Dr Helen Mason, Clinical Director for Medicine, faces a challenge translating her recently completed systematic review on discharge planning into practice improvements across her hospital trust. Her review, organised using the Theoretical Domains Framework (TDF), found strong evidence that structured discharge planning reduces readmissions by 25% and improves patient satisfaction. However, 18 months after introducing new discharge protocols based on her review findings, implementation remains inconsistent across wards. Some wards achieve excellent compliance (>90%) with discharge protocols and demonstrate reduced readmission rates, while others show minimal change (<30% compliance) with no improvement in outcomes. Dr Mason's quality improvement team has conducted focus groups with nurses, doctors, pharmacists, and patients, revealing complex barriers: time pressures, role confusion, technology challenges, and cultural resistance to change. She's now considering whether to conduct formal research using the Behaviour Change Wheel framework to study implementation systematically or whether this should remain a quality improvement initiative. The trust board is pressuring Dr Mason to demonstrate return on investment, while she recognises that successful implementation requires understanding and addressing the human factors behind variation in uptake. 1. Distinguish between what constitutes research versus quality improvement when using theoretical frameworks like TDF or CFIR. If Dr Mason wanted to study implementation systematically using these frameworks, what would push this work into research territory requiring ethical approval? Please discuss how theoretical frameworks can guide both QI activities and research studies across clinical practice, medical education, and organisational leadership contexts, and why this distinction matters for healthcare professionals planning improvement initiatives 2. Consider how theoretical frameworks enable more effective translation of research evidence into practice across diverse contexts. Using examples from your own professional environment – whether clinical practice, educational delivery, or organisational leadership – discuss how frameworks like TDF or CFIR would help identify specific barriers to evidence implementation and guide targeted intervention design. How does this systematic, theory-driven approach differ from traditional 'education and training' approaches to practice change, and what economic considerations should inform implementation decisions?
Reading File
Finding Sources
Finding Sources
Searching PubMed
"Theoretical Domains Framework" AND "implementation" AND "behaviour change"
Searching the Web
research vs quality improvement distinction ethical approval TDF CFIR implementation science 2024
Finding Sources
Finding Sources
Searching PubMed
"discharge planning" AND "readmission" AND "implementation"
Reading Web Page
research vs quality improvement distinction ethical approval theoretical frameworks
Reading Article
Searching PubMed
"Behaviour Change Wheel" AND "implementation" AND "healthcare"
Searching the Web
NHS HRA research vs quality improvement distinction criteria ethical approval implementation 2024 2025
Reading Article
Reading Web Page
quality improvement vs research distinction ethical approval criteria NHS implementation frameworks
Searching PubMed
"implementation science" AND "return on investment" AND "healthcare"
Reading Article
Reading Article
Searching PubMed
"CFIR" AND "barriers facilitators" AND "hospital" AND "evidence based practice"
I now have comprehensive material to construct a thorough, well-evidenced response. Let me compile the full answer.
Dr Mason's Implementation Science Challenge: A Critical Analysis
Question 1 - Research vs Quality Improvement: Distinguishing the Boundaries
The Core Definitional Distinction
The distinction between research and quality improvement (QI) is one of the most contested and practically consequential boundaries in healthcare. Both involve systematic data collection, use of theoretical frameworks, and aim to improve care - yet they carry fundamentally different ethical, regulatory, and governance obligations.
Quality Improvement is defined as systematic, data-guided activity designed to bring about immediate, positive changes in healthcare delivery within a particular setting. Its primary purpose is to improve a local service. The knowledge generated may not be intended to travel beyond the organisation. Clinical audit, PDSA (Plan-Do-Study-Act) cycles, and process mapping are the classic tools of QI.
Research is defined by the UK Policy Framework for Health and Social Care Research (NHS HRA, updated July 2025) as an attempt to derive generalisable or transferable new knowledge. Research is characterised by: a formal hypothesis; data collected that goes beyond what is needed for immediate care; prospective design aimed at producing knowledge that others can use; systematic methods intended for dissemination; and exposing participants to procedures or interventions beyond routine care.
