Publication

• Title: Early active mobilization during mechanical ventilation in the ICU
• Acronym: TEAM
• Year: 2022
• Journal published in: New England Journal of Medicine
• Citation: TEAM Study Investigators and ANZICS Clinical Trials Group. Early active mobilization during mechanical ventilation in the ICU. N Engl J Med. 2022;387(19):1747-1758.

Context & Rationale

• Background

  Critical illness and prolonged invasive mechanical ventilation are associated with rapid muscle wasting, ICU-acquired weakness, and long-term functional disability (a core domain of post–intensive care syndrome). Early physiotherapy and mobilisation had become widely promoted and implemented in some ICUs, but prior RCTs were generally smaller, heterogeneous in “dose” and timing, and often conducted against evolving background standards (lighter sedation, delirium prevention, earlier weaning), making incremental benefit uncertain. Mobilising ventilated patients can also provoke haemodynamic/respiratory instability and device-related events, so a definitive, pragmatic safety-and-effectiveness trial was needed.
• Research Question/Hypothesis

  In adults invasively ventilated in ICU and considered stable enough for mobilisation, does an enhanced, physiotherapist-led programme of early active mobilisation delivered during mechanical ventilation increase days alive and out of hospital at 180 days compared with usual care rehabilitation?
• Why This Matters

  Early mobilisation is resource-intensive, requires multidisciplinary coordination, and can expose patients to harms. If benefit is marginal when “usual care” already includes early mobilisation, widespread implementation of more intensive mobilisation during ventilation may not be justified; conversely, demonstrable benefit would support staffing and pathway redesign.

Design & Methods

• Research Question: Among invasively ventilated ICU adults, does early active mobilisation delivered during mechanical ventilation improve days alive and out of hospital at 180 days versus usual care?
• Study Type: International, pragmatic, multicentre, randomised, parallel-group, open-label trial conducted in 49 ICUs across 6 countries (ICU setting).
• Population:
  • Core setting: Adult ICU patients receiving invasive mechanical ventilation, enrolled early after eligibility confirmation.
  • Key inclusion (protocol-specified): Age ≥18 years; intubated and expected to remain on invasive mechanical ventilation for >48 hours; cardiovascular stability compatible with mobilisation (e.g., ventricular rate ≤150 beats/min; lactate ≤4.0 mmol/L; noradrenaline/adrenaline limits as per protocol); respiratory stability compatible with mobilisation (FiO₂ ≤0.6; PEEP ≤16 cmH₂O; no current requirement for prone ventilation, neuromuscular blockers, venovenous ECMO, nitric oxide, prostacyclin, or high-frequency oscillation ventilation). ¹
  • Key exclusions (protocol-specified): Dependent for activities of daily living in the month before ICU admission; documented cognitive impairment; proven/suspected acute primary brain injury or spinal cord injury/neuromuscular disease likely to cause prolonged weakness; rest-in-bed/non-weight bearing orders; limited life expectancy (<180 days) from chronic disease or death deemed inevitable with no commitment to full active treatment; inability to communicate in the official local language; not first ICU admission during the index hospitalisation; >72 hours since inclusion criteria first fulfilled. ¹
• Intervention:
  • Enhanced early active mobilisation: Physiotherapist-directed functional rehabilitation (active exercises and mobilisation) delivered as early as possible while patients were invasively ventilated, aiming for the highest feasible ICU Mobility Scale level each day; delivered daily (in one or more sessions) for the ICU stay up to day 28.
  • Delivered “dose” achieved (trial separation): Mean daily duration of active mobilisation 20.8 ± 14.6 minutes in the intervention group versus 8.8 ± 9.0 minutes in usual care (mean difference 12.0 minutes/day; 95% CI 10.4 to 13.6).
• Comparison:
  • Usual care: Rehabilitation/mobilisation at the intensity and timing normally provided in each participating ICU; physiotherapy not protocolised by the trial.
• Blinding: Participant and clinician blinding was not feasible (mobilisation is overt); trained research staff unaware of treatment allocation ascertained patient-reported outcomes at 180 days, and statistical analyses were performed blinded to allocation.
• Statistics: A total sample size of 750 patients was planned to detect a 7-day difference in days alive and out of hospital at day 180 with 90% power at a two-sided 5% significance level (with inflation to account for non-normality and loss to follow-up); primary analysis was intention-to-treat (excluding those withdrawing consent for use of data), using median (quantile) regression with site-level clustering; multiple imputation was used for outcomes with missing data. ¹
• Follow-Up Period: 180 days after randomisation (key in-hospital secondary outcomes assessed at day 28).

