Development of a cognitive rehabilitation intervention to support patients during and after intensive care admission: a stakeholder-informed study

Abstract

Background

Patients who survive critical illness and intensive care unit (ICU) admission often experience long-lasting physical, psychological, and cognitive impairments. Cognitive impairments are common and may persist well beyond ICU discharge. Few structured interventions address these impairments across the care continuum, from the ICU to post-discharge recovery. This gap informed the development of a stakeholder-driven cognitive rehabilitation intervention. The aim was to develop a cognitive rehabilitation intervention for critically ill patients that bridges the period from ICU admission to post-discharge recovery.

Methods

The intervention was developed using a multiphase process guided by the 2021 Medical Research Council (MRC) framework for complex interventions and reported in line with the GUIDED and TIDieR checklists. Stakeholder input was gathered through: (1) a workshop (n = 11) to identify end-user needs; (2) a three-round Delphi process (n = 39) to prioritise interventions using a seven-point Likert scale; and (3) expert consultations to refine selected components. Qualitative data from the workshop and open-ended Delphi responses were analysed thematically, while quantitative analysis identified consensus, defined a priori as ≥80% agreement on importance.

Results

The process generated a range of proposed interventions, including both novel and established elements. Interventions reaching stakeholder consensus were further reviewed by experts for clinical relevance and feasibility. Two prototype interventions were selected: (1) a Mindfulness-based intervention and (2) a Brain Training intervention.

Conclusion

This study outlines a stakeholder-informed process for developing a cognitive rehabilitation intervention tailored to ICU patients across the care continuum. The co-design approach improved contextual fit and clinical applicability. These findings support the planned feasibility evaluation of the proposed interventions.

Trial registration

The overall research study received ethics approval, while formal ethics committee approval was not required for the workshop and Delphi activities under Danish legislation (osf.io/5a4xq).

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12913-026-15036-6.

Key contributions to the literature.

• This study presents a novel, stakeholder-informed cognitive rehabilitation intervention for ICU patients, addressing the prevalent issue of cognitive impairment among ICU survivors across the continuum of care from admission to post-discharge recovery.

• The study describes a transparent, systematic, and replicable intervention development process, guided by the Medical Research Council (MRC) framework and integrating existing evidence, neurorehabilitation theory, and experiential knowledge from former patients, relatives, and healthcare professionals.

• Establishes a methodological foundation for future research on feasibility, acceptability, and implementation research in ICU survivorship care, offering a context-sensitive development model with potential applicability beyond the Danish healthcare setting.

Background

Advancements in medical treatment and technology have significantly improved survival rates among patients admitted to intensive care units (ICUs). Despite these advances, ICU survivors frequently experience prolonged physical, psychological, and cognitive impairments, collectively termed ‘Post-Intensive Care Syndrome’ (PICS). The impairments associated with PICS present escalating challenges for healthcare systems, community care providers, and public health initiatives, potentially hindering patients’ ability to regain a meaningful life [1–3]. Cognitive impairments are particularly common and may affect domains such as attention, memory, and executive function [4, 5]. These impairments may emerge during ICU admission and persist for months or even years after ICU discharge [6, 7]. Studies show that 40–60% of ICU survivors demonstrate global cognitive performance comparable to that observed in individuals with moderate brain injury when assessed using validated neuropsychological tests [4, 7, 8]. However, the prevalence of cognitive impairments varies widely depending on patient demographics, the timing of cognitive assessments, and methodological approaches [1, 3, 4, 9]. The aetiology and underlying mechanisms of cognitive impairments following critical illness are multifactorial and complex. Contributing factors may involve both patient-related (e.g. age, educational level, comorbidities, frailty) and illness- and treatment-related factors such as delirium, hypoxia, systemic inflammation, medication effects, sleep disruption, pain, severity of critical illness, and environmental stressors associated with ICU [10, 11]. Importantly, such impairments are observed across a heterogeneous ICU population, including patients admitted with sepsis, acute respiratory failure, cardiac arrest, major surgery, or traumatic injury. Delirium during ICU admission is a strong predictor of long-term cognitive impairments [12–14]. Prolonged cognitive impairment is associated with reduced health-related quality of life and physical function, as well as impaired ability to live independently and increased mortality [15, 16]. Given the high prevalence, multifactorial aetiology, and long-term consequences of cognitive impairments in ICU survivors, there is a need for targeted rehabilitation interventions aimed at mitigating these effects.

The international clinical practice guidelines by Lewis et al. [17] provide current evidence-based recommendations for preventing cognitive impairments during ICU admission. These guidelines address the prevention and management of risk factors that may affect cognitive function in critically ill patients, including pain management, agitation and sedation strategies, delirium prevention and management, early mobilisation, and minimising sleep disruption [17, 18].

Most rehabilitation interventions aiming to support cognitive recovery have traditionally been implemented after ICU discharge [19]. These include physical exercises, patient diaries, music therapy, mindfulness-based strategies, and environmental or nature-based interventions [9, 20]. While not all these approaches directly target specific cognitive functions, they may support cognitive function generally through parasympathetic activation, reducing stress, improving sleep quality and enhancing emotional well-being, and enabling engagement in rehabilitation activities. In contrast, evidence regarding cognitive rehabilitation initiated during ICU admission remains limited but is gradually emerging. Early cognitive engagement may include early mobilisation, circadian rhythm or light therapy, orientation and communication strategies, music therapy, and outdoor therapy [21–25], which may help maintain attentional orientation and support continuity in rehabilitation across care settings.