Implementation Science sits in a distinctive third space. According to Eccles and Mittman's foundational definition, it is "the scientific study of methods and strategies that facilitate the uptake of evidence-based practice and research into regular use by practitioners and policymakers." This is intrinsically different from doing improvement - it is studying why and how improvement happens.
What Pushes Dr Mason's Work into Research Territory
Dr Mason's situation contains several features that, depending on design decisions, could push her work from QI into research requiring formal ethical approval. The HQIP's 2025 Guide to Managing Ethical Issues in QI and Clinical Audit Projects, alongside the NHS HRA decision tool, identifies the following key triggers:
1. Intention to generate generalisable knowledge
If Dr Mason's goal is to produce findings about why implementation fails or succeeds - knowledge applicable to other trusts, other clinical areas, or usable by others to inform policy - this constitutes research. Collecting focus group data specifically to test TDF or BCW hypotheses about human behaviour, with intent to publish, immediately shifts this toward research.
2. Systematic data collection from identifiable participants beyond what clinical care requires
Her focus groups with nurses, doctors, pharmacists, and patients involve collecting personal opinions and experiences. If these are recorded, transcribed, analysed using a formal framework, and used to draw conclusions about groups of people (rather than just the local ward), participant protection requires ethical oversight. The NHS HRA acknowledges that participants in focus groups may be identifiable and may be exposed to psychological risk.
3. Comparing wards as distinct "sites" in a designed study
If Dr Mason formalises the comparison between high-compliance (>90%) and low-compliance (<30%) wards using CFIR constructs - defining each ward as an "inner setting" and systematically comparing determinants across them - this constitutes a comparative study. The 2025 CFIR User Guide (Aarons et al., Implementation Science, 2025) explicitly distinguishes between using CFIR for a single-site QI project versus a multi-site study comparing implementation determinants, noting the latter requires broader governance.
4. Prospective randomisation or controlled allocation of interventions
If she designs an intervention (e.g., educational module, feedback mechanism, role clarification tool) and allocates it to some wards but not others to evaluate its effect on compliance, this becomes a stepped-wedge or cluster trial - unambiguously research requiring full HRA approval and Research Ethics Committee (REC) review.
5. Vulnerable participants and psychological risk
Patients included in focus groups about their discharge experiences are a potentially vulnerable population. Even in QI contexts, the HQIP guidance (2025) warns that harm can arise from qualitative data collection, and that ethical consideration - even if not full REC review - is always required.
Practical test: The NHS HRA decision tool at hra-decisiontools.org.uk/research asks a structured sequence of questions. Key discriminating factors are whether the activity is designed to produce generalisable knowledge, whether it involves allocation of participants to groups, and whether participants undergo procedures beyond routine care.
How Theoretical Frameworks Sit Across Both Activities
The TDF and BCW/CFIR can function legitimately in both QI and research - but the way they are used differs fundamentally.
| Feature | QI Use of TDF/CFIR | Research Use of TDF/CFIR |
|---|---|---|
| Purpose | Guide local service improvement | Generate transferable knowledge about implementation determinants |
| Sampling | Convenience - whoever is available in your ward | Purposive or representative - designed to answer a research question |
| Analysis | Pragmatic thematic mapping to guide action | Rigorous framework analysis with inter-rater reliability checks, saturation testing |
| Outcome | Local change in practice | Publication, policy recommendations, contribution to knowledge base |
| Governance | Local QI governance, information governance | Full HRA/REC ethical approval, informed consent, data management plan |
| Reporting standard | SQUIRE guidelines | COREQ (qualitative), CFIR reporting standards, TIDier (intervention description) |
A 2025 systematic review by Bierbaum et al. (PMID: 40241054, BMC Health Services Research) found that across 12 studies integrating QI and implementation science in tertiary care, the most common pairing was TDF/BCW/CFIR with PDSA cycles, process mapping, and audit and feedback. Crucially, only some studies obtained ethical approval - and the review identified inconsistent use of terminology and ethics reporting as a major gap. This matters for Dr Mason: if she uses TDF in focus groups without ethical oversight, she risks both governance failures and reduced publishability.
Application Across Clinical Practice, Medical Education, and Organisational Leadership
Clinical Practice: Using TDF for a ward-level medicines reconciliation QI project (mapping barriers to the 14 domains, then designing local interventions) is firmly QI. Using TDF to conduct a formal qualitative study across multiple trusts on why prescribers resist electronic decision support, with intent to publish, is research. The same framework, different governance pathway.
