Contents
On this page
- Purposes of a qualitative strand
- Prevalence of embedded qualitative research
- Qualitative diagnosis of recruitment failure
- Process evaluation and its distinction from fidelity measurement
- Qualitative work in feasibility and pilot studies
- Designing a deliverable qualitative strand
- Findings accessible only through qualitative methods
- Reporting the qualitative strand
- Common failures of embedded qualitative research
- Key points
- Application to your study
- Frequently asked questions
Academy · Research Data Analytics & AI · Qualitative work in trials
Qualitative Research in Clinical Trials: Purposes, Design and Reporting
Qualitative research within a trial is neither a token addition nor a patient-experience appendix. When conducted rigorously, it addresses questions that the randomised comparison cannot: why eligible patients decline, what clinicians say in the recruitment consultation, and whether the intervention delivered was the intervention designed. On the published record, however, it is specified so thinly in most trial applications that it could not deliver any of these.
Two findings frame this subject. First, embedded qualitative research is rare: in a national randomised controlled trial database of 3,812 trials, only 89 (2%) described qualitative research undertaken with the trial. Second, where it is proposed, the proposal is often nominal: among the funded applications examined, a quarter devoted less than a single paragraph to it. By contrast, the one application supported by a strong evidence base, qualitative work to diagnose and remedy recruitment difficulties, has a record that compares favourably with most trial methods. This article examines the gap between that potential and current practice.
Learning objectives
- The three questions inside a trial that only qualitative methods answer well, and the questions they answer badly
- What the QuinteT Recruitment Intervention does, and what its evaluations showed
- How a qualitative strand should be specified in a protocol or grant application so that it can be delivered rather than merely proposed
- The timing of qualitative work (before, during or after the randomised phase) and the costs associated with each timing
- Which reporting standard applies to the qualitative strand, and what the 32 COREQ items ask for
- Where embedded qualitative research tends to fail, including the one setting where a well-tested intervention performed less well
Common misconceptions
- “Qualitative work is for exploring patient experience.” It does so, but its highest-value use in trials is operational diagnosis: establishing why recruitment is failing, what is said in consultations, and whether the intervention delivered matches the one designed. The recruitment application has the strongest supporting evidence of any embedded qualitative method.
- “A paragraph in the protocol is enough to commission it.” Among funded trial proposals examined, a quarter devoted less than a single paragraph to the qualitative component, only 56% gave an explicit rationale for it, and only 28% stated that qualitative researchers would conduct it. A strand specified at that level cannot be held to any standard afterwards.
- “Any member of the trial team can run the interviews.” Interviewing is a method that requires specific training, and analysis requires more. The proposal data suggest that the gap is widest here: fewer than a third of proposals stated who would conduct the work.
- “Qualitative findings are not generalisable, so they cannot change the trial.” Qualitative findings change trials routinely and measurably. All five UK publicly funded randomised trials that applied the QuinteT Recruitment Intervention recruited to target, with significant improvement in randomisation rates after intervention in three of them.
- “Reporting it is a matter of writing up some quotes.” COREQ sets out 32 reporting items for interview and focus-group studies, covering the research team, the methods, the sampling, the analysis and the relationship between data and interpretation. Quotes are one small part of item-level reporting.
Purposes of a qualitative strand
Three question types inside a trial are poorly served by quantitative data and well served by qualitative method.
Why the trial is not recruiting. A screening log may record that 300 eligible patients were approached and 40 randomised. It does not record what was said, which arm sounded less favourable when described aloud, whether the clinician conveyed equipoise, or whether the word “randomised” was understood as “experimental”. Audio-recorded consultations, interviews with recruiters and interviews with patients who declined address those questions, and no other method does 11,17.
Whether the intervention delivered was the intervention designed. Fidelity measured as a checklist proportion indicates how often components were ticked. It does not indicate what the components became in practice — a structured exercise programme implemented as unsupervised advice, a decision aid handed over rather than discussed. The UK MRC process evaluation guidance, written for the Population Health Sciences Research Network by 11 researchers from 8 universities and running to 134 pages across theory, practice and case studies, exists because this problem is both common and tractable 24.
What the outcome measure is measuring. Where an outcome is patient-reported, qualitative work can establish whether participants interpret the items as the developers intended — a question that psychometric validation addresses statistically and qualitative work addresses directly. Our pages on developing a PRO instrument and PRO instrument types cover that interface.
Qualitative work does not estimate effects, establish prevalence or substitute for a comparison group. A qualitative strand that reports how many participants expressed a view usually indicates that the design question was quantitative from the outset.
Prevalence of embedded qualitative research
The scale of non-use can be stated precisely. In a national randomised controlled trial database containing 3,812 trials, 89 — 2% — described qualitative research undertaken with the trial 22. Whatever its merits, the method is not in routine use.
The quality of the proposals that do include it is the more actionable finding. Of 36 funded proposals received for examination, 32 were included 3,35. A quarter devoted less than a single paragraph to the qualitative component. Only 56% gave an explicit rationale for including it, and only 28% stated that qualitative researchers would conduct it.
The last figure warrants particular attention, because it determines what is delivered. In nearly three quarters of proposals, no one was named to conduct the work. This is not merely a documentation defect; it is a resourcing decision taken by default, and the predictable outcome is interviews conducted by whoever has time, analysed thematically by one person and reported as a handful of illustrative quotations in the discussion section. Such a strand consumes budget without producing anything that could change the trial.
