Post-intensive care unit clinics: models and implementation - a systematic review

Abstract

Background

Advances in critical care have shifted the focus from survival alone to addressing Post-Intensive Care Syndrome (PICS), which includes persistent physical, cognitive, and psychological challenges after discharge from the intensive care unit (ICU). While post-ICU clinics have been established in high-income countries (HICs), their adoption in low- and middle-income countries (LMICs) remains limited, with structured follow-up care still under development.

Objective

To systematically review models of post-ICU clinics, examine barriers and facilitators to their implementation, and explore their potential applicability in LMICs.

Methods

This review was prospectively registered with PROSPERO (CRD42024536147) and conducted according to PRISMA guidelines. A comprehensive search of Medline, Embase, and CINAHL was completed on April 24, 2024. Studies published after 2000 describing adult post-ICU clinic models addressing PICS were included. Nineteen studies—comprising randomized controlled trials, observational studies, and quasi-experimental designs—met inclusion criteria. Risk of bias was assessed using the Joanna Briggs Institute checklists. Thematic synthesis was guided by the Consolidated Framework for Implementation Research (CFIR).

Results

Three primary models emerged: (1) hospital-based physical interviews (2), hybrid models incorporating both in-person and telehealth consultations, and (3) fully remote models using telehealth or home visits. Telehealth/home-visit models reported the highest mean attendance (88.7%), followed by hybrid (59%) and physical interview models (51.9%). Common barriers included resource constraints, transportation difficulties, limited awareness, inadequate insurance coverage, and poor interdisciplinary coordination. Facilitators included flexible scheduling, early stakeholder engagement, multidisciplinary team involvement, and use of telehealth technologies. While hybrid models appeared promising for LMICs due to their balance of accessibility and comprehensiveness, the evidence for clinical outcome benefit remains inconclusive, and questions about cost-effectiveness and sustainability persist.

Conclusion

Hybrid post-ICU clinic models may offer a feasible pathway for improving follow-up care in LMICs, especially when tailored to local constraints. However, their implementation must consider significant barriers, particularly related to funding and infrastructure, and be guided by emerging but still limited evidence on long-term patient outcomes. These findings aim to inform cautious, context-specific development of post-ICU care strategies in resource-limited settings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13054-025-05634-x.

Introduction

Advances in critical care [1] have led to a paradigm shift from merely ensuring survival to improving post-discharge functionality and long-term outcomes for patients and their families. Post-Intensive Care Syndrome (PICS) [2, 3] describes new or worsening problems in physical, cognitive, or mental health status that arise following a critical illness and persist beyond hospitalisation. Various strategies [4–7] have been proposed to support ICU survivors, including nurse navigators, pulmonary rehabilitation, ICU diaries, peer support networks, and post-ICU clinics, also known as ICU recovery clinics.

Post-ICU clinics are crucial in assessing PICS symptoms, providing consultations with clinicians specialised in PICS, reviewing ICU courses (including ICU diaries where available), and connecting patients to appropriate support services such as psychiatry, cognitive or occupational therapy, and physical rehabilitation [8]. The first PICS follow-up clinic, “Intensive aftercare after intensive care,” was established in Reading [9] in 1993. Approximately 30% of ICUs in the UK conduct outpatient follow-ups [10], with national guidelines [11] recommending reviews two to three months post-discharge. In the U.S., the first PICS follow-up clinic [12] was launched in 2011. Notwithstanding the equipoise within the scientific community over the impact of these services on patient outcomes, with studies reporting a reduced mortality [13], physical benefit [14], no effect on QOL [14–23], to worsening QOL [24] the concept is gaining traction in both high and low middle income countries (HICs and LMICs)with the Dutch [25] and Indian Society of Critical Care Medicine recently issuing a position statement [26].

Building upon our prior work [27–29] on post-ICU patient outcomes, we conducted a literature review to inform the establishment of a post-ICU clinic within the context of a publicly funded teaching hospital in an LMIC. While systematic reviews have underscored the importance of PICS over long term follow up [30], and that of follow-up systems their impact on patient outcomes [31], and the feasibility of interventions [32], few studies have detailed the specific design, assessment methods, and operational processes of post-ICU clinics, particularly in resource-limited settings. Most reviews have focused on isolated interventions, such as ICU diaries or physiotherapy clinics, rather than a holistic post-ICU model. Therefore, further exploration of innovative program delivery methods, particularly telemedicine-driven models [33], is necessary.

