Qualitative Analysis of Safety Reporting: AI tools

Qualitative analysis of safety reporting is essential for understanding why UK trial staff struggle to implement risk-proportionate reporting and what practical steps reduce burden. This post refracts the PLOS One qualitative study for researchers and CTU managers, then shows how AI-enabled qualitative research methods can accelerate synthesis, training, and cross-site learning. Readable extracts, verbatim quotes, and concrete numbers from the study are paired with actionable recommendations you can operationalise with AI-powered tools.

Key Takeaways

According to the PLOS One study, PLOS One, UK trials staff experience a "tightrope" between over-reporting and under-reporting that drives inefficiency and anxiety about decision-making.

23 CTU staff participated across four focus groups held between 01-July-2024 and 04-March-2025, according to PLOS One.
The study reported a mean participant trial experience of 13.3 years (SD 9.5) in July 2026, showing that experienced staff still face uncertainty, per PLOS One.
UK regulatory reform came into force on 28-Apr-2026, and the study recommends targeted training, clearer guidance, and shared CTU resources to realise those changes, as reported in PLOS One.

What Happened and how the study was done

Answer: The PLOS One study used qualitative focus groups to identify barriers and solutions to efficient safety reporting in UK academic trials.

According to the PLOS One study, PLOS One, researchers ran four online focus groups between 01-July-2024 and 04-March-2025 with 23 CTU staff from 10 of 52 registered UK trials units. The sample included operational and clinical staff with a mean of 13.3 years' experience (SD 9.5).

According to the PLOS One study, participants reported one over-arching theme, “Walking on a tightrope: Making justifiable decisions”, and five subthemes explaining why risk-proportionate reporting is hard to enact.

Direct quotations from participants illustrate the lived problem: "Generally, you’ve got to balance what you need to collect in terms of safety reporting and in terms of what’s needed to monitor the safety of the drug" (Participant 16, FG3), and "It feels difficult and there is always that element of anxiety that you’ve got it wrong" (Participant 20, FG4), both cited in PLOS One.

Findings Snapshot

Date / Period	Metric	Value	Implication
01-Jul-2024 to 04-Mar-2025	Focus groups conducted	4 online focus groups	Qualitative depth from multi-site staff perspectives
July 2026 (published 30-Jul-2026)	Participants	23 CTU staff (consented 24, 1 unable to attend)	Experienced sample but weighted toward oncology; findings transferable in CTIMP contexts
Reported in study	Mean trial experience	13.3 years (SD 9.5)	Even experienced staff report uncertainty about reporting decisions
28-Apr-2026	Regulatory change in force	New UK Clinical Trials Regulations (safety reporting simplified)	Study positions recommendations to support implementation of April 2026 reforms
Study results	Primary theme	"Walking on a tightrope: Making justifiable decisions"	Tension between risk-proportionate reporting and risk-aversion drives inefficiency

Implications for CTU researchers and trial managers

Answer: CTU researchers should prioritise role-specific training, explicit decision records, and cross-unit sharing to operationalise the April 2026 regulatory changes.

According to the PLOS One study, PLOS One, key barriers were: uncertainty about consequences, bureaucratic red-tape, inconsistent stakeholder requirements, lack of clarity, and gaps in experience.

According to the PLOS One study, recommended practical steps include: using MHRA, HRA and ICH resources for training; building CTU networks to share templates and case studies; and developing mentoring and accessible web resources to transfer tacit knowledge.

Actionable first steps for CTUs: 1) Run short, example-based training on SAE/SUSAR decisions tied to protocols; 2) document decision rationale on case report forms or a central log; 3) agree safety-reporting parameters with Sponsors at trial set-up to limit duplication.

How Evidano Helps

Evidano: definition and fit

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

Evidano ingests transcripts, documents and spreadsheets, then produces thematic, frequency, content and cross-segment analyses that convert messy qualitative data into auditable evidence for decisions.

Problem: Unclear patterns across sites → Solution: Thematic consolidation

Problem: CTUs reported inconsistent practices and duplicated reporting across trials, per PLOS One.

Solution: Evidano automatically codes verbatim transcripts and aggregates themes across focus groups, producing shareable codebooks and visualisations that make cross-unit inconsistencies visible and actionable (see features).