A separate finding points in the same direction. The qualitative strand of a systematic review of decentralised trial methods analysed 117 documents and produced four themes, but found participant and stakeholder experience incompletely represented across the literature; even where qualitative material exists, the perspectives it was intended to capture are unevenly represented 4.
Use and specification of embedded qualitative research. The 2% figure comes from a national randomised controlled trial database of 3,812 trials, in which 89 described qualitative research undertaken with the trial. The proposal figures come from 32 included proposals of 36 received.
Qualitative diagnosis of recruitment failure
The QuinteT Recruitment Intervention is the best-evidenced embedded qualitative method in trials. It was developed further across 13 randomised controlled trials, operates in two major phases — understanding the recruitment problem, then acting on it — and can be integrated into a feasibility, pilot or main trial phase 3,17. The first phase draws on audio-recorded recruitment consultations, interviews with recruiting staff and patients, screening log analysis and documentation review; the second feeds findings back as targeted, often individualised, actions 10,17.
Its evaluation record is unusually concrete for a trial method. All five UK publicly funded randomised controlled trials that applied the intervention recruited to target 11. Randomisation rates improved significantly after action in three of them, and significantly more patients were approached in two. For a method frequently described as exploratory, recruitment to target in all five trials is a notable result.
A streamlined version, QRI-Two, was developed from QuinteT interventions in 12 randomised trials run between 2007 and 2017, then applied to 20 randomised trials in the USA and Europe in 2018 and 2019 5,14,35. It identified previously unknown or underestimated recruitment barriers in 14 of those trials and rapidly confirmed feasibility in 4. It performed less well in 6 trials that were at the start-up stage, which is the most practically useful finding in this literature for planning when to deploy it. The method requires recruitment behaviour to observe, and at start-up there is none.
Where recruitment is the principal problem, this is the appropriate starting point, and the trial support and protocol stages are where it should be specified.
The QuinteT Recruitment Intervention: development and evaluation record
| Study | What was done | What was found |
|---|---|---|
| Method development | Developed further across 13 randomised controlled trials; operates in two major phases and can be integrated into a feasibility, pilot or main trial phase | Phase one understands the recruitment problem; phase two feeds back targeted actions |
| Evaluation in UK publicly funded trials | Five UK publicly funded randomised controlled trials that applied the intervention | All five recruited to target; significant improvement in randomisation rates after action in three; significantly more patients approached in two |
| QRI-Two development | Streamlined from QuinteT interventions in 12 randomised trials run between 2007 and 2017 | Produced a shorter form intended for wider deployment |
| QRI-Two application | Applied to 20 randomised trials in the USA and Europe during 2018 and 2019 | Previously unknown or underestimated barriers identified in 14 trials; feasibility rapidly confirmed in 4; performed less well in 6 trials that were at start-up stage |
Compiled from the QuinteT Recruitment Intervention development and evaluation literature. The start-up finding is the practical constraint: the method needs recruitment activity to observe.
Process evaluation and its distinction from fidelity measurement
Process evaluation examines what happened in practice: how the intervention was implemented, what mechanisms it worked through, and how context shaped both 24,28. The UK MRC guidance — 134 pages across theory, practice and case studies, written by 11 researchers from 8 universities — is the reference document, and its central argument is that implementation, mechanism and context are three distinct questions that a single fidelity percentage collapses into one number 24.
The distinction matters for interpretation. A trial with a null result and high measured fidelity invites the conclusion that the intervention does not work. A process evaluation may instead show that the intervention was delivered as specified but did not engage the mechanism it was designed to engage, or that it engaged it in one setting and not another. Those are different conclusions with different implications for what to do next, and the trial result alone cannot distinguish them.
Two practical points follow. Process evaluation data must be collected during the trial, which means that the design must exist before the first participant is enrolled; retrospective reconstruction yields recollection, not process 24,29. Its analysis should also be separated from the outcome analysis in time or in team, because knowledge of the trial result alters how implementation data are read. The TIDieR checklist for intervention description is the natural companion: a process evaluation can only assess delivery against a specification that was written down.
Qualitative work in feasibility and pilot studies
Feasibility studies are where qualitative work most readily justifies its place, because the question a feasibility study asks — can this trial be run, and should it be — is not primarily a question about effect sizes. Published guidance on qualitative research within feasibility studies for randomised controlled trials comprises 16 items across five domains: research questions, data collection, analysis, teamwork and reporting 21.
The teamwork domain is the one most often omitted, and the proposal data suggest that it matters most, given that only 28% of examined proposals stated who would conduct the qualitative work 21,22. The items in that domain concern who is involved, how qualitative and quantitative team members communicate, and how disagreements between the two kinds of evidence are resolved, which is an organisational question with a methodological consequence.
The research questions domain is the next most consequential. A feasibility study with a qualitative strand and no stated qualitative research questions will produce general impressions, which cannot inform a go or no-go decision. A useful formulation names the specific uncertainty: whether recruiters can describe the arms in a balanced way, whether the outcome measure can be completed at home, and whether the control condition is acceptable to clinicians asked to deliver it.