Medical Education: Using the BCW to redesign a clinical handover training programme based on observed performance gaps is educational development (QI equivalent). Conducting a randomised study comparing two training interventions with pre/post competency assessments and follow-up at six months is educational research requiring ethics approval.
Organisational Leadership: A trust using CFIR to assess its own readiness for electronic patient record implementation, using staff surveys analysed internally to guide rollout strategy, is organisational QI. A trust partnering with an academic institution to produce a peer-reviewed study on CFIR constructs predicting implementation success across 15 trusts is health services research.
Why the distinction matters practically:
- Legal protection: Incorrectly classifying research as QI to avoid ethics review can expose institutions to regulatory sanctions under the UK Policy Framework for Health and Social Care Research and GDPR.
- Participant safety and trust: Research participants have rights to informed consent, withdrawal, and data protection that QI participants may not be formally afforded.
- Credibility and publishability: Journals increasingly require evidence of ethical approval even for QI studies; SQUIRE and EQUATOR guidelines distinguish what is expected.
- Funding eligibility: Research Council and NIHR funding requires full research classification and ethical approval.
- Intellectual property and knowledge contribution: Work classified as QI cannot always be disseminated as generalisable evidence - limiting Dr Mason's ability to contribute to the wider implementation science evidence base.
Question 2 - Theory-Driven Implementation vs "Education and Training" Approaches
Why Theoretical Frameworks Enable Better Translation
The traditional approach to practice change in healthcare has been straightforward: identify the evidence-practice gap, deliver education (a lecture, a guideline, a training session), and expect behaviour change to follow. Decades of evidence show this approach fails. The core reason is that it assumes knowledge deficit is the primary barrier to behaviour change, when the reality is far more complex.
The COM-B model - the behavioural hub of the Behaviour Change Wheel - proposes that any behaviour requires three elements: Capability (physical and psychological), Opportunity (physical and social), and Motivation (automatic and reflective). Traditional education addresses only one component of capability (knowledge). It ignores time pressures (physical opportunity), team culture (social opportunity), habit and identity (automatic motivation), and self-efficacy (psychological capability).
Mather, Pettigrew and Navaratnam's 2022 systematic review of reviews (PMID: 36042457, Systematic Reviews) - which used TDF and BCW to map barriers and facilitators to clinical behaviour change in primary care across 19 reviews - found that the most consistently identified barriers relate to "Knowledge," "Environmental Context and Resources," and "Social Influences." Critically, Motivation factors were least often identified - not necessarily because they are less important, but because standard research methods are poorly designed to capture them, and practitioners may lack insight into their own motivational barriers.
Applying TDF/CFIR to Dr Mason's Discharge Planning Problem
Dr Mason's situation is a near-perfect case study for theory-driven implementation analysis. Her ward-level variation (>90% vs <30% compliance) is not random noise - it reflects systematic differences in the implementation environment that CFIR and TDF can characterise precisely.
Using the TDF to map individual-level barriers (from Dr Mason's focus groups):
| TDF Domain | Dr Mason's Likely Barriers (from Focus Group Findings) |
|---|---|
| Knowledge | Staff may understand the protocol exists but not its rationale or their specific role in it |
| Skills | Pharmacists may lack training in the new e-prescribing module used in discharge; nurses may lack communication skills for patient counselling |
| Social/professional role and identity | Doctors may not see discharge planning as "their" responsibility; role confusion identified by focus groups maps directly here |
| Beliefs about capabilities (self-efficacy) | Staff on low-compliance wards may believe they can't complete the protocol given their workload |
| Motivation and goals | Competing demands mean discharge planning is deprioritised relative to acute care tasks (automatic motivation bias) |
| Memory, attention, decision processes | Time-pressured clinical environments mean well-intentioned staff forget steps without prompts or checklists |
| Environmental context and resources | Technology challenges = physical opportunity barrier; inadequate staffing = resource constraint |
| Social influences | Cultural resistance reflects social opportunity barriers - norms on some wards where "we don't do it that way" |
| Emotion | Staff may feel anxious or frustrated with change, particularly if they perceive implementation as top-down imposition |
| Behavioural regulation | High-compliance wards likely have self-monitoring mechanisms (e.g., discharge checklists, audit feedback); low-compliance wards do not |
Using CFIR to map organisational and contextual barriers:
CFIR's five domains add a layer TDF cannot fully capture:
- Inner Setting: Ward culture, leadership engagement, available data infrastructure, and history of change (wards that have successfully changed before are more ready)
- Outer Setting: Trust-level pressure from the board (creating both opportunity and anxiety), peer benchmarking, commissioner incentives
- Innovation characteristics: Is the discharge protocol perceived as credible? Is it a good fit with existing workflows, or does it require major redesign?