Furthermore, one pilot study found that introducing cognitive exercises during ICU admission was feasible, practical, and manageable for both patients and staff [26]. However, current research lacks consistency in key aspects such as intervention complexity (single- vs. multi-component), timing (initiation and duration), dosage (frequency and intensity), and delivery methods [19, 27]. Limited evidence is therefore available regarding the optimal approach to cognitive rehabilitation for patients during and after ICU admission [19], which challenges implementation in clinical practice and creates uncertainty around the development of targeted interventions.

Although cognitive rehabilitation should ideally be tailored to individual impairments, the acute ICU context is characterised by fluctuating clinical stability and limited opportunity for cognitive assessment. In the present intervention development, a structured and standardised approach was chosen to enable a systematic exploration of intervention feasibility within routine care. Interventions targeting commonly affected cognitive domains such as attention, memory, and executive functioning may therefore represent a pragmatic starting point for early cognitive rehabilitation. In future clinical application, greater individualisation and adaptation to patients’ functional needs and recovery will be essential.

Consequently, there is a clear need to develop evidence-based cognitive rehabilitation interventions for ICU patients. This need is further underscored by evidence from individuals recovering from stroke [28], multiple sclerosis [29] and traumatic brain injury [30], where structured cognitive rehabilitation has been shown to support recovery of attention, memory, and executive functioning. Although the underlying pathophysiological mechanisms differ, critically ill patients present with similar cognitive symptom profiles that partially overlap with those observed in these conditions. Similar patterns of cognitive impairments have also been described following severe systemic illness or infection, perioperative neurocognitive disorders, and broader critical illness-related neurocognitive dysfunction [31–34]. Together, these observations suggest that selected rehabilitation principles from neurological and somatic populations may inform the development of cognitive rehabilitation interventions across the ICU recovery phases.

Therefore, the aim of the present study was to develop a theory-informed cognitive rehabilitation intervention for ICU patients throughout their care pathway, with the specific structure and content of the intervention expected to emerge through a stakeholder-informed development process. The intervention aims to address current knowledge gaps to improve ICU care pathways by establishing a transparent foundation for future evaluation.

Methods

Design

This intervention development study was guided by the Medical Research Council’s (MRC) framework for the development and evaluation of complex interventions [35]. The focus was on intervention development, while also addressing implementation early in the process to enhance adoption, as outlined in the MRC Framework [35] and by O’Cathain et al. [36]. The process was informed by the six core elements of the MRC Framework including context, iterative development and testing of programme theory, active stakeholder engagement, identification of key uncertainties, refinement of the intervention, and attention to potential economic considerations [35]. To strengthen the development process, we also applied the co-production and prototyping framework for public health interventions by Hawkins et al. [37]. The study is reported in accordance with the GUIDED recommendations for intervention development and the Template for Intervention Description and Replication (TIDieR) checklist for detailed intervention description [38, 39].

The research group

The intervention development was initiated as part of the Intensive Care Unit Cognitive Rehabilitation (ICU CogHab) research programme, which aims to enhance cognitive rehabilitation for ICU survivors. This programme is a collaboration between Aarhus University Hospital and the University Hospital of Southern Denmark, Kolding. The multidisciplinary research group included ICU professors, senior researchers, clinical nurse specialists, an ICU physician, an occupational therapist, and a physiotherapist. This group held decision-making authority regarding the content and, to some extent, the delivery of the cognitive rehabilitation, with approval from ICU management. Additionally, a project team, including the first author (KA), two ICU nurses, and two occupational therapists, was responsible for delivering the intervention and providing on-site support to both ICU nurses and patients.

Setting and target populations

We adopted the Context and Implementation of Complex Interventions (CICI) framework definition of setting as the physical environment in which an intervention interacts with contextual factors across macro, meso, and micro levels [40]. The development process was intended to prepare a cognitive rehabilitation intervention to be ready for evaluation and implementation in four general ICUs at Aarhus University Hospital and the University Hospital of Southern Denmark in Kolding.

Daily patient care in these ICUs is guided by the principles of PADIS [17, 18], which include minimal sedation, routine delirium screening, early mobilisation, and non-pharmacological interventions such as sleep promotion, reorientation, and communication strategies. These systemic practices were considered contextual supports underpinning the intervention, as they were expected to facilitate patients’ attentional engagement and readiness to participate in the structured cognitive activities introduced in the study.

Delirium severity was assessed using the Confusion Assessment Method for the Intensive Care Unit-7 (CAM-ICU-7) [41] and the Intensive Care Delirium Screening Checklist (ICDSC) [42]. Family members can be actively involved in communication, care planning, and early rehabilitation, with no restrictions on visiting hours. These practices reflect an evidence-based, patient-centred approach and create a clinical environment amenable to initiating cognitive rehabilitation during ICU admission. The intervention was designed for critically ill adult patients (≥18 years) expected to require an ICU stay of at least 24 hours. ICU nurses played a key role in delivering and supporting the intervention in clinical practice due to their continuous bedside presence and responsibility for identifying suitable patients.

The stakeholders

The different stages in the development process involved various stakeholders, including former ICU patients, their relatives, and healthcare professionals (HCPs). Former patients and relatives were recruited through established ICU follow-up networks, such as ICU café meetings. HCPs were recruited via clinical and research networks. HCPs included nurses, physicians, physiotherapists, and occupational therapists from both the ICU, hospital wards, and primary care settings. Some HCPs worked in clinical practice, while others were researchers with extensive knowledge of the ICU context or the interventions. The stakeholders involved represented various stages of the rehabilitation pathway for ICU patients or possessed specialised expertise in cognitive rehabilitation or psychology. Before participation, all stakeholders received detailed information about the study and their role in providing feedback throughout the ongoing development process.