This systematic review aims to examine the various post-ICU clinic models that cater to the complete spectrum of post-ICU patient care—encompassing physical, cognitive, and mental health aspects—and identify the barriers and enablers to their successful implementation, to establish a post-ICU clinic at our centre.

Materials and methods

This systematic review was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO), ID-CRD42024536147, and conducted following PRISMA [34] guidelines.

Research questions

What are the different post-ICU clinic models reported in the literature (with a specific focus on reports from LMICs?

What are the facilitators and barriers to implementing these post-ICU clinics successfully?

Inclusion criteria

We included studies that reported models catering to adult patients and offering physical, psychological and cognitive follow-up interventions post-ICU discharge. Eligible studies identified target populations, intervention designs, outcome measures, involved professionals, duration of follow-up, and assessment tools for PICS evaluation. Process evaluations discussing barriers and facilitators to implementation were included, even if they lacked effectiveness data.

Exclusion criteria

Studies were excluded if published before 2000, were not in English, involved only paediatric patients, lacked clear intervention descriptions, or did not address comprehensive PICS follow-up.

Eligible study designs

This review included RCTs, observational cohort studies, and descriptive non-experimental publications such as editorials.

Comprehensive search strategy

Studies were retrieved from Medline, Embase, and CINAHL (search date: April 24, 2024), using relevant search terms. Search terms included: design or model or development or strategy or set-up or planning, implement* or service utilisation or usage or operation or function*, Post-ICU Clinics or post-intensive care outpatient clinic or post-ICU recovery clinic or post-critical care clinic or ICU follow-up clinic or post-ICU recovery centre. The complete search strategy is outlined in the supplementary material (Supplementary Document 1: Chart 1). Lists of selected articles were screened for additional citations. Support was sought from the librarian for extraction and search as required.

Data extraction and analysis

Two reviewers (SC, SN) assessed study eligibility and extracted data using a standardised form. Discrepancies were resolved by discussion (ST, BR). Data extracted were: author name, year of publication, the country where the study was done, type of study, journal in which it was published, research funding, start and end date, duration of the study, urban or rural setup, aim of the study, nature of the hospital, type of ICU included, the specific name of the project, departments and persons involved, education provided, inclusion and exclusion criteria to enrol in the study, person(s) responsible for enrolment and follow up, mode of contact for fixing appointments, baseline demographics of patients, the timing of the first clinic appointment, frequency of interviews, mode of interview, number of enrolments and attendance, percentage of attendance, causes of nonattendance, type of patient assessment performed, screening tools used, barriers and facilitators, and any modifications made. For the quantitative studies, data were presented as means, ranges, or numbers and proportions as appropriate. Study outcomes included the attendance rate of participants at the end of the final assessment, dimensions of health measured, and screening tools used for assessment. No assumptions were made about any missing or unclear information.

Qualitative data were compiled and analysed thematically according to the Consolidated Framework for Implementation Research (CFIR) [35]. The CFIR offers a structured set of constructs organised into five domains, applicable across various contexts. This practical framework supports the systematic identification of potential barriers and facilitators, customising implementation strategies, making necessary adaptations, and explaining outcomes.

Quality assessment

The risk of bias was assessed using the Joanna Briggs Institute (JBI) critical appraisal checklists [36].

Results

Study selection

After removing duplicates, 122 studies were screened for eligibility. Twelve studies were included from hand searching and snowball techniques. After exclusion of studies that were irrelevant or did not discuss all components of post-ICU care, 19 studies were included for data synthesis in this systematic literature review. (See Fig. 1. PRISMA flow diagram.)

Fig. 1.

PRISMA flow diagram

Risk of bias

Four RCTs had a low and four a moderate risk of bias. All ten observational studies, except one and the final quasi-experimental study had a low risk of bias. (Supplementary document 1: chart 2). The majority of the RCTs had concerns related to domains 4 and 5, specifically blinding of participants and personnel. Observational studies had concerns with domains 3 and 8 pertaining to group allocation and follow-up losses.

