qualitative analysis
qualitative analysis of contraceptive uptake
LARC uptake Bangladesh

Keep reading

Browse all articles