Guidance domains for qualitative research within feasibility studies
| Domain | What the items address |
|---|---|
| Research questions | Whether the qualitative questions are stated, and whether they bear on the feasibility decision |
| Data collection | Methods, sampling, timing relative to the randomised element |
| Analysis | Named approach, number of analysts, resolution of disagreement |
| Teamwork | Who is involved, how qualitative and quantitative team members communicate, how conflicting evidence is handled |
| Reporting | What is reported, where, and against which standard |
The published guidance comprises 16 items across these five domains. Domain names as given; the item detail sits in the source guidance.
Designing a deliverable qualitative strand
Six decisions determine whether an embedded qualitative strand produces usable findings. All of them belong in the protocol.
The question. Stated as a question, not a topic. “Patient experience of the intervention” is a topic. “How do participants decide whether to continue the home monitoring regimen after the first month” is a question, and it determines sampling and timing.
Sampling. Purposive against a stated frame, and stated in advance: which participants, from which arms, at which sites, selected on what dimensions of variation. Interviewing whichever participants agreed to be interviewed produces a convenience sample reported as a purposive one.
Timing relative to randomisation. Before randomisation for decision-making and equipoise; during the trial for implementation and adherence; afterwards for interpretation of outcomes. Each timing has a cost. Interviewing during the trial risks influencing behaviour in a way that affects the outcome; interviewing only afterwards yields recollection filtered through knowledge of the result.
Who conducts it. Named roles with relevant training, and someone who is not the treating clinician. This is the decision three quarters of proposals leave unstated.
Analytic approach. Named — thematic analysis, framework, grounded theory — with how many analysts, whether independently, and how disagreement is resolved. “Thematic analysis will be used” names a family, not a method.
How it integrates with the outcome analysis. Whether the qualitative findings are reported alongside, within or separately from the main results, and what happens if the two disagree. A mixed-methods implementation study of a risk-based monitoring approach in an academic trial unit is a worked example of integration done deliberately rather than by juxtaposition 14.
Our protocol writing and SPIRIT pages cover where each of these sits in the document.
The qualitative work package in a trial protocol or grant application (figures below are illustrative)
A paragraph that commissions nothing
An embedded qualitative study will explore the experiences and views of participants and staff involved in the trial. Semi-structured interviews will be conducted with a sample of participants and with members of the clinical team, and will be analysed using thematic analysis. The findings will provide important context for the interpretation of the trial results and will inform the implementation of the intervention in routine practice.
No research question. “Experiences and views” is a topic. It cannot determine who to sample, when to interview or when to stop, so every later decision becomes ad hoc.
No sampling frame. “A sample of participants” permits anyone who volunteers, which is a convenience sample that will be reported as purposive.
Nobody named. Only 28% of examined proposals stated that qualitative researchers would conduct the work; this paragraph is in the other 72%, and the predictable outcome is interviews done by whoever has capacity.
No analytic detail. “Thematic analysis” names a family of methods. Number of analysts, independence and disagreement resolution are all unspecified.
No route from finding to action. “Will provide important context” commits nobody to doing anything, so the strand cannot change the trial while the trial is still running.
No timing relative to unblinding. If the qualitative analysis happens after the result is known, the themes will drift toward explaining it.
A work package that can be delivered
The embedded qualitative study addresses two questions: how recruiting clinicians describe the two arms in the recruitment consultation, and why eligible patients decline randomisation. Data comprise up to 60 audio-recorded recruitment consultations across all six sites, sampled to include every recruiting clinician and over-sampling the two sites with the lowest randomisation ratio; interviews with all recruiting staff at months 3 and 9; and interviews with up to 25 patients who declined, approached by a researcher unconnected to their care within two weeks of the consultation. Sampling continues until no new barriers are identified across two consecutive site reviews, and the stopping point will be reported. Data are collected and analysed by a named qualitative researcher (0.5 FTE for 24 months) with a second analyst independently coding a 20% subsample; disagreements are resolved by discussion with a third team member and recorded. Analysis uses framework analysis against an initial matrix derived from the recruitment literature, revised inductively. Findings are fed back to the trial management group at months 6 and 12 as specific, site-level actions, and the feedback and any resulting changes are documented as part of the trial record. The qualitative strand will be reported against COREQ in a companion paper, with a summary in the main trial report; the analysis of consultation data will be completed before the outcome data are unblinded.
Two specific questions determine the whole design, which is why sampling, timing and the stopping rule can all be stated in advance.
The sampling frame is explicit and deliberately over-samples the sites where the problem is worst, which is where the informative variation lives.
A named researcher with an FTE allocation converts the strand from an aspiration into a resourced work package.
The analytic approach, the second coder, the subsample proportion and the disagreement procedure are all stated, so the analysis is assessable.
Feedback points and documented actions give the findings a route into the running trial — the mechanism by which QRI-type work improves recruitment.
Completing consultation analysis before unblinding protects the interpretation, and naming COREQ sets the reporting standard before anyone drafts.
A reviewer assessing the first version has nothing to assess and no basis on which to fund it at any particular level. The second version states what will be collected, by whom, with what budget, and what will be done with the findings, and it is also the only version that could still be delivered as written in year two.
Findings accessible only through qualitative methods
A concrete example from the diagnostics literature. A two-country qualitative study of home- and remote-based medical testing interviewed 47 people — 37 patients, 2 caregivers and 8 healthcare professionals — and ran a validation workshop with 32 more, comprising 13 patients, 7 caregivers and 12 healthcare professionals 3,5,17,22,35. It produced five themes across two countries. Among its findings: convenience did not influence patients’ decisions in the way the framing of remote testing assumes 3.