Characteristics of included studies : (supplementary document 2)

Details of the 19 studies, including the ICU clinics’ characteristics such as country, geographic locale, and type of institute, as well as the types of duration and study design, names of the clinics with their inclusion criteria, team composition, and follow-up processes, are detailed in Table 1. Three distinct models for patient follow-up in Post ICU Clinics could be identified based on the mode of interview performed:

Table 1.

Summary of characteristics of different studies

Parameter	Details
Countries of Origin	• USA: Seven studies [17, 38, 39, 41, 42, 46, 51] • UK, Europe, Australia, New Zealand: [15, 16, 21, 40, 43–45, 47, 48, 50] • Low/Middle-Income Country: 1 study in Argentina [49]
Type of Studies	• RCTs: Eight studies [15, 16, 19, 21, 44–47] • Observational: Ten studies [17, 38–42, 49–51] • Quasi-Experimental: One study [48]
Aim of Studies	• Exploring outcomes among COVID-19 survivors [40, 49–51] • Describing the design, implementation, barriers and facilitators of ICU recovery programs [38, 39, 44] • Transition to primary care [42] • Examining the effects of ICU recovery project [15–17, 19, 21, 46–48]
Urban or Rural	• Urban: Eleven studies [15, 17, 38, 39, 43, 45–48, 50, 51] • Urban and Rural/Semi-urban: Five studies [16, 21, 40, 42, 44] • Unspecified: Three studies [19, 41, 49]
Type of ICU	• Medical ICU: Three studies [38, 42, 46] • General/Multiple ICUs: Twelve studies [15–17, 19, 21, 39, 41, 43–46, 48] • COVID ICU: Four studies [40, 49–51]
Location of Clinic	• Academic/University Hospitals: Ten studies [30, 38–40, 42, 44–46, 48, 51] • Mixed (Academic and Non-Academic Hospitals): Seven studies [15–17, 19, 21, 47, 49] • Unspecified: One study [41] • Non-Academic Tertiary Centre: One study [40]
Name of Clinic	• ICU Recovery Clinic: Most common [38, 42, 46] • ICU Follow-up Clinic [44, 47]
Timeline of Studies	• Before 2010: Four studies [15–17, 47] • Pre-COVID pandemic (after 2010): Ten studies [19, 21, 38, 39, 41–43, 45, 46, 48] • Post-COVID pandemic: Five studies [40, 44, 49–51]
Frequency of Follow-Up (supplementary document 1: chart 3)	• One-Month Post-Discharge: Most studies [21, 38, 40, 41, 49, 51] • Early or Baseline Assessment: Some studies [16, 46–48] • Different Time Points: Multiple follow-ups [17, 38, 39, 43, 45, 46, 51]
Duration of Follow-Up	Ranged from 7 weeks [49] to more than 3 years [16, 19, 21, 39, 43, 48]
Departments Involved	Multidisciplinary Approach: Common except one study [16] (Including ICU, Pulmonology, Physical Medicine, Rehabilitation, Psychiatry, Psychology, Neuropsychology, ICU-trained Nurse, Research Nurse, Nutritionist, Social Worker, Biostatistics)
Team Leader	• ICU Physician: Most common [17, 29, 38, 40–44, 51] • Physiotherapists: Two studies [16, 19] • Nurse-Led Projects: Three studies [15, 21, 48]
Time of Recruitment	• At ICU discharge: [15–17, 21, 39, 42–46, 48] • At hospital discharge: [38] • After hospital discharge: [40, 43]
Person/Team responsible for Recruitment	• ICU Physician or ICU Nurse/Health worker: Most common [21, 38, 40–51] • Independent Nurse Practitioner: [46] • NHS Pathway (UK): In some studies [15] • ICU Recovery Center Coordinator: [39]
Health Education	• Introductory Letter: [38] • Health Education Flyers or Materials: [39, 42–44, 46] • Physical Rehabilitation Handbook: [15, 16] • Self-help Manual: [47]
Method of Scheduling	• Telephone-Based: All studies • Reminder Mail additional to telephone call: One study [38]
Frequency of Follow-Up (supplementary document 1: chart 3)	• One-Month Post-Discharge: Most studies [21, 37, 39, 40, 48, 50] • Early or Baseline Assessment: Some studies [16, 45–47] • Different Time Points: Multiple follow-ups [17, 37, 38, 42, 44, 45, 50]
Duration of Follow-Up	Ranged from 7 weeks [49] to more than 3 years [16, 19, 21, 38, 42, 47]
Departments Involved	Multidisciplinary Approach: Common except one study [16] (Including ICU, Pulmonology, Physical Medicine, Rehabilitation, Psychiatry, Psychology, Neuropsychology, ICU-trained Nurse, Research Nurse, Nutritionist, Social Worker, Biostatistics)
Team Leader	• ICU Physician: Most common [17, 29, 37, 39–43, 50] • Physiotherapists: Two studies [16, 19] • Nurse-Led Projects: Three studies [15, 21, 47]
Time of Recruitment	• At ICU discharge: [15–17, 21, 38, 41–45, 47] • At hospital discharge: [37] • After hospital discharge: [39, 42]
Person/Team responsible for Recruitment	• ICU Physician or ICU Nurse/Health worker: Most common [21, 37, 39–50] • Independent Nurse Practitioner: [45] • NHS Pathway (UK): In some studies [15] • ICU Recovery Centre Coordinator: [38]
Health Education	• Introductory Letter: [37] • Health Education Flyers or Materials: [38, 41–43, 45] • Physical Rehabilitation Handbook: [15, 16] • Self-help Manual: [46]
Method of Scheduling	• Telephone-Based: All studies • Reminder Mail, in addition to the telephone call: One study [37]