Problem: Training gaps and tacit knowledge → Solution: Searchable case libraries

Problem: The study recommended practical, example-based training and mentoring to raise confidence, per PLOS One.

Solution: Evidano extracts exemplar quotes and decision rationales, enabling CTUs to build searchable case-study libraries and short training modules from real transcripts and documents.

Problem: Time-consuming transcription and redaction → Solution: Fast, compliant transcripts

Problem: Focus groups in the study were auto-transcribed then manually cleaned, adding time and risk of PHI exposure, per PLOS One.

Solution: Evidano offers secure speech-to-text with custom dictionaries and PII redaction to deliver analysis-ready transcripts faster and safer (see speech-to-text).

Problem: Need for rapid synthesis for regulators → Solution: Exportable evidence packs

Problem: CTUs need audit-ready summaries to justify risk-proportionate choices to regulators, a central concern in the study, per PLOS One.

Solution: Evidano generates downloadable summaries, frequency tables, and verbatim quote sets that support transparent, traceable decision-making for MHRA/HRA submissions.

FAQ: qualitative analysis of safety reporting

How does the PLOS One study show safety reporting is inefficient?

Answer: The study identifies a core tension (staff balance risk-proportionate reporting against risk-aversion) which produces over-reporting, duplication, and confusion.

According to the PLOS One study, participants described bureaucratic burdens, inconsistent stakeholder expectations, and uncertainty about consequences; these factors led to excessive reports and time spent on administrative tasks rather than signal detection (PLOS One).

Will the April 2026 UK regulatory changes fix these problems?

Answer: The April 2026 regulations simplify several reporting requirements, but the study finds implementation depends on training and clarity at the CTU and Sponsor level.

According to the PLOS One study, legislative change (in force 28-Apr-2026) reduces duplication like extra REC reporting, yet participants said clear guidance, mentoring, and shared resources are still needed to change daily practice (PLOS One).

How can AI qualitative research speed adoption of risk-proportionate reporting?

Answer: AI-assisted coding and synthesis turn scattered qualitative evidence into actionable recommendations and training materials faster than manual methods.

AI tools can extract decision-rationales, cluster common confusions, and produce standardised templates and exemplars from transcripts, which addresses the study's call for case studies and shareable web resources (PLOS One).

Is this research relevant if my trials are not oncology-focused?

Answer: Yes, the study focused on CTIMP trials with an oncology skew, but the core mechanisms (bureaucracy, ambiguity, and training gaps) are common across CTIMPs.

According to the PLOS One study, the authors note the sample was weighted toward oncology yet argue many safety-reporting definitions and challenges are universal across CTIMP contexts (PLOS One).

Conclusion & Next Steps

The PLOS One study shows that even experienced CTU staff feel caught between over-reporting and under-reporting, and that April 2026 regulatory changes remove some barriers but not the need for training, clarity, and shared case-based resources (PLOS One).

AI-enabled qualitative analysis can accelerate synthesis of focus groups, generate auditable evidence for decision-making, and produce searchable training libraries that the study recommends.

Next steps for CTUs: adopt example-driven training, document decision rationales, and pilot AI-assisted synthesis on a small set of transcripts to create shareable templates.

If you want to convert qualitative evidence into decision-ready outputs and training materials, Try Evidano for free.

Topics

qualitative analysis of safety reporting
AI qualitative research
clinical trial safety reporting
safety reporting qualitative study

Keep reading

Browse all articles

Qualitative Analysis of Safety Reporting: AI tools

Key Takeaways

What Happened and how the study was done

Findings Snapshot

Implications for CTU researchers and trial managers

How Evidano Helps

Evidano: definition and fit

Problem: Unclear patterns across sites → Solution: Thematic consolidation

Problem: Training gaps and tacit knowledge → Solution: Searchable case libraries

Problem: Time-consuming transcription and redaction → Solution: Fast, compliant transcripts

Problem: Need for rapid synthesis for regulators → Solution: Exportable evidence packs

FAQ: qualitative analysis of safety reporting

How does the PLOS One study show safety reporting is inefficient?

Will the April 2026 UK regulatory changes fix these problems?

How can AI qualitative research speed adoption of risk-proportionate reporting?

Is this research relevant if my trials are not oncology-focused?

Conclusion & Next Steps

Topics

Keep reading

Company

About

Preview Evidano

Account

Newsletter