This post explains how clinical trials units (CTUs), trial managers, and pharmacovigilance teams can use qualitative analysis to make safety reporting more efficient under the new UK framework. The primary keyword for this guide is qualitative analysis of safety reporting. The audience is UK-based and international academic trial teams who must implement risk-proportionate approaches introduced into UK law in April 2026. The payoff is a short, practical workflow that combines reflexive thematic insights with AI-enabled transcription and coding to reduce noise, increase transparency, and accelerate decisions.
Key Takeaways
According to the PLOS One study published July 30, 2026, trial staff describe safety reporting as a ‘‘tightrope’’ between over-reporting and under-reporting, driven by uncertainty, red tape, inconsistent requirements, lack of clarity, and gaps in experience. The PLOS One study is the source for the participant quotes and statistics below. Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents.
- 23 CTU staff participated in four focus groups run between 01-July-2024 and 04-March-2025, according to PLOS One published July 30, 2026.
- Participants came from 10 of 52 UK-registered trials units and had a mean of 13.3 years’ trial experience (SD 9.5), as reported in PLOS One on July 30, 2026.
- The PLOS One authors note that legislative changes signed in 2025 and coming into force 28 April 2026 aim to reduce duplicative reporting and codify risk levels for trials.
- Practical solutions reported in PLOS One include clearer regulator guidance, role-specific training, and CTU-level knowledge sharing platforms.
What happened and how the PLOS One study measured it
The PLOS One study asked: what are the barriers and solutions to efficient safety reporting for academic trial staff. The answer is a qualitative focus group study with 23 participants in four online focus groups held between 01-July-2024 and 04-March-2025, as described in PLOS One published July 30, 2026.
The PLOS One study recorded and auto-transcribed Microsoft Teams sessions, then cleaned and pseudonymised transcripts before applying Reflexive Thematic Analysis, using NVivo for coding. The analytic output produced one overarching theme, "Walking on a tightrope: Making justifiable decisions, " and five subthemes that explain sources of inefficiency.
Direct quotes from participants illustrate the tensions identified in PLOS One, for example: "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).
Findings Snapshot
| Date / Source | Metric | Value | Implication |
|---|---|---|---|
| 01-July-2024 to 04-March-2025, PLOS One | Focus groups | 4 groups, 23 participants | Qualitative breadth across CTU roles, supports transferability within UK academic trials |
| July 30, 2026, PLOS One | CTUs represented | 10 of 52 registered UK CTUs | Inconsistent SOPs and interpretations across CTUs identified as a source of duplication |
| Reported in PLOS One | Participant experience | Mean 13.3 years (SD 9.5) | Experienced staff report bureaucratic friction and training gaps among less experienced staff |
| Legislative timeline | UK clinical trial regulations | Signed 2025, in force 28 April 2026 | New rules reduce duplicative reporting to REC and codify trial risk levels |
Implications for CTU teams and trial managers
CTU teams must shift from documenting every event to documenting justified, risk-proportionate information as required by the April 28, 2026 regulations, according to PLOS One and MHRA guidance.
- Action: map who in your team makes causality and reporting decisions, and document decision rules to increase transparency and defendability, as recommended by PLOS One.
- Action: standardise SOP language across similar trials to avoid site burden and double data entry that participants described in PLOS One.
- Action: implement role-specific, bite-sized training and example-based templates because PLOS One participants recommended practical, illustrative training to build confidence.
How Evidano helps CTUs operationalise qualitative findings
Problem: No shared visibility into past decisions
Solution: Use Evidence synthesis to centralise transcripts, decision notes, and templates so teams can review past justifications quickly.
Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. For CTUs this means ingesting meeting transcripts, investigator notes, and case report narratives to produce searchable themes and decision trails.
Example: upload cleaned focus group transcripts and use Evidano to generate thematic clusters and extract verbatim decision quotes to include in SOPs and investigator brochures, reducing repeated interpretation work.
Problem: Transcription and cleaning overhead
Solution: Use automated transcription with domain-specific dictionaries to preserve drug names and redact PII.
Evidano offers transcription and redaction features that match the workflow used in the PLOS One study, and supports custom dictionaries and PII redaction to speed transcript cleaning while keeping data safe. See Speech to text for details.
Problem: Slow thematic synthesis across trials
Solution: Apply AI-enabled thematic, frequency, and cross-segment analyses to identify which low-grade harms recur and which reports are noise.
Evidano can produce hierarchical codes and cross-segment comparisons so CTUs can see, for example, whether grade 1 and 2 toxicities are clustered by site, regimen, or patient subgroup and decide whether they require reporting or PROM collection. Learn about features at Evidano features.
Problem: Teams need quick Q&A over documents
Solution: Use AI chat over your documents to ask direct questions about prior decisions and quoted rationales.
Evidano’s AI chat lets trial teams ask: "When did we decide to record X, and who justified that? " and returns quotes, timestamps, and coded themes to accelerate defensible reporting choices. See AI chatbot.
FAQ: qualitative analysis of safety reporting
What is qualitative analysis of safety reporting and why does it matter?
Answer: Qualitative analysis of safety reporting is the systematic study of staff accounts, decision notes, and free-text reports to identify themes about how and why reporting choices are made.
Support: The PLOS One study (published July 30, 2026) used Reflexive Thematic Analysis on 23 participants to surface the dominant theme of balancing risk, showing how qualitative methods reveal drivers of over-reporting and under-reporting that statistics alone miss.
How can AI make qualitative safety analysis faster and more reliable?
Answer: AI speeds transcription, codes themes consistently, and finds co-occurrence patterns across large text sets, reducing manual time for synthesis.
Support: The PLOS One team used auto-transcription in Microsoft Teams and manual cleaning; replacing repetitive cleanup and code application with AI workflows frees analyst time for interpretation and training design.
What immediate steps should a CTU take after reading the PLOS One findings?
Answer: Start by documenting current decision rules, collect recent decision notes and transcripts, and run a rapid qualitative audit to find common friction points.
Support: PLOS One participants recommended clarifying expectations, improving transparency, and sharing templates. A rapid audit of 4 to 8 recent SAE decisions will quickly show whether reporting is consistent across trials.
Are there privacy or regulatory risks to using AI on safety data?
Answer: Yes, privacy and regulatory safeguards are essential when processing safety reports, so use platforms with PII redaction, encryption, and clear data use policies.
Support: The PLOS One study pseudonymised transcripts before analysis, and CTUs should likewise ensure data is de-identified and compliant with applicable law when using AI tools.
Conclusion & Next Steps
The PLOS One study published July 30, 2026, shows that CTU staff experience safety reporting as a balancing act driven by uncertainty, bureaucracy, inconsistent expectations, and training gaps.
Applied qualitative analysis, enhanced with AI transcription and thematic tools, can make reporting decisions more transparent, less duplicative, and faster to justify to regulators and sponsors.
If your team wants to pilot a rapid qualitative audit of recent SAE decisions and generate shareable templates and evidence trails, try an AI-enabled workflow that handles transcription, coding, and searchable decision logs.
To get started with a free trial, Try Evidano for free.