Model 1: Physical interview model: Patients need to come to the outpatient clinic [37, 38]of the index hospital (where the patient was admitted) for the first and all subsequent interviews for face-to-face assessments [15, 17, 37–42].

Model 2: Hybrid model: The 1 st assessment is at an outdoor clinic by a face-to-face interview, followed by a tele (phone) monitoring. The final assessment is conducted by face-to-face interview, either at the clinic or at home if the patient is unable to come to the clinic for any reason [20, 43–46]. In some hybrid models, patients had the option to choose between a face-to-face interview at the clinic, a teleassessment [47], or a session at the hospital or home [18, 48].

Model 3: Non-hospital-based model: In this model, patients did not need to come physically to the index hospital. Here, patients were provided physical and psychotherapy sessions and subsequently followed up through telephonic consultations [49, 50], or patients were paid home visits, followed by telephonic follow-up [16].

Participation rate at the time of final assessment was found to be highest in the non-hospital-based (Telehealth/Home visit) model and lowest in the Hospital-based physical interview model. The mean percentage (range) of attendance for Model 3 was the highest, with the three models scoring 51.9% (26–75), 59% (12.6–100), and 88.7% (82.5–92), respectively. (Tables 2, 3 and 4). The criteria for inclusion or exclusion of patients were similar across studies, including ICU length of stay greater than 48 h, sepsis, and multiorgan failure, and excluded patients with terminal illness or those with preexisting neurologic impairments (Chart 4). A wide variety of screening tools were used to diagnose different PICs domains. (Table 5).

Table 2.

Attendance rate of model 1: Hospital-based physical interview model

Author	No of patients enrolled	No of patients attended	% of attendance
Rita N. Bakhru et al. [37]	196	101	51.5%
Carla M. Sevin [38]	307	62	28%
Burçin HALAÇLI [39]	NA	NA	NA
Ariel M Modrykamien [40]	NA	NA	NA
Kirby P. Mayer [41]	59	38	64.4%
B H Cuthbertson [15]	286	192	67.1%
Lise Fonsmark [42]	101	26	26%
James C. Jackson [17]	821	382	75%
	Mean and range		51.9%(26- 75)

Table 3.

Attendance rate of model 2: hybrid model (part in hospital or optional)

Author	No of patients enrolled	No of patients attended	% of attendance
M. Rohr et al. [43]	NA	NA	NA
Sarah L. Bloom [45]	302	38	12.6%
Christina Jones [46]	126	106	84%
Janet F. Jensen [20]	386	205	53.1%
Diane Friedman [44]	546	380	69.5%
Kathryn McDowell [18]	60	49	81.6%
Ezequiel Martínez [48]	22	22	100%
Jónasdóttir [47]	168	105	71%
	Mean and range		59% (12.6-100)

Table 4.