Key Results

This trial was not stopped early. A planned interim analysis occurred after 400 participants had completed day-28 follow-up; recruitment continued to the target sample size.

• Effectiveness: Early active mobilisation did not increase days alive and out of hospital at 180 days (median difference −2.0 days; 95% CI −10 to 6; P=0.62), and did not improve key in-hospital or 180-day functional outcomes.
• Intervention delivery: The intervention increased active mobilisation time by ~12 minutes/day and accelerated achievement of some mobility milestones, but the proportion achieving standing was similar in both groups.
• Safety signal: Adverse events potentially attributable to mobilisation were more frequent with early active mobilisation (OR 2.55; 95% CI 1.33 to 4.89; P=0.005), including more arrhythmias and oxygen desaturation events.

Internal Validity

• Randomisation and allocation concealment: Centralised, web-based randomisation stratified by site with permuted blocks of varying size (concealed until randomisation); open-label thereafter. ¹
• Post-randomisation exclusions and missingness: 750 participants were randomised; primary outcome was available for 733 participants (369 early mobilisation; 364 usual care).
• Differential loss: Participants without a primary outcome were more frequent in usual care (14) than early mobilisation (3), mainly from withdrawal of consent and loss to follow-up; missingness was low overall, and multiple imputation was used for incomplete outcomes.
• Performance/detection bias: Mobilisation could not be blinded; however, the primary outcome (days alive and out of hospital) is largely objective, and patient-reported outcomes were collected by trained staff unaware of allocation.
• Protocol adherence and separation of the variable of interest: Mean active mobilisation time differed materially (20.8 ± 14.6 vs 8.8 ± 9.0 minutes/day; mean difference 12.0 minutes/day; 95% CI 10.4 to 13.6).
• Contamination / “high-quality” usual care: Standing (ICU Mobility Scale ≥4) occurred in 77.4% of early mobilisation patients and 77.3% of usual care patients; peak ICU Mobility Scale score was identical (median 6 [IQR 4–8] in both groups; difference 0; 95% CI −1 to 1), consistent with substantial mobilisation occurring in the control arm.
• Baseline comparability: Groups were well balanced at baseline (e.g., mean age 60.5 vs 59.5 years; mean APACHE II 18.2 vs 18.6; sepsis 66.3% vs 66.2%; vasopressor use 61.5% vs 62.4%).
• Timing and feasibility constraints: Enrolment required cardio-respiratory stability (FiO₂ ≤0.6; PEEP ≤16 cmH₂O; no prone ventilation/neuromuscular blockers/ECMO), which reduces non-adherence from “too sick to mobilise”, but narrows the eligible population and may reduce detectable incremental benefit.
• Statistical rigour: Sample size target was achieved; primary analysis used median regression with clustering by site; secondary outcomes were not multiplicity-adjusted and are best interpreted as supportive/ exploratory.

Conclusion on Internal Validity: Overall, internal validity appears moderate-to-strong: randomisation and follow-up were robust and outcomes were largely objective, but open-label delivery and substantial mobilisation within usual care reduced treatment separation and plausibly diluted any true effect.