This illustrates the form of a genuinely useful qualitative finding: it contradicts a premise that a quantitative study would have assumed. A survey asking patients to rate the importance of convenience would have returned a number, and that number would have been uninformative about whether convenience influenced the actual decision. See the Academy page on home diagnostics in trials for the implications for decentralised designs.
In general, qualitative work is most valuable where it can falsify an assumption built into the trial design. It is least valuable where it is used to illustrate or confirm a result the trial has already produced, which is the use to which it is most often put.
Reporting the qualitative strand
COREQ — the consolidated criteria for reporting qualitative research, published in 2007 — sets out 32 reporting items for interview and focus-group studies, and it is the standard applicable to the qualitative strand of a trial 3,22,35. The items cover the research team and their reflexivity, the study design including theoretical framework and sampling, the setting and data collection, and the analysis and findings, including how themes were derived and how quotations are identified.
Two items are routinely missed in trial-embedded work. The first is the relationship between the researcher and the participants — whether the interviewer was involved in the participant’s care or in the trial’s delivery, which bears directly on what a participant will say about a trial they are in 35. The second is data saturation or the stopping rule: why interviewing stopped when it did. Neither is onerous to report, and both are among the first items a methodological reviewer examines.
Where the qualitative strand is reported as part of the main trial publication, space is the constraint and the usual outcome is an item-light account with quotations. A better approach is a short, transparent account in the main paper with a complete COREQ-reported companion paper, cross-referenced. Our COREQ checklist page and SRQR checklist set out the items; CONSORT 2025 governs the trial report itself.
Common failures of embedded qualitative research
Four failure modes are described below, three of them avoidable.
Deployed too early. QRI-Two performed less well in 6 of the 20 trials to which it was applied, and those were trials at the start-up stage 5. A method that analyses recruitment behaviour needs recruitment behaviour to analyse. The same logic applies to process evaluation before any implementation has occurred.
Unresourced. A quarter of examined proposals gave the qualitative component less than a paragraph and fewer than a third named qualitative researchers to conduct it 22. A strand with no named analyst and no stated analytic approach does not fail conspicuously; it produces a thin thematic account on which no one can act, which is harder to detect than an outright failure.
Analysed after the result is known. Once the trial outcome is known, implementation data are read differently and themes drift toward explaining the result. The analyses should be separated in time or in team.
Asked to do a quantitative job. A qualitative strand cannot establish how common a view is, and reporting counts of participants holding a theme invites exactly that misreading. If the question concerns prevalence, a different design is required; see the pages on cohort studies and epidemiological designs for trials.
Key points
Key points
Embedded qualitative research addresses three questions a trial cannot: why recruitment is failing, whether the intervention delivered matches the one designed, and what the outcome measure is capturing. It is rare (2% of 3,812 trials in a national database described any) and, where proposed, it is often nominal, with a quarter of funded proposals allocating it less than a paragraph and only 28% naming qualitative researchers to conduct it. The best-evidenced application is recruitment diagnosis: all five UK publicly funded trials applying the QuinteT Recruitment Intervention recruited to target, with significant improvement in randomisation rates in three, and the streamlined QRI-Two identified previously unknown barriers in 14 of 20 trials, although it performed less well in the 6 that were at start-up, because the method requires recruitment behaviour to observe. Protocols should specify the question, the sampling frame, the named analyst, the analytic approach, the feedback route and the timing relative to unblinding, and reporting should follow COREQ’s 32 items.
Application to your study
Application to your study
- The qualitative research question should be formulated as a question. If it reads as a topic, such as experiences, views or perceptions, it will not determine sampling or timing, and subsequent decisions will be improvised.
- The researcher and the full-time equivalent (FTE) allocation should be named in the application. This is the single most common omission and the one that determines whether the work is delivered.
- The sampling frame and the stopping rule should be stated before data collection begins. The statement should specify which participants, from which arms and sites, are selected on which dimensions, and what condition ends data collection.
- The timing should be decided relative to randomisation and to unblinding. Qualitative work should be conducted before randomisation for questions of equipoise and decision-making and during the trial for implementation; in all cases, the analysis of process data should, where possible, be completed before the outcome is known.
- A feedback route should be built into the trial management structure. The route should specify named meetings, named months and documentation of what changed. Findings that arrive after the trial closes cannot remedy its problems.
- Where recruitment is the problem, qualitative work should be deployed once recruitment is under way. QRI-Two performed less well at trial start-up because there was no recruitment behaviour to analyse.
- The strand should be mapped against COREQ’s 32 items during drafting rather than at submission. Particular attention should be paid to the researcher-participant relationship and the stopping rule, the two items most often missing.
Frequently asked questions
Uncommon. In a national randomised controlled trial database of 3,812 trials, 89 — 2% — described qualitative research undertaken with the trial. Where it is proposed, the specification is often minimal: of 32 funded proposals examined, a quarter devoted less than a single paragraph to the qualitative component, only 56% gave an explicit rationale for its inclusion, and only 28% stated that qualitative researchers would conduct it. The overall picture is therefore of a method with good evidence behind specific applications that is rarely used and, when used, often under-specified to the point where it cannot deliver what it was included to deliver.