Development process and data collection

The process was designed to generate and prioritise potential intervention components, with the expectation that the final prototype structure would be determined through stakeholder input and feasibility considerations. The development process consisted of three stages, in accordance with the methodology for co-designing interventions [37]. Stage 1) Establish an evidence base, a programme theory, and stakeholder involvement, stage 2) Co-designing the intervention, and stage 3) Prototyping the intervention (see Fig. 1).

Fig. 1.

Intervention development using co-design and prototyping with inspiration from Hawkins et al. (2017) [37]

Figure 1 illustrates the three stages of the development process as inspired by Hawkins et al. [37, 43].

Establishing the evidence base (stage 1)

Findings from previously conducted studies within the ICU CogHab research programme informed the development process and prioritisation of intervention components in the present study (see Results section).

Theories informing the intervention (stage 1)

The intervention prototypes were specifically designed to be accessible and usable for patients with cognitive impairments, with a focus on common post-ICU deficits, including reduced attention, memory, and executive function. To achieve this, the intervention drew on the evidence-based principles from cognitive rehabilitation and instructional methods developed for patients recovering from brain injury [43], which emphasise person-centred, functionally oriented strategies to support individuals with cognitive impairments in daily life. Person-centred principles were operationalised by allowing flexibility in intervention timing, intensity, and content, and by emphasising activities that could support engagement in meaningful everyday functions during recovery. Cognitive Load Theory [44] further informed the intervention design by highlighting the need to minimise unnecessary cognitive demands when presenting new information, especially for patients with impaired working memory. Additionally, evidence from neuropsychological rehabilitation supports the use of compensatory strategies, such as task simplification, external cues, and repetition, to enhance engagement and functional participation in cognitively demanding exercises [45]. These principles guided our development phase and the design of simple, clearly structured, and visually supported training exercises, tailored to patients’ cognitive capacity during the ICU and early post-ICU phases. Furthermore, elements of Mindfulness-Based Stress Reduction [46] informed the intervention’s emphasis on present-moment awareness and attentional control, which may support cognitive engagement and emotional regulation during recovery.

The intervention prototypes did not rely on full cognitive recovery but were designed to promote early engagement by directing the patient’s limited attentional resources toward the exercises.

The theoretical rationale underpinning this approach was supported by the cognitive pyramid model proposed by Stuss and Benson [47], which conceptualises cognitive functioning as a hierarchical system. Basic functions such as arousal, attention, and processing speed serve as foundational elements that support higher-order cognitive processes. While ICU care traditionally prioritises the bottom-up recovery of basic functions, higher-order cognitive abilities are rarely addressed systematically during ICU admission. However, early and graded engagement in simple, goal-directed cognitive activities may help maintain attentional orientation, support emerging executive functions, and reduce the risk of cognitive decline during critical illness. From a rehabilitation perspective, initiating cognitive engagement at this stage may also facilitate continuity of cognitive rehabilitation that can be sustained and further individualised across the ward and post-discharge phases. These assumed mechanisms of action, including enhanced attention, increased engagement in meaningful activities, and improved confidence in managing cognitive challenges are illustrated in Fig. 2. Accordingly, the intervention was designed to operationalise these theoretical assumptions by providing simple, structured cognitive activities that could be initiated during ICU admission and adapted across the recovery trajectory.

Fig. 2.

Logic model illustrating the rationale, mechanisms, and expected outcomes of the ICU CogHab intervention

Figure 2 presents a logic model illustrating the theoretical rationale, mechanisms, and anticipated outcomes of the ICU CogHab intervention. The intervention is part of ICU CogHab - a cognitive rehabilitation programme developed to support ICU survivors experiencing cognitive impairments and rehabilitation needs. The model is informed by cognitive rehabilitation theory, prior evidence from neurological rehabilitation, and stakeholder input. It outlines the intervention components (Mindfulness and Brain Training) and activities, their mechanisms of action, and expected short- and long-term outcomes. The model also reflects implementation assumptions, which will be evaluated in the feasibility study.

Stakeholder involvement (stage 1 and 2)

A broad group of stakeholders was invited to participate in the development process to enhance the intervention’s relevance, acceptability, and feasibility, thereby ensuring alignment with clinical practice as well as with theoretical and evidence-based principles in cognitive rehabilitation and intervention development [35, 36, 48]. Stakeholders were involved through three processes: A workshop, a Delphi process, and expert consultations.

A workshop was initially conducted, involving 11 stakeholders (five HCPs, three former patients, and three relatives). The workshop was moderated by an experienced ICU research nurse who introduced the agenda facilitated overall discussion. Participants were allocated to three mixed groups and given a time to exchange experiences, promote creative idea generation, and explore perceived needs related to cognitive rehabilitation after critical illness. A member of the research group was present in each group to support the discussion process and to document suggestions through structured notetaking. Notes focused on proposed intervention components. Following the group discussions, key suggestions were summarised in a plenary session, allowing participants to reflect on similarities and differences across groups.

Following this, a three-round Delphi process involving 39 mixed stakeholders (21 HCPs, seven researchers, five patients and six relatives) was undertaken. The first round was exploratory and open-ended, allowing panellists to propose potential interventions to support cognitive function during and after ICU admission. Responses were analysed and synthesised into a structured list of candidate intervention components. In the subsequent two rounds, panellists rated the importance of each proposed intervention using a seven-point Likert scale, ranging from “very unimportant” to “very important” [49]. Participants were also able to provide written comments to justify their ratings or suggest refinements. Consensus was defined a priori as ≥80% of participants rating an intervention as either “important” or “very important.” Interventions reaching this threshold in the second and third Delphi rounds were retained and further discussed in subsequent stages of the development process, in accordance with established Delphi methodology [50].