External Validity

• Population representativeness: Participants were typical of many medical–surgical ventilated ICU cohorts (mean APACHE II ~18; ~66% sepsis), but important groups were excluded (baseline dependency/cognitive impairment; acute primary neurological injury; and patients requiring prone ventilation, neuromuscular blockade, or ECMO).
• Setting and resources: Conducted in 49 ICUs across 6 countries; recruitment was predominantly from Australia/New Zealand (with smaller contributions from the United Kingdom, Germany, Ireland, and Brazil), in systems where physiotherapy-led mobilisation programmes may be relatively well resourced.
• Intervention feasibility: The intervention required daily, structured physiotherapy mobilisation during ventilation and institutional capability to mobilise ventilated patients safely; this may not translate directly to low-resource ICUs or settings without routine physiotherapy coverage.

Conclusion on External Validity: Generalisability is good to well-resourced ICUs caring for stable mechanically ventilated adults, but limited for patients with severe ARDS requiring prone ventilation/neuromuscular blockade/ECMO, those with major premorbid disability, and units with constrained rehabilitation staffing.

Strengths & Limitations

• Strengths:
  • Large, pragmatic, multinational ICU RCT (n=750) with clinically meaningful, patient-centred 180-day outcome.
  • Delivered a measurable increase in mobilisation “dose” and earlier achievement of some mobility milestones.
  • Transparent safety reporting with quantification of mobilisation-attributable adverse events and serious events.
  • Objective primary outcome, with blinded ascertainment of patient-reported outcomes and blinded statistical analysis.
• Limitations:
  • Open-label intervention with potential for performance bias and co-intervention differences.
  • High mobilisation in the control arm (standing achieved in ~77% in both groups; identical peak mobility), limiting separation and increasing risk of “ceiling effects”.
  • Eligibility required cardio-respiratory stability and excluded patients receiving prone ventilation/neuromuscular blockers/ECMO, limiting applicability to the sickest ventilated patients.
  • The primary endpoint (days alive and out of hospital) is influenced by health-system discharge practices and post-acute care pathways, although it captures survival and healthcare utilisation.
  • Secondary outcomes were not multiplicity-adjusted; interpretation should focus on effect sizes and confidence intervals rather than nominal significance testing.

Interpretation & Why It Matters

• Clinical practice

  In ICUs where usual care already includes frequent early mobilisation (including standing/walking while ventilated), augmenting the “dose” by an additional ~12 minutes/day did not translate into better 180-day recovery or reduced mortality. This supports a pragmatic expectation: “more mobilisation during ventilation” is not, by itself, a reliable lever for improving long-term outcomes.
• Safety and implementation

  Mobilisation-attributable adverse events were more frequent with early active mobilisation, particularly arrhythmias and oxygen desaturation; implementation should emphasise safety screening, staffing skill-mix, and real-time physiological monitoring.
• Research direction

  The null result in the presence of high-quality usual care implies that future benefit, if present, may depend on patient selection (who is most likely to respond), intervention phenotype (what type of activity), and dose/intensity (how much, how early, for how long), rather than simple “early vs not early” mobilisation.

Controversies & Subsequent Evidence

• Interpretation of a neutral, pragmatic trial: A contemporaneous editorial emphasised that TEAM likely tested an incremental increase in mobilisation against an already proactive mobilisation culture, and highlighted the need to refine patient selection and intervention components rather than assuming “early mobilisation” is a uniformly effective, dose-agnostic treatment. ²
• Safety signal and bedside feasibility: Correspondence focused on mobilisation-attributable adverse events and the clinical interpretation of increased events without patient-centred benefit. ³
• Author response: The trialists’ reply underscored protocolised screening and event adjudication, and framed the findings as evidence that “enhanced” mobilisation during ventilation should not be expected to improve long-term outcomes when usual care is already mobilisation-forward. ⁴
• Conceptual reframing: A subsequent intensive care commentary argued that “mobilise vs not” may be the wrong binary, and that rehabilitation effects likely depend on modifiable barriers (sedation/delirium, catheters, ventilation mode) and individualised goals across the ICU-to-ward-to-home continuum. ⁵
• Guidelines incorporating contemporary evidence: An international guideline on positioning and early mobilisation in critically ill adults continues to recommend structured mobilisation approaches, but acknowledges that evidence for long-term outcome improvement is inconsistent and implementation must be safety-led and context-specific. ⁶
• Critical care society guidance: The 2025 Society of Critical Care Medicine focused update (PADIS domain: immobility) reinforces early mobilisation as part of comprehensive ICU care, while reflecting uncertainty about hard clinical endpoints and emphasising standardised pathways and interdisciplinary delivery. ⁷
• Observational signal vs causal effect: A recent prospective observational analysis of mobilisation variables in ventilated patients reported associations between mobilisation characteristics, adverse events, and outcomes, supporting the importance of safety monitoring while underscoring confounding risks when inferring causality outside randomised comparisons. ⁸
• Health economics: A published cost-effectiveness evaluation based on TEAM examined costs and outcomes (including quality-adjusted life years) to inform implementation decisions in settings where physiotherapy resources are constrained. ⁹