It is a qualitative method for diagnosing and remedying recruitment difficulty in randomised trials, developed further across 13 trials and operating in two major phases: understanding the recruitment problem using audio-recorded consultations, staff and patient interviews and screening data, then feeding findings back as targeted actions. The evaluation record is strong by the standards of trial methodology. All five UK publicly funded randomised controlled trials that applied it recruited to target, with significant improvement in randomisation rates after action in three and significantly more patients approached in two. The streamlined QRI-Two, applied to 20 trials in the USA and Europe in 2018 and 2019, identified previously unknown or underestimated barriers in 14 of them.
The appropriate timing depends on the question, and each timing has costs. Questions about decision-making, equipoise and consent require data from before or around randomisation. Questions about implementation and adherence require data collected during delivery, because retrospective accounts yield recollection, not process. Questions about how participants interpreted an outcome measure can be addressed later. One timing rule holds generally: process and consultation data should be analysed before the outcome is unblinded where possible, because knowledge of the result changes how implementation data are read. A recruitment-diagnosis method should not be deployed at start-up: QRI-Two performed less well in exactly the 6 trials where it was applied before recruitment behaviour existed to observe.
COREQ, the consolidated criteria for reporting qualitative research, published in 2007, with 32 items covering the research team and reflexivity, study design and sampling, setting and data collection, and analysis and findings. It is the applicable standard for interview and focus-group studies, which covers most trial-embedded work. Two items are missed most often in this setting: the relationship between researcher and participant — specifically whether the interviewer was involved in the participant’s care or the trial’s delivery — and the stopping rule for data collection. Where space in the main trial paper is limited, a workable approach is a short, transparent summary there and a fully COREQ-reported companion paper, cross-referenced from both.
Yes, and that is the application with the best evidence. The purpose of a recruitment intervention is to feed findings back during recruitment so that they alter what happens next, which is why the evaluations report improved randomisation rates after action rather than merely a better understanding of why recruitment was poor. The requirement is structural: a protocol-specified feedback route, named points in the trial calendar at which findings go to the trial management group, and documentation of what changed as a result. Without that route, the findings appear in a paper after the trial closes, when they can serve only as a retrospective account.
No number holds across studies, and a proposal that states one without a basis offers an unsupported figure. A reviewer looks for a defensible stopping rule stated in advance: data collection continues until no new barriers or themes emerge across a defined review interval, with the stopping point reported and justified. In practice, sampling adequacy depends on the heterogeneity of the setting and the narrowness of the question: a single-site study of one consultation type reaches adequacy far sooner than a multi-country study of implementation across differing health systems. Authors should state a planned maximum for budgeting and a stopping condition for method, and should report what actually occurred.
Fidelity measurement asks how much of the specified intervention was delivered, usually as a proportion of components completed. Process evaluation asks three separate questions — how the intervention was implemented, what mechanisms it worked through, and how context shaped both — and the UK MRC guidance, running to 134 pages across theory, practice and case studies and written by 11 researchers from 8 universities, exists because collapsing those three into one number loses the information needed to interpret a trial. A trial can have high fidelity and a null result because the delivered intervention did not engage its intended mechanism, which is a different conclusion from the intervention not working, and only a process evaluation distinguishes them.
No; designing it as though it does is a common error. Qualitative sampling is purposive, not representative, so the number of participants expressing a view is a property of who was sampled rather than of the population. Reporting theme counts invites the misreading. Qualitative work is well suited to identifying that a view or a barrier exists, describing its structure and reasoning, and falsifying assumptions built into the trial design — for example, the two-country study of home and remote testing, which interviewed 47 people and ran a validation workshop with 32 more, and found that convenience did not influence patients’ decisions in the way remote-testing arguments assume. If prevalence is the question, a design built to estimate it should be used.
References
- O'Brien BC, Harris IB, Beckman TJ, Reed DA, Cook DA. Standards for reporting qualitative research: a synthesis of recommendations. Academic Medicine. 2014;89(9):1245-1251. read 9 October 2026. equator-network.org
