Editorial 

The Future of Survivorship in Critical Illness

Kevin Moncion, Jenna Smith-Turchyn, Michelle E. Kho

JAMA Published Online: May 18, 2026

doi: 10.1001/jama.2026.5761

In this issue of JAMA, O’Neill and colleagues¹ report the results of a pragmatic, multicenter, randomized clinical trial evaluating a remotely delivered, 6-week, multicomponent, rehabilitation intervention for critical illness survivors. Across 52 National Health Service hospitals in the UK, 429 adults who received at least 48 hours of invasive mechanical ventilation, could speak English or had a translator, and were directly discharged home from the hospital were randomized to a rehabilitation intervention (n = 231) or standard care (n = 198).¹ Specially trained interventionists led weekly 1-on-1 remote appointments encompassing exercise, symptom management, and psychological well-being support. The interventionists had regular meetings (1-2 per month) with a multiprofessional team to address participants’ queries and identify further management strategies. Additional program offerings included weekly group-based remote exercise sessions and an online peer support group.

The primary outcome, the EuroQoL 5-dimension 5-level questionnaire (EQ-5D-5L) utility score, was centrally measured by trial staff blinded to treatment allocation at 8 weeks.¹ Investigators calculated the sample size to detect a minimum clinically important difference of 0.08 between groups. Even though participants randomized to the rehabilitation intervention had higher EQ-5D-5L scores, the trial did not meet the prespecified clinical or statistical significance thresholds at 8 weeks (adjusted mean difference, 0.04 [95% CI, −0.001 to 0.09]; P = .05).¹The complier average causal effect analyses revealed a statistically significant difference for participants with higher adherence to the 1-on-1 sessions (88% of participants fully adherent, ≥80% fidelity; adjusted mean difference, 0.05 [95% CI, 0.005 to 0.10]; P = .03).¹ A priori subgroup analyses revealed better primary outcome scores in participants 56 years of age or younger (n = 158; 85 in the rehabilitation intervention group and 73 in the standard care group) and in those who required mechanical ventilation for 7 days or less in the intensive care unit (ICU) (n = 175; 89 and 86, respectively).¹

High-quality survival after ICU discharge is important to patients, their caregivers, researchers, and society. High-quality survival goes beyond simply being alive after time spent in the ICU and reflects a high level of recovery in function and participation after an individual experiences critical illness.² The iRehab trial¹ addresses an important clinical issue and has several strengths, including being a multicenter study with sites across the UK, having a thoughtful and rigorous study design, having a centralized intervention and outcome assessment, and having blinded outcome assessments. However, of the 1672 individuals offered participation in the study, 1109 (66%) declined the offer.¹ The intervention delivery in the iRehab trial¹ is novel—a remotely delivered, holistic intervention addressing critical illness survivorship concerns. Its premise is potentially scalable and customizable across health care systems in high- and middle-income countries. We highlight 3 areas of importance when considering the results of the iRehab trial¹ and rehabilitation from ICU stays: the promise of remote rehabilitation, rehabilitation as a complex intervention, and the physical recovery trajectory after an ICU stay.

The Promise of Remote Rehabilitation

Remote rehabilitation can be provided in various forms, including video conferencing, mobile apps, virtual reality, and other information and communication technologies. The benefits of remote rehabilitation include improved accessibility for individuals with geographic or mobility barriers, reduced travel burden, flexible therapy delivery, and the ability to extend services into the home environment.^3-5 The challenges include technology connectivity and quality issues, complexity for users with low health and tech literacy, limited ability to monitor safety, limitations in clinician preparation for providing remote care, inability to provide hands-on assessment and therapy, and concerns about data privacy and security.^3-5 The intervention in the iRehab trial¹ overcame the challenges of remote rehabilitation by giving participants the option of connecting using an online platform or by telephone, providing them with material to supplement the remote intervention, incorporating substantial training and fidelity checks for the specially trained interventionists, and using a multicomponent, patient-centered intervention to address individual participant needs and improve self-efficacy and self-management skills.

Remote rehabilitation approaches have also been implemented across other clinical populations (such as cancer and stroke) where they have improved accessibility, supported functional recovery, and addressed barriers to transitional care delivery.^6-8 Similar to the intervention in the iRehab trial,¹ these approaches typically include symptom monitoring, education, and exercise components and are delivered over multiple weeks (most often 4-12 weeks). The multicomponent intervention in the iRehab trial¹ included additional support for psychosocial well-being. As described above, baseline function among participants in the iRehab trial¹ was low, whereas other studies typically included participants with higher functional status. The iRehab trial¹ brings awareness to the opportunity of using remote rehabilitation across varying physical function levels, with careful safety monitoring.

