Abstract
Background: An increasing number of patients require prolonged stays in the intensive care unit (ICU), defined in this thesis as exceeding seven days. Although these patients represent a relatively small proportion of total ICU admissions, prolonged stays are associated with substantial physical, psychological, cognitive, and emotional burdens for patients and their families, as well as significant resource implications for health and social care systems. Numerous quality improvement (QI) tools are currently employed in ICUs to enhance patient safety, clinical outcomes, and the delivery of relevant care processes. However, none are specifically tailored to address the unique and complex needs of patients experiencing prolonged ICU stays within the UK context. This gap in targeted quality improvement risks omissions in care and may contribute to poorer outcomes for a patient population already known to face significant challenges in recovery.Objectives:
Objective one: To identify relevant actionable processes of care (interventions over which healthcare professionals have direct control) that improve the experience of a prolonged ICU stay and may improve outcomes, as identified by patients who have experienced a prolonged ICU stay, family members, and healthcare professionals.
Objective two: To build consensus on which of these identified actionable processes of care are the most important for inclusion in a quality improvement tool for adult patients with a prolonged ICU stay.
Objective three: To establish feasibility and healthcare professional acceptability of the resulting quality improvement tool in a central London ICU to understand potential uptake and utility in clinical practice.
Methods: A participatory approach was employed across three phases of research: Phase one: Guided by principles of experience-based co-design, semi-structured interviews were conducted with patients, family members, and healthcare professionals with direct experience with prolonged ICU stays. Interviews were conducted in person, via telephone, or through videoconferencing platforms. Data were analysed using the framework method, enabling the systematic identification and categorisation of actionable care processes that influenced the ICU experience.
Phase two: A two-round electronic Delphi survey was undertaken, with participants asked to rate the importance of each actionable process of care on a Likert scale from 1-9. An online meeting employing the nominal group technique then took place to further prioritise the most important of these actionable processes of care.
Phase three: Feasibility and acceptability testing of the resulting tool was conducted in an academic ICU in London. Mixed methods were employed, using the Theoretical Framework of Acceptability (TFA) to structure findings from interviews with and questionnaires from healthcare professionals using the tool. Metrics including completion rates, time required to complete the tool and actions generated by tool usage were collected. The Expert Recommendations for Implementing Change (ERIC) framework was used to describe what implementation strategies were employed, and the Template for intervention description and replication (TiDieR) to demonstrate how the tool was implemented in practice.
Results:
Phase one: 37 actionable processes of care were identified from 38 interviews with patients, family members, and healthcare professionals. Participants described the need for humanised care including involving patients and family in care decisions, support for emotional needs, and the presence of family. 2 additional actionable processes of care identified in a previously published systematic review but not identified from interviews were added to these.
Phase two: 116 participants were recruited to the first electronic Delphi round, including 45 (39%) patients. 78% of all participants were retained in round two. From 39 actionable processes of care, 32 were rated of critical importance after two electronic Delphi survey rounds. 15 participants of the online consensus-building meeting further prioritised this list, resulting in the final 25-item tool, ‘Recover25’. Items focusing on patient comfort, recovery, and humanisation were most highly rated, with family involvement items deprioritised.
Phase three: Recover25 was used with 26 patients with a median ICU stay of 12.5 days over a twelve-week period. Recover25 generated a median of one new action and took 9.8 minutes to complete. Eleven interviews were conducted with healthcare professionals, who returned 33 TFA questionnaires. Recover25 was deemed acceptable, with 94% participants rating it an ethical intervention because of the person-centred focus.
Discussion: This thesis addresses a critical gap in the evidence: the absence of a quality improvement (QI) tool specifically designed to guide the care of the UK prolonged-stay intensive care unit (ICU) population. This lack posed a risk of omissions in essential, context-specific aspects of care. To address this, robust participatory methods were employed to identify and prioritise actionable care processes for inclusion in a novel QI tool; ensuring it was grounded in the experiences of patients and families, relevant to frontline healthcare professionals, and aligned with the best available evidence. The resulting tool, Recover25, is distinct from existing ICU QI frameworks, which overlook vital humanising processes of care that are pertinent to prolonged ICU stays. Recover25 was found to be both feasible and acceptable in clinical practice, prompting the consideration of, on average, one additional care process per patient when implemented. A comprehensive, multistage evaluation is now required to assess its impact on patient-, family-, and healthcare professional- reported outcomes, guided by the Medical Research Council’s updated framework for evaluating complex interventions.
Conclusions: This thesis presents an innovation for UK prolonged stay ICU patients: a rigorously developed and stakeholder-centred quality improvement tool, Recover25. It can prevent omissions in care and was feasible and acceptable to healthcare professionals, who felt it represented ethical care. Future multistage evaluation of the effectiveness of Recover25 using the Medical Research Council’s updated complex intervention framework is proposed.
| Date of Award | 1 Apr 2026 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Louise Rose (Supervisor) |
Cite this
- Standard