Model 3: Non-hospital-based model (Telehealth/Home visit)

Author	No of patients enrolled	No of patients attended	% of attendance
Miguel A. Martillo	49	45	92%
Alberto Zangrillo [49]	61	56	91.8%
Doug Elliott [16]	195	161	82.5
	Mean and range		88.7% (82.5 −92)

Table 5.

Screening tools used to diagnose different PICs domains

Dimensions of health	Screening tools
Cognitive	Montreal Cognitive Assessment (MOCA): [39–42, 49] Mini Mental Status Assessment: [39, 40] T-MOCA: [51] SOC-13 (sense of coherence): [21]
Depression	Hospital Anxiety and Depression Scale (HADS): [15, 39, 40, 42, 43, 45, 47, 50] General Anxiety Disorder-7 (GAD-7): [49]
Post-Traumatic Stress Disorder (PTSD)	Post-Traumatic Stress Disorder checklist [39, 40] Impact of Events Scale-Revised (IES-R) [47] Harvard Trauma Questionnaire (HTQ-IV)
Physical health	Physical component SF-36 [16, 19, 41, 47, 48] Physical component SF-12 [44] Chair rise test [44] Jamar plus Digital hand dynamometer [40, 44] Medical Research Council (MRC) Scale [40] Modified Rankin Scale (mRS) [51]
Activity of Daily Living (ADL)	Barthel index [45, 49]
Quality of Life (QOL)	EQ-5D-3 L [40, 51] EQ-5D-5 L [42] SF-36 [21]
Respiratory function	6 MWT [16, 39, 48]
Nutrition	Nutritional Risk Screening (NRS)−2002 and Malnutrition Universal Screening Tool (MUST), Mini Nutritional Assessment (MNA-SF) questionnaire [40, 50]
Medication Reconciliation	Medication Reconciliation Form [39, 41]

Barriers and facilitators to implementation of a post-ICU clinic: (chart 5)

The studies reported various enablers for setting up post-ICU follow-up clinics. For instance, clinics that adapted their programs to prior stakeholder experiences [38, 39] of formal or informal follow-up and provided evidence of the need and success of their efforts to patients and carers, catered to their ‘felt needs’, such as by allowing flexibility [37, 43, 50] in visit time and types, were better accepted by the users. If the clinic offered an ‘added advantage’ to patient-carers, such as arranging appointments with parent teams, it was likely to be well received.

Successful process of implementation often included elements of careful planning [17, 37, 39], sustained, deliberate, and early stakeholder engagement [15, 17, 40, 44] with frequent actionable feedbacks [37–39] and timely reminders to patient families to attend clinics [37, 39]. (Implementation Process)

Similarly, aligning the sociocultural values and beliefs of the clinical team, the wider multidisciplinary setting, and hospital administration with the post-discharge needs of critically ill patients encouraged the implementation and delivery of the innovation. These included effective engagement with the community [38] to which the patient returned, availability of transportation services [16, 48], or good internet connectivity [50] that could leverage the flexible scheduling provided by adaptive program policies. Post-ICU clinic designs that received support from health bodies, professional societies [38] or public policies [15, 17, 39] improved reach and sustainability. (Outer setting)

The general characteristics of the setting in which the clinic is established are a critical determinant of implementation success. For instance, in our review, we found that academic settings [37, 39, 40, 45] reported a structured follow-up programme that improved coordination and acceptance. Teams that can ensure an environment of strong internal communication [37, 40, 42, 43, 50] often backed by a supportive organisational climate [44, 48, 49] can run more successful post-ICU programmes with better family involvement and more effective patient tracking. (Inner Setting)

The spectrum of stakeholders —patients, caregivers, and healthcare workers who engaged with the ICU clinic — also affected the implementation. Those who were convinced about the utility of post-ICU follow-up were better motivated and more responsive to appointment reminder systems, and they stayed engaged with teleconsultations, as did stakeholders educated about the long-term benefits of staying engaged. (Individual Characteristics)