Rehabilitation as a Complex Intervention

The iRehab trial¹ was a complex intervention, codesigned with patients with lived experience of critical illness, grounded in program theory, and built to deliver patient-centered, individually tailored care.⁹ Consistent with the UK Medical Research Council’s framework for complex interventions,^10,11 the intervention in the iRehab trial¹involved multiple interacting components delivered within a complex setting. Successful implementation across 52 sites required real-time adaptation and clinical reasoning from interventionists, self-management and adherence from participants, and targeted outcome collection across physical, psychological, and social domains. Indeed, this complexity has direct implications for the interpretation of trial results.

The Medical Research Council’s framework cautions that a single primary outcome may not adequately capture the effects of a complex intervention.^10,11 Although the EQ-5D-5L is recommended in core outcome sets for critical illness,¹² O’Neill et al¹ acknowledge it is a generic outcome measure that may not be responsive to a complex intervention that simultaneously targets multiple health domains.¹³ However, when the findings are considered as a pattern across outcomes, rather than through the EQ-5D-5L end point alone, the overall interpretation becomes more favorable. Across secondary outcomes, which serve as hypothesis-generating, the participants randomized to the intervention in the iRehab trial¹ had statistically higher 30-second sit-to-stand scores, less fatigue, and less anxiety. At 6 months, the difference in the EQ-5D-5L scores was statistically and clinically significant, favoring the intervention group (adjusted mean difference, 0.08 [95% CI, 0.03 to 0.13]; P = .003),¹ although the study was underpowered for this time point. Across outcomes, this consistency suggests that the intervention achieved its intended therapeutic targets.

However, among ICU survivors offered the trial, only one-third agreed to participate. This is lower than the recruitment rate (median, 82.6%; IQR, 60%-100%) reported in a systematic review¹⁴ of follow-up care after ICU stay for interventions without hospital attendance. A complementary analysis of qualitative studies documenting barriers and facilitators to follow-up after ICU stay identified the importance of offering patients flexibility in choosing when to attend follow-up.¹⁴ In the context of a clinical trial, flexibility may be impractical. Furthermore, patients declining follow-up had more comorbid disease, and the authors¹⁴ hypothesized that these patients already had established health care contacts or had lower expectations for full health recovery. Important considerations for future research include tailoring the initiation of rehabilitation after an ICU stay and characterizing individuals who decline to participate in research.

The Physical Recovery Trajectory After an ICU Stay

The iRehab trial¹ offers an important window into the physical impairment and trajectory of physical function recovery in ICU survivors. The iRehab trial¹ included a performance-based measure, the 30-second sit-to-stand test,¹⁵ a measure of physical function with age- and sex-based norms. Notably, the investigators in the iRehab trial¹ conducted these assessments virtually. The mean age of this cohort was 55 years. Prior to the critical illness event, more than 95% of the participants¹ could walk independently or with a mobility aid. After a mean of 11 days of mechanical ventilation, 19 days in the ICU, and 36 days in the hospital, participants had 30-second sit-to-stand test scores that were approximately 8 repetitions in the iRehab trial,¹ which are well below age- and sex-based norms (24 repetitions in females and 25 in males),¹⁶ and lower than norms for community-dwelling people aged 80 to 90 years (14 repetitions in females and 17 in males).¹⁶

By 8 weeks, the participant sit-to-stand test scores had improved to approximately 10 repetitions and, by 6 months, to approximately 12 repetitions.¹ However, these scores still fall below the functional threshold for community-based independence (sit-to-stand test score of 15 repetitions in females and 17 in males).¹⁷ Impaired physical function is associated with worse mobility, an increased risk of falls, and an increased risk of future disability.^18,19 These results demonstrate that ICU survivors experience important physical impairment, and some survivors need ongoing rehabilitation and community-based exercise programs beyond 6 months after hospital discharge. Wherever possible, the use of performance-based outcome measures as part of longitudinal follow-up in any cohort of ICU survivors is critical to understanding recovery and identifying future intervention targets.

Conclusions

The intervention in the iRehab trial¹ provides a model for future research to build on and may offer one option for ICU survivors to receive structured recovery support at home without the barriers of travel, fatigue, or limited local resources. Further work is needed to establish long-term efficacy, determine cost-effectiveness, and optimize usability for diverse patient needs. There is a need to explore why many individuals declined to participate in the iRehab trial and how to engage ICU survivors in future rehabilitation studies. Last, to improve recovery for ICU survivors, organizational and funding structures are needed to support the sustained integration of various rehabilitation modalities and community-based exercise programs into care pathways after ICU stays.