AI-enabled qualitative analysis: IEN ecosystem

Primary keyword: AI-enabled qualitative analysis. Healthcare researchers and HR leaders need fast, reliable syntheses of small qualitative datasets such as focus groups about internationally educated nurses, and AI-enabled qualitative analysis can shorten the path from audio to actionable themes. According to PLOS ONE (Ryan et al., 2026), a reflexive thematic analysis of four focus groups with 21 IENs produced a multilayered ecosystem framework (macro, exo, meso, micro) that links national policy to day-to-day retention. This post explains the PLOS ONE methods and findings, gives concrete extraction steps that qualitative teams can apply with AI tools, and maps common project bottlenecks to specific Evidano features for secure, audit-ready analysis.

Key Takeaways

According to PLOS ONE (Ryan et al., 2026), a qualitative study of 21 internationally educated nurses (IENs) used four semi-structured focus groups and reflexive thematic analysis to show that retention is shaped by interacting macro, exo, meso, and microsystem forces.

21 participants, four focus groups, conducted between June 8 and June 29, 2023, provided the dataset used in the PLOS ONE analysis (Ryan et al., 2026).
PLOS ONE (Ryan et al., 2026) reports 206 initial analytic codes generated during coding, showing high thematic detail to synthesise.
Ireland’s national context was stressed: PLOS ONE (Ryan et al., 2026) cites OECD data that 51.8% of the Irish nursing population were IENs in 2023, and 78% of new nursing registrants in 2023 were IENs.
PLOS ONE (Ryan et al., 2026) includes verbatim participant concerns such as “We have colleagues who have waited for more than a year to just have your family in. That’s a very big challenge” (P2, FG2), which signals policy-level barriers affecting retention.

What happened and how the PLOS ONE study worked

Answer: PLOS ONE (Ryan et al., 2026) ran four face-to-face semi-structured focus groups with 21 IENs across one Irish hospital group to map how layered ecosystems affect talent management.

According to PLOS ONE (Ryan et al., 2026), focus groups took place between June 8 and June 29, 2023, each lasting about 56 minutes on average, with audio recorded and professionally transcribed.

According to PLOS ONE (Ryan et al., 2026), NVivo™ version 14 was used to manage transcripts and the authors followed Braun and Clarke’s six-phase reflexive thematic analysis, producing 206 initial codes that were iteratively clustered into ecosystem-level themes.

According to PLOS ONE (Ryan et al., 2026), results show macrosystem barriers (visa delays, housing, childcare), exosystem gaps (recruitment reality gaps, rushed induction), mesosystem operational pressures (understaffing, inconsistent leave), and microsystem interpersonal dynamics (microaggressions and allyship).

Findings snapshot

Date	Metric	Value	Implication
June 8-29, 2023	Focus group dates	4 focus groups, N = 21 participants	Dataset size typical of intensive qualitative study; suitable for deep reflexive TA
2026-08-20	Publication	PLOS ONE article (Ryan et al., 2026)	Peer-reviewed, open access evidence for policy and practice
During analysis	Initial analytic codes	206 codes	High code granularity; needs structured synthesis and frequency analysis
2023	National IEN share	51.8% of nursing population in Ireland	High reliance on IENs makes retention a strategic priority
2023	New registrants who were IENs	78% of new entrants	Recruitment-success but retention and integration risks remain

Implications for qualitative teams and healthcare researchers

Answer: Research teams should treat small focus-group datasets like the PLOS ONE study as high-value but analysis-intensive, and plan for coding depth, cross-level synthesis, and stakeholder-ready outputs.

According to PLOS ONE (Ryan et al., 2026), the study generated 206 codes from 21 participants, which implies that manual synthesis without tooling will be time-consuming and error prone.

According to PLOS ONE (Ryan et al., 2026), policy-relevant themes (visa, housing, childcare) emerged at the macrosystem layer, meaning qualitative reports should explicitly link participant quotes to policy recommendations and timelines.