- de Viron S, Trotta L, Steijn W, Young S, Buyse M. Does Central Statistical Monitoring Improve Data Quality? An Analysis of 1,111 Sites in 159 Clinical Trials. Therapeutic Innovation & Regulatory Science. 2024;58(3):483-494. doi:10.1007/s43441-024-00613-w read 9 October 2026. doi.org/10.1007/s43441-024-00613-w
- Lalanza S, Pena C, Bezos C, Yamauchi N, Taffner V, Rodrigues K, Garcia Manrique M, Rubio Jareno A, Lemos Gil A. Patient and Healthcare Professional Insights of Home- and Remote-Based Clinical Assessment: A Qualitative Study from Spain and Brazil to Determine Implications for Clinical Trials and Current Practice. Advances in Therapy. 2023;40(4):1670-1685. doi:10.1007/s12325-023-02441-0 read 9 October 2026. doi.org/10.1007/s12325-023-02441-0
- Rogers A, De Paoli G, Subbarayan S, Copland R, Harwood K, Coyle J, et al. A Systematic Review of Methods used to Conduct Decentralised Clinical Trials. British Journal of Clinical Pharmacology. 2022;88(6):2843-2862. doi:10.1111/bcp.15205 read 9 October 2026. doi.org/10.1111/bcp.15205
- Donovan JL, Jepson M, Rooshenas L, Paramasivan S, Mills N, Elliott D, Wade J, Reda D, Blazeby JM, Moghanaki D, Hwang ES, Davies L. Development of a new adapted QuinteT Recruitment Intervention (QRI-Two) for rapid application to RCTs underway with enrolment shortfalls to identify previously barriers and improve recruitment. Trials. 2022;23(1):1-14. doi:10.1186/s13063-022-06187-y read 9 October 2026. doi.org/10.1186/s13063-022-06187-y
- de Viron S, Trotta L, Steijn W, Young S, Buyse M. Does Central Monitoring Lead to Higher Quality? An Analysis of Key Risk Indicator Outcomes. Therapeutic Innovation & Regulatory Science. 2022;57(2):295-303. doi:10.1007/s43441-022-00470-5 read 9 October 2026. doi.org/10.1007/s43441-022-00470-5
- de Jong AJ, Grupstra RJ, Santa-Ana-Tellez Y, Zuidgeest MGP, de Boer A, Gardarsdottir H. Which decentralised trial activities are reported in clinical trial protocols of drug trials initiated in 2019-2020? A cross-sectional study in ClinicalTrials.gov. BMJ Open. 2022;12(8):e063236. doi:10.1136/bmjopen-2022-063236 read 9 October 2026. doi.org/10.1136/bmjopen-2022-063236
- Cragg WJ, Hurley C, Yorke-Edwards V, Stenning SP. Dynamic methods for ongoing assessment of site-level risk in risk-based monitoring of clinical trials: A scoping review. Clinical Trials. 2021;18(2):245-259. doi:10.1177/1740774520976561 read 9 October 2026. doi.org/10.1177/1740774520976561
- Davis K, Minckas N, Bond V, et al. Beyond interviews and focus groups: a framework for integrating innovative qualitative methods into randomised controlled trials of complex public health interventions. Trials. 2019;20:329. doi:10.1186/s13063-019-3439-8 read 9 October 2026. doi.org/10.1186/s13063-019-3439-8
- Rooshenas L, Paramasivan S, Jepson M, Donovan JL, et al. Intensive Triangulation of Qualitative Research and Quantitative Data to Improve Recruitment to Randomized Trials: The QuinteT Approach. Qualitative Health Research. 2019;29(5):672-679. doi:10.1177/1049732319828693 read 9 October 2026. doi.org/10.1177/1049732319828693
- Rooshenas L, Scott LJ, Blazeby JM, Rogers CA, Tilling KM, Husbands S, Conefrey C, Mills N, Stein RC, Metcalfe C, Carr AJ, Beard DJ, Davis T, Paramasivan S, Jepson M, Avery K, Elliott D, Wilson C, Donovan JL. The QuinteT Recruitment Intervention supported five randomized trials to recruit to target: a mixed-methods evaluation. Journal of Clinical Epidemiology. 2019;106:108-120. doi:10.1016/j.jclinepi.2018.10.004 read 9 October 2026. doi.org/10.1016/j.jclinepi.2018.10.004
- Agrafiotis DK, Lobanov VS, Farnum MA, Yang E, Ciervo J, Walega M, et al. Risk-based monitoring of clinical trials: an integrative approach. Clinical Therapeutics. 2018;40(7):1204-1212. read 9 October 2026. journals.sagepub.com
- Stenning SP, Cragg WJ, Joffe N, Diaz-Montana C, Choudhury R, Sydes MR, et al. Triggered or routine site monitoring visits for randomised controlled trials: results of TEMPER, a prospective, matched-pair study. Clinical Trials. 2018;15:600-609. read 9 October 2026. journals.sagepub.com
- von Niederhausern B, Orleth A, Schadelin S, Rawi N, Velkopolszky M, Becherer C, et al. Generating evidence on a risk-based monitoring approach in the academic setting - lessons learned. BMC Medical Research Methodology. 2017;17(1):26. doi:10.1186/s12874-017-0308-6 read 9 October 2026. doi.org/10.1186/s12874-017-0308-6
- Brosteanu O, Schwarz G, Houben P, Paulus U, Strenge-Hesse A, Zettelmeyer U, et al. Risk-adapted monitoring is not inferior to extensive on-site monitoring: results of the ADAMON cluster-randomised study. Clinical Trials. 2017;14:584-596. read 9 October 2026. journals.sagepub.com
- Buyse M, Squifflet P, Coart E, Quinaux E, Punt CJA, Saad ED. The impact of data errors on the outcome of randomized clinical trials. Clinical Trials. 2017;14(5):499-506. read 9 October 2026. journals.sagepub.com
- Donovan JL, Rooshenas L, Jepson M, et al. Optimising recruitment and informed consent in randomised controlled trials: the development and implementation of the Quintet Recruitment Intervention (QRI). Trials. 2016;17:283. doi:10.1186/s13063-016-1391-4 read 9 October 2026. doi.org/10.1186/s13063-016-1391-4