- Individuals involved: Are ward champions present? Do senior clinicians actively model the behaviour?
- Implementation process: Was the protocol co-designed with ward staff or delivered top-down? Were implementation leads trained? Is there ongoing support?
The contrast between Dr Mason's high and low compliance wards almost certainly reflects systematic differences across these CFIR constructs. The 2025 CFIR User Guide recommends that even in single-trust QI contexts, defining the "inner setting" clearly (in this case, the ward) enables structured comparison of implementation determinants.
Targeted Intervention Design Using the BCW
Once barriers are mapped using TDF/COM-B, the BCW provides a systematic mechanism to select interventions. The eight intervention functions of the BCW are: Education, Persuasion, Incentivisation, Coercion, Training, Restriction, Environmental Restructuring, Modelling, and Enablement.
For Dr Mason's scenario:
- TDF: Memory/attention/decision → BCW: Environmental restructuring - add discharge checklist as a default field in the electronic patient record; create a visual prompt at the nurses' station
- TDF: Social influences/culture → BCW: Modelling - identify a clinical champion on each low-compliance ward; use high-compliance ward staff as peer educators rather than management-led training
- TDF: Role identity → BCW: Persuasion - engage ward doctors through clinical grand rounds showing their ward's readmission data compared to high-compliance peers
- TDF: Environmental context (time) → BCW: Enablement - restructure ward rounds to include a dedicated daily discharge planning review slot; reduce documentation burden
- TDF: Skills → BCW: Training - bespoke, ward-based simulation of the e-prescribing discharge module rather than generic e-learning
This systematic mapping produces what is called a Behaviour Change Technique (BCT) inventory - a specific, theoretically-justified set of active ingredients for the intervention. The BCT Taxonomy v1 (Michie et al.) identifies 93 distinct BCTs linked to specific COM-B targets, enabling precise intervention specification and later replication.
Theory-Driven vs Traditional Education and Training: The Key Contrasts
| Dimension | Traditional Approach | Theory-Driven (TDF/BCW/CFIR) Approach |
|---|---|---|
| Assumption | Knowledge deficit drives non-compliance | Multiple interacting barriers at individual, team, and system levels |
| Diagnosis | None - assumes problem is known | Systematic barrier assessment using validated frameworks before intervention design |
| Intervention design | Generic training for all staff | Targeted interventions matched to specific barriers for specific groups |
| Contextual sensitivity | One-size-fits-all protocol across all wards | Allows differentiation of intervention by ward-level CFIR profile |
| Evidence base | Weak - simple education rarely changes complex behaviour long-term | Growing - BCW-designed interventions show superior effect sizes in several clinical domains |
| Sustainability | Low - behaviour reverts when training pressure is removed | Higher if structural and cultural levers are addressed alongside individual capability |
| Accountability | Hard to explain why it failed | If intervention fails, the theoretical model predicts where the breakdown occurred |
The 2022 systematic review (PMID: 36042457) notes that "PCPs may lack insight into the role that 'Motivation' and aspects of psychological Capability have in behaviour change" - which is precisely why education alone fails. It addresses what staff can see as a barrier but misses the deeper motivational and environmental determinants they cannot easily articulate.
Economic Considerations for Dr Mason's Trust Board
The trust board's demand for return on investment (ROI) data is legitimate and can be addressed systematically.