Co-design of the intervention (stages 1 and 2)

The workshop and Delphi process were further supplemented by a series of expert consultations (e.g. experts within mindfulness and neuropsychology). These consultations were conducted through structured meetings with members of the research group to review and refine the intervention components identified in the earlier phases. Findings from the workshop and Delphi were discussed iteratively with focus on feasibility, clinical relevance, and potential mechanisms of action. During this process, proposed intervention elements were refined, adapted, and operationalised into preliminary prototype formats.

Prototyping the intervention (stages 2 and 3)

The initial expert feedback was reviewed by the research group, followed by additional consultations with other relevant clinical and methodological experts to refine and finalise the proposed intervention prototype. These iterative discussions focused on aligning the intervention content with theoretical assumptions, clinical feasibility, and anticipated implementation within routine ICU practice. The refinement process also involved developing introductory materials to familiarise patients and ICU nurses with the intervention, and revising intervention materials to improve clarity, usability, and perceived relevance across recovery phases.

Data analysis

Qualitative data from the workshop discussions and open-ended responses in the Delphi rounds were analysed using inductive content analysis inspired by Elo and Kyngäs [51]. Workshop data were organised through structured notes, while written comments from Delphi participants were extracted directly from the survey platform. The analysis proceeded through stages of open coding, grouping in topics, and abstraction. The main and last author conducted the primary work, and emerging interpretations were discussed iteratively within the research group to enhance analytical rigor and ensure relevance to the intervention development process. Initially, meaningful categories related to perceived intervention needs, feasibility considerations, and contextual experiences were identified. These categories informed the subsequent prioritisation and refinement of intervention components for the further process.

Quantitative data from the Delphi survey were analysed using descriptive statistics to identify the proportion of ratings across each intervention category. All quantitative analyses were conducted using STATA 19.5 (StataCorp, College Station, TX, USA).

Ethical considerations

The study was conducted in accordance with the Declaration of Helsinki [52] and approved by the Research Ethics Committee of the Central Denmark Region (file no. 1-10-72-137-23), which covers this study and the subsequent feasibility studies. In addition, the study was registered with the legal office of the Central Denmark Region (file no. 1-16-02-133-23).

Participants in the workshop and Delphi process were recruited from a pool of former ICU patients and relatives who had previously given informed consent to be contacted for research purposes. In accordance with Danish law and The Danish Healthcare Act §42, surveys and workshops of this nature do not require formal ethics committee approval. All data were stored and handled in compliance with the General Data Protection Regulation (GDPR).

Results

Development process

The development process lasted from November 2022 until May 2024 and followed an iterative and dynamic three-stage process, integrating current evidence, stakeholder input, and considerations regarding implementation (see Fig. 1). These processes, as detailed below, led to the identification of two distinct prototype interventions, as the most promising approaches for further evaluation. These prototypes comprised a Mindfulness intervention and a Brain Training intervention.

The current evidence base (stage 1)

The studies conducted in the first part of Stage 1 included a scoping and an umbrella review of the literature [19, 53], two qualitative studies on patients and relatives’ experiences of cognitive impairments both during and after ICU admission [54, 55], one qualitative study on ICU nurses’ experiences of managing cognitive problems in clinical practice [56], and a concept analysis of the term cognitive impairments in critically ill patients [57]. The studies formed a solid knowledge base and informed the development of the intervention. Specifically, they contributed to identifying core patient and system needs, relevant timing, delivery modes, and contextual barriers and facilitators for cognitive rehabilitation in the ICU setting. Together, the empirical work and conceptual insights (see Table 1) informed the design of the cognitive rehabilitation intervention.

Table 1.