- Hurley C, Shiely F, Power J, Clarke M, Eustace JA, Flanagan E, Kearney PM. Risk based monitoring (RBM) tools for clinical trials: a systematic review. Contemporary Clinical Trials. 2016;51:15-27. read 9 October 2026. journals.sagepub.com
- Knepper D, Lindblad AS, Sharma G, Gensler GR, Manukyan Z, Matthews AG, Ahn H. Statistical monitoring in clinical trials: best practices for detecting data anomalies suggestive of fabrication or misconduct. Therapeutic Innovation & Regulatory Science. 2016;50:144-154. read 9 October 2026. journals.sagepub.com
- Olsen R, Bihlet AR, Kalakou F, Andersen JR. The impact of clinical trial monitoring approaches on data integrity and cost - a review of current literature. European Journal of Clinical Pharmacology. 2016;72(4):399-412. read 9 October 2026. journals.sagepub.com
- O'Cathain A, Hoddinott P, Lewin S, et al. Maximising the impact of qualitative research in feasibility studies for randomised controlled trials: guidance for researchers. Pilot and Feasibility Studies. 2015;1:32. doi:10.1186/s40814-015-0026-y read 9 October 2026. doi.org/10.1186/s40814-015-0026-y
- Drabble SJ, O'Cathain A, Thomas KJ, Rudolph A, Hewison J. Describing qualitative research undertaken with randomised controlled trials in grant proposals: a documentary analysis. BMC Medical Research Methodology. 2014;14:24. doi:10.1186/1471-2288-14-24 read 9 October 2026. doi.org/10.1186/1471-2288-14-24
- Sheetz N, Wilson B, Benedict J, Huffman E, Lawton A, Travers M, et al. Evaluating source data verification as a quality control measure in clinical trials. Therapeutic Innovation & Regulatory Science. 2014;48(6):671-680. read 9 October 2026. journals.sagepub.com
- Moore G, Audrey S, Barker M, Bond L, Bonell C, Hardeman W, Moore L, O'Cathain A, Tinati T, Wight D, Baird J. Process Evaluation of Complex Interventions Guidance: UK Medical Research Council (MRC) Guidance. MRC Population Health Sciences Research Network, London. 2014:134 pp. read 9 October 2026. eprints.gla.ac.uk
- Tudur Smith C, Williamson P, Jones A, Smyth A, Hewer SL, Gamble C. Risk-proportionate clinical trial monitoring: an example approach from a non-commercial trials unit. Trials. 2014;15:1-10. read 9 October 2026. journals.sagepub.com
- Macefield RC, Beswick AD, Blazeby JM, Lane JA. A systematic review of on-site monitoring methods for health-care randomised controlled trials. Clinical Trials. 2013;10(1):104-124. read 9 October 2026. journals.sagepub.com
- Kirkwood AA, Cox T, Hackshaw A. Application of methods for central statistical monitoring in clinical trials. Clinical Trials. 2013;10(5):783-806. read 9 October 2026. journals.sagepub.com
- Grant A, Treweek S, Dreischulte T, et al. Process evaluations for cluster-randomised trials of complex interventions: a proposed framework for design and reporting. Trials. 2013;14:15. doi:10.1186/1745-6215-14-15 read 9 October 2026. doi.org/10.1186/1745-6215-14-15
- Chan AW, Tetzlaff JM, Altman DG, Laupacis A, Gotzsche PC, Krleza-Jeric K, Hrobjartsson A, Mann H, Dickersin K, Berlin JA, Dore CJ, Parulekar WR, Summerskill WSM, Groves T, Schulz KF, Sox HC, Rockhold FW, Rennie D, Moher D. SPIRIT 2013 Statement: Defining Standard Protocol Items for Clinical Trials. Annals of Internal Medicine. 2013;158(3):200-207. doi:10.7326/0003-4819-158-3-201302050-00583 read 9 October 2026. doi.org/10.7326/0003-4819-158-3-201302050-00583
- Bakobaki J, Joffe N, Burdett S, Tierney J, Meredith S, Stenning S. A systematic search for reports of site monitoring technique comparisons in clinical trials. Clinical Trials. 2012;9(6):777-780. read 9 October 2026. journals.sagepub.com
- Tudur Smith C, Stocken DD, Dunn J, Cox T, Ghaneh P, Cunningham D, Neoptolemos JP. The value of source data verification in a cancer clinical trial. PLoS ONE. 2012;7(12):e51623. read 9 October 2026. journals.sagepub.com
- Journot V, Pignon JP, Gaultier C, Daurat V, Bouxin-Metro A, Giraudeau B, et al. Validation of a risk-assessment scale and a risk-adapted monitoring plan for academic clinical research studies - the Pre-Optimon study. Contemporary Clinical Trials. 2011;32:16-24. read 9 October 2026. journals.sagepub.com
- Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. Trials. 2010;11(1):32. doi:10.1186/1745-6215-11-32 read 9 October 2026. doi.org/10.1186/1745-6215-11-32
- Brosteanu O, Houben P, Ihrig K, Ohmann C, Paulus U, Pfistner B, et al. Risk analysis and risk adapted on-site monitoring in noncommercial clinical trials. Clinical Trials. 2009;6(6):585-596. read 9 October 2026. journals.sagepub.com
- Tong A, Sainsbury P, Craig J. Consolidated criteria for reporting qualitative research (COREQ): a 32-item checklist for interviews and focus groups. International Journal for Quality in Health Care. 2007;19(6):349-357. read 9 October 2026. researchnow.flinders.edu.au
- von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP; STROBE Initiative. Strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ. 2007;335(7624):806-808. doi:10.1136/bmj.39335.541782.AD read 9 October 2026. doi.org/10.1136/bmj.39335.541782.AD
- von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Epidemiology. 2007;18(6):800-804. doi:10.1097/EDE.0b013e3181577654 read 9 October 2026. doi.org/10.1097/EDE.0b013e3181577654
Every source above is a peer-reviewed article, a publisher’s record of one, or a standard issued by the body that owns it. Each carries the date it was last read. Where a standard moves, this article is revised rather than left up with a note.