In contrast, the reports in this review present various barriers across multiple domains that hinder the establishment of effective post-ICU clinics. Due to a lack of prior research and pilot evaluations, clinics faced challenges in building trust with both patients and healthcare providers. Additionally, rigid protocols and inflexibility hindered their ability to meet the specific needs of patients transitioning from the ICU to follow-up care. The need for interdisciplinary communication and collaboration is crucial, but logistical hurdles [37, 38, 43, 48] complicated this process. (Innovation Domain) Financial constraints [37, 40, 49] and limited funding hindered effective clinic operations, while geographical challenges [37, 38, 40, 42, 48, 50] particularly in rural areas, restricted patient access. Additionally, a lack of insurance coverage [38, 41, 43, 44, 48] and low awareness [49] of recovery programs reduced patient engagement. (Outer Domain) Inadequate resource allocation [38, 40–42, 46] perceived lack of benefits, and poor communication [42] between ICU teams and general practitioners disrupted continuity of care. Inconsistent follow-up reminders, caused by the de-prioritisation [48] of the program, led to missed appointments and disengagement. (Inner Domain) Individually, many patients faced physical or cognitive impairments [37, 38, 40, 43, 45] that made attendance and self-management [15, 41, 48, 50] difficult, and misconceptions [15, 41, 48, 49] about the benefits of follow-up care contributed to reluctance to participate. (Individual Domain) Finally, poor planning and scheduling conflicts [16], particularly with appointments set too soon after discharge [37], led to potential dropout of participants. High mortality and attrition [42, 50] among care recipients further complicated long-term evaluations of the clinic’s effectiveness. (Implementation Process Domain) (see Fig. 2: CFIR constructs of different domains).

Fig. 2.

CFIR constructs of different domains

Discussion

In this systematic review of studies that described post-ICU clinics as an intervention for ICU survivors, we found that the majority of ICU Recovery programs were established in high-income countries [37, 38, 40, 41]In the last decade [37–40, 43, 48], focusing on patients at a higher risk of developing Post-Intensive Care Syndrome (PICS).

Recent evidence on post-ICU follow-up services remains mixed. While a meta-analysis of 21 studies (16 RCTs) suggested some benefit—particularly short-term improvements in depression and mental health-related quality of life with physical or psychological interventions—another review found insufficient evidence due to limited and heterogeneous data [51, 52]. Notably, a recent French RCT reported that a hospital-based, intensivist-led multidisciplinary follow-up worsened quality of life at 12 months compared to standard community care. Possible reasons included limited symptom-specific expertise, unmet patient expectations, a lack of physiotherapy input, and the burden of repeated, lengthy consultations, which may have been concomitant with possibly superior, less burdensome care experienced by the control group [24]. Similarly, the recently published ‘Recovery beyond Survival’ report highlights systemic gaps and inconsistencies in rehabilitation care for ICU survivors in the UK, calling for structured, multidisciplinary, and patient-centered approaches to ensure comprehensive recovery beyond hospital discharge [52, 53].

Health education by various means [15, 37, 38, 41, 43, 45] to patients and their families is a common component in the clinics and improves adherence to the post-ICU program. Very early scheduling [16] for follow-up may not be feasible considering physical frailty and the psychological issues of both the patients and their caregivers. However, follow-up left too late may delay diagnosis and be a setback [38] for the follow-up program. Patients need to be followed up for a reasonably longer period to monitor progress and detect late-onset PICS. Periodic telephonic consultations [43] between physical interviews may be perceived as confidence-boosting for patients, improving participation. Multidisciplinary approaches [37, 38, 43] with ICU physician leaders were most successful [39–41, 43], set up in tertiary care centres [15, 16, 18, 20] or academic hospitals [37, 39, 43, 45] to overcome the lack of resources and infrastructural barriers.

Attendance in clinics has been a pervasive problem across studies, and hybrid models have been found to perform better than hospital-based physical models in terms of participant engagement. However, participation rates were higher in pure telehealth models that offered personalised care, combining in-person and digital interactions [18, 43, 46, 47], allowing for tailored, individualised care and patient experiences. This approach improved accessibility, countering barriers related to transportation and scheduling [48]. Additionally, Hybrid models offered the advantage of multidimensional evaluation [46, 48] which is a key objective of any post-ICU follow-up program.

The physically based models [15, 37–41]faced challenges such as physical frailty [41], the absence of caregivers [40], accessibility issues [36, 37], including transportation barriers [37, 38], which made it difficult for some patients to attend appointments, and a lack of insurance policy benefits [37, 38].