Data sources informing the intervention development

Stage	Evidence base, Study type or process	Aim of the study	Main findings	How the study/process informed the intervention development
Stage 1 – Part 1. Evidence reviews and primary data collection	Scoping review based on 11 included studies [53].	To review the literature on patients’ experiences of cognitive impairment following critical illness treated in an ICU.	Four themes were found: 1. Experiencing poor memory. 2. Managing everyday life. 3. Feeling unsupported by the healthcare system. 4. Strategies for support in recovery.	Overall: • Highlighted key cognitive challenges post-ICU. Concrete - what informed the intervention: • Emphasized need for structured support and information for patients, which was incorporated in the information pamphlets (psychoeducation).
	Qualitative study based on participant observations and interviews with patients (n = 20) and relatives (n = 15) [55].	To explore patients’ and relatives’ experiences of patients’ cognitive impairments while in the ICU.	Cognitive impairments leave patients feeling vulnerable. Relatives provide crucial support. Four themes: 1. Having a hazy memory and a foggy brain. 2. Frustrations due to difficulties in speaking 3. An altered sense of self. 4. Feeling of disconnect between body and mind.	Overall: • Reinforced need for early cognitive support. Concrete - what informed the intervention: • ‘Brain training’ (with exercises targeted the memory & speech difficulties). • ‘Mindfulness’ (with exercises targeted at altered sense of self & improve mind-body connection).
	Qualitative study based on 3- and 6-month follow-up interviews with patients (n = 18) and relatives (n = 14) [54].	To explore patients’ and relatives’ experiences of patients’ cognitive impairment due to critical illness following an ICU admission.	Cognitive impairments affect family, work, and social life. Three themes: 1. It feels like living in a parallel world. 2. Getting back to a normal everyday life with a vulnerable self. 3. Managing everyday life using self-invented strategies.	Overall: • A need for targeted rehabilitation strategies post-ICU. Concrete - what informed the intervention: • With the role of ICU nurses highlighted in early rehabilitation, this supported the intervention to be nurse-led/delivered by the nurses in the ICU.
	Qualitative study based on interviews of ICU nurses (n = 10) [56].	To explore nurses’ experiences of cognitive problems in patients admitted to the ICU.	Nurses see cognitive problems and delirium as interconnected. Three themes: 1. Cognitive problems and delirium are seen as two sides of the same coin. 2. Searching for the person behind the patient. 3. Maintaining a sense of normality in a confusing environment.	Overall: • ICU nurses lack structured cognitive support tools. Concrete - what informed the intervention: • Led to development of informational pamphlets to support communication with patients on cognitive impairments. • ICU nurses play a key role in intervention delivery.
	Concept analysis based on Rodgers’ evolutionary method [57].	To clarify the meaning of the concept of cognitive impairment in critically ill patients throughout the trajectory of their rehabilitation during and after an ICU admission.	Identified terminology, attributes, and progression of impairment. Cognitive impairment can shift from acute dysfunction to long-term impairment.	Overall: • Supplied the theoretical foundation for defining cognitive impairment. • Clarified attributes, antecedents, and consequences of impairment. Concrete - what informed the intervention: • Provided broad inspiration for the pamphlet on ICU patients’ cognitive impairments.
	Umbrella review based on 13 research syntheses [19].	To provide an overall examination of literature concerning non-pharmacological interventions that can enhance cognitive functioning in critically ill patients or facilitate their rehabilitation pathway during and after their ICU admission.	No definitive conclusion on best non-pharmacological interventions for cognitive rehabilitation in the ICU. Four intervention types summarized: 1. Cognitive activities and training. 2. Mobilization and physical exercises. 3. Emotional, psychological and social support. 4. Information.	Overall: • Broad overview of possible interventions. Concrete - what informed the intervention: • The recommendation on developing multicomponent interventions supported the design of the two interventions ‘Brain training’ and ‘Mindfulness’. • The study showed that it is possible to develop interventions that support patients throughout the ICU pathway. • Provided inspiration on how to be transparent about the intervention components and delivery.
Stage 1 – Part 2. Stakeholder involvement and data collection + Stages 2 + 3 Co-designing the intervention	Workshop based on involvement of stakeholders (n = 11).	To exchange ideas and gain an understanding of the end-users’ needs.	During the roundtable workshop with patients, relatives and clinicians many ideas were discussed but no new cognitive interventions were suggested. Findings: Various initiatives and interventions related to information, the ICU environment, as well as physical and cognitive rehabilitation were identified.	Overall: • Identified patient transition challenges. Concrete - what informed the intervention: • Suggestions were considered relevant for further evaluation in a Delphi process.
	Delphi process based on involvement of different stakeholders (patients, relatives and clinicians, n = 39) in the Delphi panel.	Brainstorm and develop creative ideas for cognitive rehabilitation interventions.	Completion rates were: 100% (Round 1), 92% (Round 2), 85% (Round 3). Findings: Existing interventions were favored, with key focus areas: • Information (enhancing transparency and education). • ICU environment (optimizing surroundings for cognitive recovery). • Early mobilization and physical activity.	Overall: • The most novel suggestions did not reach 80% cut-off between round two and three. • Highlighted the difficulties with providing new and creative ideas on cognitive rehabilitation. Concrete - what informed the intervention: • The research development group further discussed how to integrate new rehabilitation strategies, particularly the novel approaches suggested in the first round.
	Expert consultations with different health care professionals with experience in the rehabilitation of patients with neurological conditions or mild brain injury.	To obtain expert input on cognitive rehabilitation interventions supplementary to usual care.	A neuropsychologist recommended: • Structured rehabilitation through cognitive remediation and psychoeducation for patients beyond the acute phase. A neuropsychologist recommended: • Stress reduction, using/by mindfulness (also suggested in the first Delphi round).	Overall: • These suggestions were in line with those that did not reach 80% agreement on importance in the Delphi round. Concrete - what informed the intervention: • The research group selected mindfulness and cognitive remediation as core interventions, supported by psychoeducational materials(pamphlets on cognitive impairment and post-ICU challenges).
	Expert consultations with two specialists in mindfulness and two experts in neurorehabilitation.	To refine the selected interventions and involve experts in co-design.	• Mindfulness experts suggested keeping Mindfulness and Brain Training as separate interventions due to their different mechanisms. • Suggested short (≤20 min) guided mindfulness exercises. • Recommended practical cognitive exercises (e.g., reading, writing, problem-solving, games etc.).	Interventions were finalized as two separate interventions: 1. Mindfulness (passive, stress-reduction). 2. Brain training (active engagement, skill enhancement, and memory support).

Table 1 provides an overview of the data sources that informed the development of a cognitive rehabilitation intervention targeting ICU patients.

Stakeholder contribution (stage 1 and 2)

Both the workshop and the Delphi process generated numerous suggestions for interventions important to ICU patients. Some approaches were novel in the general ICU setting, while others were already established as part of usual care to varying degrees. Novel interventions suggested included Virtual Reality, Mindfulness, Guided Interaction Therapy (The Affolter Model) [58], and Goal Management Therapy [59]. Furthermore, existing treatments used in neurological rehabilitation were proposed, including simple exercises, tasks, and games, the use of pictograms, reading and writing exercises, sensory stimulation training, and interest-based training.