How to cite this article
Pick your referencing style and copy it. Your university or target journal may vary the punctuation — check its guide if it has one.
Pubrica Academy. (2026, October 9). Qualitative Research in Clinical Trials: Purposes, Design and Reporting. Pubrica. https://pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/
Pubrica Academy. “Qualitative Research in Clinical Trials: Purposes, Design and Reporting.” Pubrica, 9 Oct. 2026, pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/.
Pubrica Academy (2026) Qualitative Research in Clinical Trials: Purposes, Design and Reporting. Pubrica. Available at: https://pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/ (Accessed: 15 October 2026).
Pubrica Academy. “Qualitative Research in Clinical Trials: Purposes, Design and Reporting.” Pubrica. October 9, 2026. https://pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/
Pubrica Academy, ‘Qualitative Research in Clinical Trials: Purposes, Design and Reporting’ (Pubrica, 9 October 2026) <https://pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/> accessed 15 October 2026.
Pubrica Academy. Qualitative Research in Clinical Trials: Purposes, Design and Reporting. [Internet]. Pubrica; 2026 Oct 9 [cited 2026 Oct 15]. Available from: https://pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/
Pubrica Academy, “Qualitative Research in Clinical Trials: Purposes, Design and Reporting,” Pubrica, Oct. 9, 2026. [Online]. Available: https://pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/
@misc{qualitative_research_in_clinical_trial_2026,
author = {{Pubrica Academy}},
title = {Qualitative Research in Clinical Trials: Purposes, Design and Reporting},
howpublished = {Pubrica},
year = {2026},
month = {oct},
url = {https://pubrica.com/academy/concepts-definitions/qualitative-research-in-clinical-trials/},
note = {Accessed: <span data-cite-acc="iso">2026-10-15</span>}
}This article is written to be learned from. Cite it if you are referring to what it says; if you are using the evidence behind it, cite the sources listed above instead.
Qualitative work package specification
This should be completed before the protocol or grant application is drafted. Any section left blank at submission is likely to be the section that goes undelivered. The proposal data indicate clearly which section that usually is: the named researcher.
EMBEDDED QUALITATIVE WORK PACKAGE
Trial: ______________________ Protocol version: ______________
1. QUESTIONS (as questions, not topics; two or three maximum)
Q1: ..........................................................
Q2: ..........................................................
Each question must pass this test: does the answer change
something someone will do?
Q1 changes: ______________________
Q2 changes: ______________________
2. WHY QUALITATIVE (explicit rationale - 56% of proposals lack one)
Why this cannot be answered with trial or routine data:
..............................................................
3. DATA SOURCES
[ ] Audio-recorded recruitment consultations target n: ______
[ ] Interviews with recruiting / delivering staff n: ______
[ ] Interviews with participants n: ______ arm(s): ________
[ ] Interviews with patients who declined n: ______
[ ] Observation of delivery sessions: ______
[ ] Documents / screening logs
[ ] Other: ______________________
4. SAMPLING
Frame: ______________________________________________________
Dimensions of variation deliberately sampled: _______________
Sites over-sampled and why: _________________________________
Stopping condition: _________________________________________
(Planned maximum for budgeting: ______ )
5. TIMING
Relative to randomisation: [ ] before [ ] during [ ] after
Relative to unblinding: analysis complete before unblinding?
[ ] yes [ ] no - justification: ___________________________
Feedback points in the trial calendar: month ___, month ___
6. WHO DOES IT (28% of proposals state this - be in that 28%)
Lead qualitative researcher: ______________ FTE: ____ for __ mo
Second coder: ______________ subsample coded: ____%
Independent of participant care? [ ] yes [ ] no - explain: __
Disagreement resolution procedure: __________________________
7. ANALYSIS
Named approach: [ ] thematic [ ] framework [ ] grounded theory
[ ] other: ______________
Initial coding frame source (if any): ________________________
Software: ______________
Who sees trial outcome data, and when: _______________________
8. INTEGRATION WITH THE TRIAL
Reported: [ ] in main paper [ ] companion paper [ ] both
What happens if qualitative and quantitative findings conflict:
..............................................................
Route from finding to action during the trial:
..............................................................
9. REPORTING
[ ] COREQ 32 items mapped during drafting, not at submission
[ ] Researcher-participant relationship reported
[ ] Stopping rule reported
[ ] Quotation identification method stated
[ ] Deviations from this specification logged with dates Pubrica AcademyWritten and maintained by Pubrica’s subject matter experts. Articles are revised when the underlying standards change, and each article states the date on which it was last reviewed.
How Pubrica can help
Pubrica prepares qualitative work packages that a funder can cost and a team can deliver, with stated questions, a sampling frame, a named analytic approach, a second-coder arrangement and a feedback route into the trial management structure. Pubrica also maps completed qualitative strands against the 32 COREQ items before submission and identifies the two items that trial-embedded studies most often omit. Where a trial is already recruiting below target, Pubrica assists in designing the qualitative diagnosis so that it reaches the trial management group while recruitment can still be influenced.