The lack of reimbursement for post-ICU services remains a significant barrier in many countries. Without financial support, hospitals often cannot justify allocating staff or establishing structured follow-up systems. This results in poorly coordinated care, limited continuity, and reinforces the misconception that post-ICU recovery services are optional or non-essential [54]. A UK-wide survey revealed significant growth in ICU recovery and follow-up services, yet persistent gaps in psychological care, physical rehabilitation, and service equity remain due to funding and staffing constraints [55].

In low- and middle-income countries, where resources are particularly scarce, cost-effective solutions are crucial. ICU “distress thermometers” [56] and digital tools, such as ICU recovery pathways [57] help prioritise care without significantly increasing workload. The use of web-based follow-up platforms and telemedicine also offers promising alternatives, minimising the need for physical clinic space and allowing care to be delivered by fewer personnel over wider regions. Together, these strategies can improve continuity of care while reducing costs and infrastructure demands. In our review, rigid scheduling without flexibility to account for stakeholder engagements and resource constraints, including staffing shortages [16, 49, 50], further limited care availability. While telehealth and completely non-hospital home-based follow-up programs demonstrated the highest participation rates (Chart 4), among the three studies reviewed [16, 49, 50], the comprehensiveness of care provided through telehealth was more limited than that of the hybrid model. High rates of cognitive impairment [16, 49] among patients affected their awareness, understanding, and comprehension of the care delivered through telehealth and also made it challenging to self-monitor [16] their progress. Although it is undeniable that the telehealth or remote screening method is gaining more popularity to overcome geographical [16] and many other barriers, multidimensional telehealth screening tools are yet to be validated [58].

Since the establishment of the THRIVE collaborative for post-ICU clinics in 2017, Haines et al. [54] interrogated clinicians from 23 sites on three continents and used the CFIR, like us, to report the barriers and facilitators for hospital-based clinicians to establish post-ICU clinics. More recently, in 2024, Flores et al. [59] described their personal experience of setting up a post-ICU clinic, and the ISCCM has published a position statement on establishing post-ICU clinics. Key themes in all these are similar to what we found in the review: the concept of “humanising survivorship” served as a facilitator. At the same time “access issues” and “funding” were major barriers to establishing clinics.

Those considering beginning a programme are advised to equip themselves with a motivated clinician leader leading a multidisciplinary team, ensure adequate funds and buy-in with the administration, align their processes with those of institute priority, identify patients who will benefit the maximum and seek advocacy, specially from patient -family groups who share a common experience of survivorship [60, 61].

Within the context of an academic setting in a public hospital of a low- and middle-income country (LMIC), we have triangulated learning from this review with local expertise and patient and carer feedback to codesign a contextually adapted model for post-ICU services. The protocol for evaluation of the feasibility and acceptability of this model is registered prospectively (CTRI number CTRI/2024/08/072677) and is attached as Supplement 3.

Strengths and limitations

The strengths of this systematic literature review include a comprehensive literature search, adherence to PRISMA guidelines, a detailed risk of bias assessment, and the use of CFIR for qualitative data analysis. However, limitations include considering only studies published in English.

Conclusion

This review highlights that hybrid post-ICU clinic models may offer a balanced and adaptable approach to delivering follow-up care, particularly suited to low-resource settings. They facilitate continuity, enhance accessibility, and accommodate patient heterogeneity in recovery trajectories. Successful implementation requires institutional commitment, multidisciplinary coordination, and context-specific adaptations. By addressing structural and logistical barriers and leveraging technology-based solutions, healthcare systems can enhance survivorship care and mitigate the long-term effects of critical illness.

Author contributions

ST conceptualised the study. ST, BR, SN, SC did the literature search SC wrote the main manuscript text. SC, SN, RC conducted data extraction, analysis and prepared figures and tables. ST supervised. All authors reviewed the manuscript.

Funding

This research received no external funding.

Data availability

No datasets were generated or analysed during the current study.

Declarations 

Ethics approval and consent to participate

Ethical approval and consent to participate were not required for this study, as it is a systematic review of previously published data.

Consent for publication

This article does not contain any personal information that could lead to the identification of the patient(s); therefore, consent for publication was not required.

Competing interests

The authors declare no competing interests.