Summary

• In 750 invasively ventilated ICU adults, enhanced early active mobilisation during ventilation did not improve days alive and out of hospital at 180 days (median difference −2.0 days; 95% CI −10 to 6; P=0.62).
• There was no clear benefit for mortality, ventilator-free days, ICU-free days, or patient-reported functional outcomes at 180 days.
• The intervention increased active mobilisation time by ~12 minutes/day and accelerated some mobility milestones, but standing rates and peak mobility were similar in both groups.
• Mobilisation-attributable adverse events were more frequent with early active mobilisation (OR 2.55; 95% CI 1.33 to 4.89; P=0.005), including more arrhythmias and oxygen desaturation.
• The findings suggest that, where “usual care” mobilisation is already high, incremental increases in mobilisation during ventilation are unlikely to deliver additional long-term benefit and may increase harms.

Overall Takeaway

TEAM is a landmark pragmatic rehabilitation trial because it tested a clinically deliverable increase in physiotherapist-led mobilisation during mechanical ventilation at international scale and with 180-day, patient-centred recovery endpoints. Its neutral effectiveness result—paired with a clear safety signal and high mobilisation in usual care—shifts modern practice from assuming “more mobilisation during ventilation is better” toward safer, context-aware, and potentially individualised rehabilitation strategies.

Bibliography

• 1Presneill JJ, Bellomo R, Brickell K, Buhr H, Gabbe BJ, Gould DW, et al. Protocol and statistical analysis plan for the Treatment of Invasively Ventilated Adults with Early Activity and Mobilisation (TEAM) trial. Crit Care Resusc. 2021;23(3):262-272.
• 2Moss M. Early Mobilization of Critical Care Patients—Still More to Learn. N Engl J Med. 2022;387(19):1807-1808.
• 3Cheung JCH, Ho L, Yip YY. Early Active Mobilization during Mechanical Ventilation in the ICU. N Engl J Med. 2023;388(6):572-573.
• 4Hodgson CL, Bailey MJ, Young PJ. Early Active Mobilization during Mechanical Ventilation in the ICU. Reply. N Engl J Med. 2023;388(6):573-574.
• 5Hodgson CL, Kho ME, da Silva VM. To mobilise or not to mobilise: is that the right question? Intensive Care Med. 2023;49(8):1000-1004.
• 6Schaller SJ, Anstey M, Blobner M, et al. Guideline on positioning and early mobilisation in the critically ill adult in the ICU. Intensive Care Med. 2024;50(8):1211-1227.
• 7Lewis K, et al. A focused update to the clinical practice guidelines for the prevention and management of pain, agitation/sedation, delirium, immobility, and sleep disruption in adult patients in the ICU: Society of Critical Care Medicine guidelines. Crit Care Med. 2025;53(3):e711-e727.
• 8Paton M, et al. Association of active mobilisation variables with adverse events and mortality in mechanically ventilated patients: a prospective observational study. Lancet Respir Med. 2024;12(5):403-415.
• 9Higgins AM, Summers MT, Coulter EC, Needham DM, Hodgson CL, et al. A cost-effectiveness analysis of early active mobilisation in adults who are mechanically ventilated in an ICU: the TEAM trial. Crit Care Med. 2025;53(9):e1725-e1735.