Examples of established approaches included activity-based training, strategies to support circadian rhythms (e.g. light exposure using natural and artificial light), orienting cues (e.g. visible clocks and calendars), the use of communication aids, environmental adaptations of the patient room, noise reduction initiatives, music therapy, outdoor excursions, and early mobilisation. Additionally, some of the suggestions were not directly related to cognitive rehabilitation but focused more on providing support and information. Some interventions were identified as important at different stages of the recovery process, with some relevant during ICU admission, others during ward admission, and some only after discharge to home.

Most interventions that reached the ≥80% cut-off in the third round of the Delphi process were, to varying degrees, already incorporated into usual ICU care; some were standard practice, while others were used sporadically or only in specific patient groups. Among the novel interventions that did not reach the ≥80% cut-off after the second Delphi round, Mindfulness and Brain Training still received moderate support from the panel. Agreement for Mindfulness was 25% for ICU use, 32% for general ward-based use, and 49% for post-discharge use, while corresponding ratings for Brain Training were 32, 65, and 78%, respectively (see more in Additional file 2). As a result, Mindfulness and Brain Training were selected by the research group, based on their perceived clinical relevance, potential mechanisms of action, and feasibility of implementation within existing ICU resources. Expert consultations subsequently informed iterative refinement of these two interventions into preliminary prototype interventions (Fig. 1).

Prototyping the interventions (stage 3)

Based on the processes in Stages 1 and 2, we designed the intervention prototypes, which were specified into two separate interventions: “Mindfulness” and “Brain Training” (Fig. 3). The development of two distinct intervention formats reflected different but complementary theoretical foci. The Mindfulness component primarily aimed to support parasympathetic activation, attentional regulation and present-moment awareness, whereas the Brain Training component was designed to provide structured cognitive stimulation through simple task-based activities targeting attention, memory, and executive functioning. Each intervention was delivered in a separate cardboard box with a distinct colour design.

Fig. 3.

Visual representation of the ICU CogHab intervention and its core elements

The Mindfulness box contained a pamphlet with QR codes linking to three different audio recordings: one specifically designed for ICU patients by mindfulness experts for this study, and two additional audio files recommended and delivered by these experts. The audio files varied in length from approximately ten minutes to a maximum of 30 minutes. This box also included a set of headphones with an adapter for smartphones or tablets. Additionally, a website link providing access to further audio recordings was shared upon request.

The Brain Training box contained an activity book with a range of cognitive exercises, including reading, crosswords, structured colouring tasks (e.g., patterns and figures), simple arithmetic tasks, and various games. The exercises were of varying difficulty levels to accommodate the patients’ diverse needs and skill levels. The book’s final section included recommendations for similar exercises or more advanced tasks. Additionally, the box included a week planner to help track appointments, set goals, and record personal notes. Furthermore, the Brain Training box contained a stress-relief squeeze ball, a puzzle, a deck of cards, and coloured pencils.

Both boxes also included two informational pamphlets developed explicitly for the study: one providing information about common post-ICU symptoms and self-management strategies, and another explaining cognitive impairments and offering guidance on how patients can manage these challenges independently. The design of the activity book, calendar, pamphlets, and box labels was developed in collaboration with the graphics department of the lead hospital site, based on sketches provided by the research group.

The prototype versions were reviewed by ICU nurses, the research team, and one ICU patient to gather feedback on usability and content. The feedback informed the refinement of the interventions to improve face validity alignment, ultimately resulting in two final intervention prototypes for feasibility testing. The intervention activities were designed to be delivered in short, flexible sessions adapted to patients’ energy levels and clinical stability. This was typically achieved through brief daily engagement during ICU admission with continued use after discharge, with progression in frequency and duration guided by patients’ recovery and tolerance. To facilitate transfer to everyday functioning, the intervention components were designed to introduce simple cognitive strategies that could be applied in meaningful daily activities, such as maintaining orientation in time and place, structuring daily tasks, and supporting attentional focus during routine interactions. Mindfulness exercises were intended to promote emotional regulation and present-moment awareness, whereas the Brain Training activities aimed to encourage graded cognitive engagement across care settings. The use of accessible training materials and involvement of relatives were further intended to support continuation and adaptation of these strategies during recovery at home. A detailed description of the interventions is provided in the Additional file 1 (TIDieR).

Figure 3 presents the two intervention boxes developed for ICU patients [1]: a Mindfulness Box, containing a pamphlet with audio-guided mindfulness exercises (accessible via QR codes), headphones, a mobile phone adaptor; and [2] a Brain Training Box, which included cognitive training exercises (an activity book), a week planner, coloured pencils, deck of playing cards, a puzzle, and a squeeze ball. Both boxes also include recovery-oriented information pamphlets: one titled ‘Cognitive Difficulties’ addressing cognitive impairments, and another titled ‘Life after Intensive Care’, describing patient experiences and common sequelae.

Mode of delivery or use during test phase

The interventions were designed for continuous use from ICU admission through six months post-discharge, based on evidence of persistent cognitive impairments following critical illness [3, 4]. Each patient will receive one of the two interventions, either Mindfulness or Brain Training, allocated at randomisation, and provided with a corresponding intervention box (see Additional file 1 for detailed content and instructions). As part of the development process, strategies for delivering the intervention in clinical practice were defined (see also Fig. 2). These include introducing ICU nurses to the intervention through group sessions and, when needed, individual instruction. An instructional video and the possibility of on-site support from the project team were also developed to assist nurses in facilitating patients’ daily use of the intervention. To support patient engagement, the materials are designed for introduction at the point of ICU enrolment through verbal explanation, supported by written material and a QR code linking to a short video designed to enhance understanding and usability. As part of the intervention delivery, use during ICU admission will be supported by ICU nurses when patients are able to engage. Continued use after discharge will be supported either with assistance from relatives or, for approximately half of the patients allocated to the intervention arm, with additional support from the project team. Patients will be encouraged to engage with the intervention activities at least once daily, preferably during daytime hours, with session duration adapted to needs and tolerance. Flexibility in frequency and intensity of use will be emphasised, allowing adaptation to individual capacity and recovery trajectory.