Practical steps for teams: (1) preserve verbatim quotes with identifiers as PLOS ONE did, (2) quantify code frequency by theme to show scope (PLOS ONE’s 206-code stage is an example), (3) map themes to ecosystem layers for multi-stakeholder audiences.

How Evidano helps: from audio to policy-ready themes

Evidano definition

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.

Problem: Focus groups produce rich text and many initial codes, as seen in PLOS ONE where researchers generated 206 initial codes from 21 participants; Solution: Evidano ingests transcripts, auto-extracts themes and code frequencies, and lets teams validate reflexively.

Feature mapping: Problem: Time-consuming verbatim transcription and PII risk; Feature: Evidano offers secure transcription with custom dictionaries and PII redaction (speech-to-text).

Feature mapping: Problem: Managing 200+ codes and cross-level mappings; Feature: Evidano provides thematic, content, frequency, and cross-segment analyses plus hierarchical code→subcode visualizations (features).

Feature mapping: Problem: Need for transparent audit trail to support policy asks; Feature: Evidano keeps encrypted, non-training data logs and exportable decision records for reproducibility and ethics review (data security).

FAQ: AI-enabled qualitative analysis

What is AI-enabled qualitative analysis and when should I use it?

Answer: AI-enabled qualitative analysis combines human-led coding with AI assistance to speed transcription, code suggestion, and frequency counts while preserving researcher interpretation.

According to PLOS ONE (Ryan et al., 2026), reflexive thematic analysis required iterative human judgment across six phases; AI should therefore assist tasks like transcription and initial code clustering, not replace reflexive interpretation.

Can AI reproduce Braun and Clarke’s reflexive thematic analysis?

Answer: AI can accelerate parts of reflexive TA, but human reflexivity and analytic memos remain essential.

According to Braun and Clarke and as applied in PLOS ONE (Ryan et al., 2026), reflexive TA is researcher-led; AI is best used to surface patterns and support transparent documentation of code merges and splits.

How do I securely transcribe and analyze sensitive focus groups?

Answer: Use a platform that supports encrypted storage, PII redaction, and controlled access for transcripts.

Evidano’s transcription features include custom dictionaries and PII redaction to match the ethical safeguards described by PLOS ONE (Ryan et al., 2026) for vulnerable participants, and evidence export supports ethics audits.

How reliable are AI-suggested themes for policy recommendations?

Answer: AI-suggested themes are reliable as a starting point when combined with human validation and reflexive memos.

According to PLOS ONE (Ryan et al., 2026), authors used NVivo™ for organisation but relied on human-led theme development; AI can reproduce clustering and frequency metrics but researchers must define final themes and policy links.

How can I reproduce the PLOS ONE ecosystem mapping with my dataset?

Answer: Map codes to macro, exo, meso, and microsystem layers, then quantify code counts per layer and triangulate with participant quotes.

PLOS ONE (Ryan et al., 2026) provides a worked example: 206 initial codes were reduced into ecosystem themes with illustrative quotes; replicate by running code frequency, co-occurrence networks, and exporting theme grids for stakeholders.

Conclusion & Next Steps

Answer: AI-enabled qualitative analysis turns the intensive coding work in studies like PLOS ONE (Ryan et al., 2026) into faster, auditable insight streams that researchers and policymakers can act on.

The PLOS ONE study (Ryan et al., 2026) shows that small but deep focus-group datasets generate many codes and policy-relevant themes; using AI for transcription, code frequency, and cross-segment synthesis preserves researcher reflexivity while accelerating delivery.

If your team needs secure transcription, rapid thematic and cross-segment analysis, and exportable audit trails to inform retention policy for IENs, consider tooling that is built for qualitative research workflows and compliance.

Ready to try this on your next focus-group dataset? Try Evidano for free.

Topics

AI-enabled qualitative analysis
qualitative analysis of internationally educated nurses
AI thematic analysis
focus group transcription

Keep reading

Browse all articles