The cost of non-implementation is quantifiable. Dr Mason's own systematic review demonstrates that structured discharge planning reduces readmissions by 25%. In England, the NHS tariff for emergency readmissions carries financial penalties for trusts when readmission occurs within 30 days for many conditions. If the trust has, say, 500 potentially avoidable readmissions annually, each costing approximately £3,000-£5,000 (including bed-day costs and penalty payments), a 25% reduction represents £375,000-£625,000 in avoided costs annually.
The cost of theory-driven implementation is modest by comparison. A structured TDF/BCW implementation programme - including focus groups, barrier analysis, champion training, and environmental modifications - typically costs significantly less than its clinical outcome value. Connor et al.'s 2023 scoping review (PMID: 36751881, Worldviews on Evidence-Based Nursing) found that of studies measuring ROI on evidence-based practice implementation, 94% showed a positive ROI, with none showing negative ROI.
The cost of failed implementation is also quantifiable - and often ignored. Landi et al.'s 2024 systematic review (PMID: 38195545, BMC Health Services Research) found that while transitional care programmes (which include discharge planning) have generally positive economic impact, the number of organisational components implemented predicts success - not merely whether a protocol exists. Low-compliance wards are spending implementation resources without realising the benefit. This is a waste the board should be motivated to address.
APEASE criteria provide a structured economic framework for intervention selection within the BCW:
- Affordability - what is the cost of the intervention relative to benefit?
- Practicability - can it be delivered within current resource constraints?
- Effectiveness - what is the evidence base?
- Acceptability - will staff and patients accept it?
- Safety/side effects - does the intervention itself cause harm?
- Equity - does it reduce or worsen inequalities between patient groups or staff groups?
Using APEASE alongside BCW intervention selection allows Dr Mason to present the trust board with a prioritised, economically justified, theory-grounded implementation plan rather than another round of generic training.
A final note on the research question: If Dr Mason proceeds to formally study this using BCW and CFIR - publishing findings about implementation determinants in medicine discharge planning across her trust - she not only satisfies her own institution's board but contributes to a field where evidence is thin. The 2025 CFIR User Guide notes that implementation research is "iterative and evolving," and that contributions from clinical settings are needed. This dual purpose - local QI and generalisable knowledge - is achievable, but requires the governance pathway of research (HRA approval, REC review, informed consent) from the outset, not added retrospectively when publication is contemplated.
Summary
The critical insight for Dr Mason and clinicians in similar positions is that the TDF, BCW, and CFIR are not inherently research tools or QI tools - they are analytical and intervention-design frameworks that can serve both purposes. What determines the governance pathway is the intent, design, and scope of the work: is it generating local knowledge to improve a service, or generalisable knowledge to inform others?
Using these frameworks as QI tools: local barrier assessment, targeted intervention, PDSA cycles, audit feedback - no ethics approval required beyond standard information governance.
Using these frameworks as research tools: multi-site comparisons, formal participant recruitment and consent, framework analysis for publication, theory testing - full HRA/REC ethical approval required.
Dr Mason's most prudent path is to use the HRA decision tool early, design her focus group analysis to dual-qualify as both ethical QI and publishable research from the start, and present the board with an APEASE-justified implementation plan grounded in the economic cost of 18 months of inconsistent compliance.
Key References:
- Bierbaum M et al. (2025). Integration of QI and implementation science in healthcare: systematic review. BMC Health Services Research [PMID: 40241054]
- Mather M et al. (2022). Barriers and facilitators to clinical behaviour change using TDF/BCW: systematic review of reviews. Systematic Reviews [PMID: 36042457]
- Landi S et al. (2024). Organisational levers and economic benefits in transitional care: systematic review. BMC Health Services Research [PMID: 38195545]
- Connor L et al. (2023). Evidence-based practice and healthcare ROI: scoping review. Worldviews on Evidence-Based Nursing [PMID: 36751881]
- HQIP (2025). Guide to Managing Ethical Issues in QI and Clinical Audit Projects
- Aarons GA et al. (2025). CFIR User Guide. Implementation Science [link.springer.com/article/10.1186/s13012-025-01450-7]
- NHS HRA (2025). UK Policy Framework for Health and Social Care Research; Decision tool: hra-decisiontools.org.uk/research
This is a shared conversation. Sign in to Orris to start your own chat.