Safety concerns

The intervention is designed to be simple, user-friendly, and supported by brief, clear instructions. Mindfulness-based interventions for stress reduction have been extensively studied and are generally considered safe, for example, among individuals with psychiatric disorders [60]. Based on this evidence, no specific safety concerns were anticipated regarding the use of mindfulness with ICU patients, either during admission or in the post-ICU period. Likewise, cognitive exercises have been demonstrated to be feasible and well-tolerated in critically ill patients. For instance, a study by Wassenaar et al. [26] reported no adverse effects associated with cognitive exercises during ICU admission, and similar findings have been observed in post-ICU rehabilitation settings [21, 61]. This supports our assumption that including structured cognitive exercises is safe for this patient population. While adverse effects are uncommon and typically mild, such as transient fatigue or frustration, we acknowledge that they may occur, particularly in individuals with cognitive or emotional vulnerabilities. These potential effects will be closely monitored during the clinical evaluation of the intervention.

Discussion of the development process and intervention

This study presents a transparent, systematic, stakeholder-informed development of two cognitive rehabilitation intervention prototypes, Mindfulness and Brain Training, tailored for ICU patients and designed to span the continuum from ICU admission to post-discharge. Guided by the MRC framework and informed by clinical expertise, patient experiences, and existing neurorehabilitation theory, the development process integrated iterative feedback from former patients, relatives, HCPs, and researchers through a workshop, Delphi rounds, and expert consultations. Similar to our process, previous studies have also developed and tested cognitive exercises [26] and mindfulness interventions [62] for ICU patients. These interventions were feasible and not burdensome to ICU patients [26] and were associated with improved well-being [62]. However, none of these interventions were designed to span the entire care pathway from ICU to home, a dimension of increasing interest in ICU research and practice [19]. This study aimed to address that gap. Whether this full-pathway design is suitable and feasible remains to be seen, pending clinical evaluation.

The growing interest in post-ICU rehabilitation has led to an increasing number of studies evaluating complex interventions delivered after hospital discharge. However, as emphasised in a recent systematic review by Sloan et al. [63], the existing evidence base remains methodologically limited and inconsistent, particularly concerning the development and implementation processes of such interventions. The review identified frequent omissions in the reporting of development procedures, fidelity measures, recruitment strategies, and safety considerations. Notably, only a few studies referenced established intervention development frameworks, despite evidence suggesting that such frameworks can improve effectiveness, implementability, and scalability of interventions [35, 64–66]. Of the 71 studies reviewed by Sloan et al. [63], only six studies (10%) explicitly employed the MRC framework. The updated MRC framework offers structured and practical guidance for the development and evaluation of complex interventions [35]. It supports researchers in selecting appropriate methods and perspectives and highlights six core elements to be considered across the iterative phases: (1) context, (2) development of a programme theory, (3) stakeholder involvement, (4) identification of key uncertainties, (5) ongoing refinement of the intervention, and (6) economic considerations. In line with the updated MRC framework, the following discussion reflects on how the six core elements were addressed during the development of the ICU CogHab intervention.

1. Context: Addressing contextual complexity was particularly challenging, as the ICU CogHab intervention was designed for delivery across multiple settings from the ICU to post-discharge environments. The ICU patient population represents a highly heterogeneous group in terms of diagnosis, severity of illness, the extent of cognitive impairments and rehabilitation needs, all of which may influence both implementation fidelity and intervention outcomes. To inform the intervention’s design and optimise contextual relevance and fit with the current practice, we drew on the findings from the previous exploratory studies [54–56] and reviews [19, 53]. These collectively highlight the importance of initiating cognitive rehabilitation during ICU admission, similar to physical rehabilitation, to ensure that the intervention is established prior to ICU discharge.

2. Development of programme theory: To guide development and visualise how the intervention was expected to work, we defined a programme theory accompanied by a logic model, in line with current recommendations [67]. Although Sloan et al. [63], in their systematic review, found that all included studies presented a programme theory or rationale, only 8% included a logic model. We emphasised both elements in the present study to support transparency and to delineate the causal assumptions underpinning intervention components and outcomes.

3. Stakeholder involvement: While stakeholder involvement is increasingly recognised as essential to intervention relevance and sustainability in development studies, Sloan et al. [63] found that only 39% of reviewed studies reported such involvement, and when present, it was often poorly described, with limited information on the methods used and at which level stakeholders were involved. In contrast, our approach was characterised by systematic, multiphase engagement of diverse stakeholders, including former ICU patients, relatives, clinicians, and researchers. Through a workshop, Delphi process, and expert consultations, stakeholders contributed to the prioritisation and refinement of intervention components. Nevertheless, the workshop and Delphi process may have challenged participants’ creativity by focusing on components perceived as feasible within current practice, which may have limited the generation of novel or broader ideas in favour of practical strategies. Additionally, the predominance of clinicians in the Delphi panel may have introduced selection bias. Still, combining insights from the stakeholder processes with empirical findings from earlier studies helped ensure that the intervention reflected lived experiences, patient-identified needs, and preferences. This approach aimed to enhance early ownership among ICU nurses and facilitate future implementation, consistent with recommendations that advocate for early stakeholder involvement in the development of sustainable, scalable interventions [35, 37]. The level of stakeholder integration applied in this study represents a broader involvement compared to several previous influential ICU intervention studies, where the interventions were primarily developed based on expert consensus and literature reviews [21, 68, 69].

4. Key uncertainties and 5) Refinement of the intervention: The continuity embedded in the ICU CogHab intervention, starting cognitive rehabilitation in the ICU and continuing after discharge, represents a novel contribution. This longitudinal design aligns with emerging evidence indicating that early and sustained interventions may be critical in mitigating the long-term cognitive sequelae associated with PICS [19]. Due to limited existing evidence on optimal dosing (e.g., frequency, intensity), decisions regarding intervention recommendations were based on consensus within the research group. The multi-component nature of the intervention, while intended to be simple and accessible, may impose a cognitive burden on patients already experiencing deficits, particularly in the ICU phase, potentially affecting adherence and engagement. Nonetheless, previous research supports the feasibility and potential benefits of similar interventions [26, 62], justifying further evaluation. Moreover, acknowledging that a ‘one-size-fits-all’ model is unlikely to succeed, we intentionally designed the ICU CogHab intervention with inherent flexibility, allowing for individual adaptation based on cognitive capacity and contextual factors. In the forthcoming feasibility study, we are particularly interested in examining key uncertainties related to fidelity, acceptability, and contextual fit. To support transfer to everyday functioning, the intervention incorporates simple, functionally oriented strategies such as maintaining orientation, structuring daily activities, and supporting attentional focus in routine situations. These strategies were intentionally embedded within both the Mindfulness and Brain Training components to facilitate gradual application in real-life contexts and to support continuity of cognitive engagement across the recovery trajectory. By promoting repeated use in familiar environments and encouraging involvement of relatives, the intervention is intended to enhance generalisation and functional integration of cognitive skills beyond the clinical setting.

5. Economic considerations: While less prominent in our development process, economic considerations were addressed at a general level. For example, we prioritised a low-cost delivery and integration into routine care to minimise additional resource demands. However, as this study focused on early development, economic evaluations were not prioritised but remain essential for later-stage testing and scaling.

Limitations of the study

Despite the systematic and stakeholder-informed development process, several limitations should be acknowledged. First, the predominance of clinicians in the Delphi panel may have introduced selection bias, and the structured format of stakeholder engagement may have constrained the generation of ideas. Second, key intervention parameters, such as optimal dose and intensity, were based on research group consensus due to limited existing evidence. Third, while economic considerations were incorporated at a general level, no formal cost-effectiveness evaluation was conducted. Furthermore, a limitation relates to the decision to retain and further develop intervention components that did not reach the predefined ≥80% consensus threshold in the Delphi process. Although this decision was guided by qualitative stakeholder input as well as considerations of clinical relevance, novelty, and feasibility, it represents a pragmatic deviation from the initial consensus criterion. This introduces an element of researcher judgement in prioritising intervention components and should be considered when interpreting the development process. Finally, the intervention was developed within a Danish ICU context, which may limit its immediate generalisability. Cultural and contextual adaptations will be necessary before implementation in other healthcare settings.

Conclusion

This study describes a structured, evidence-informed, theory- and stakeholder-driven development process for two cognitive rehabilitation interventions - Mindfulness and Brain Training - designed to support ICU patients’ early cognitive engagement and facilitate continuity of rehabilitation from ICU admission through post-discharge recovery. By introducing simple attentional regulation strategies and task-based cognitive activities that may be integrated into daily routines, the interventions are expected to facilitate patients’ gradual return to meaningful everyday functioning following critical illness. This work provides a methodological foundation for subsequent feasibility evaluation of acceptability, fidelity, and implementation, before broader clinical adoption can be considered.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

ICU

Intensive Care Unit

ICU CogHab

Intensive Care Unit Cognitive Rehabilitation

PICS

Post-Intensive Care Syndrome

MRC

Medical Research Council

GUIDED

Guidance for reporting intervention development studies in health research

TIDieR

Template for Intervention Description and Replication

PADIS

Clinical Practice Guidelines for the Prevention and Management of Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption in Adult Patients in the ICU

HCP

Healthcare professionals

GDPR

General Data Protection Regulation

Author contributions

KAS led the overall study conceptualisation, design, coordination, data analysis, and communication with the development group (including AH, PD, NR, HKN, HIJ, ASÅ, ABA, HS, MM and ALH). PD, AH, HKN, and NR provided continuous supervision and critical feedback throughout all phases of the intervention development. KAS, HS, AH, and ALH conducted the initial workshop; KAS led the Delphi process and expert consultations, with PD and AH participating in selected consultations. KAS developed all intervention materials with input from all authors. Pamphlets were co-developed by KAS, MM, and AH and reviewed by the whole group. HS, JD and MOC contributed specifically to the adaptation of the mindfulness component for the ICU setting. All authors participated in the critical appraisal of the intervention and contributed to iterative refinement. KAS drafted the manuscript, and all authors contributed to revisions, approving the final version.

Funding

This study was funded by the Novo Nordisk Foundation, Denmark [NNF20OC0061394], and Aarhus University, Denmark [2022-0367-531].

Data availability

The data supporting this article are available from the first author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki [52] and approved by the Research Ethics Committee of the Central Denmark Region (file no. 1-10-72-137-23), which covers this study and the subsequent feasibility studies. In addition, the study was registered with the legal office of the Central Denmark Region (file no. 1-16-02-133-23). Participants in the workshop and Delphi process were recruited from a pool of former ICU patients and relatives who had previously given informed consent to be contacted for research purposes. In accordance with Danish law and The Danish Healthcare Act §42, surveys and workshops of this nature do not require formal ethics committee approval. All data were stored and handled in compliance with the General Data Protection Regulation (GDPR).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.