Table 2 Evaluations of non-pharmacological interventions for improving long-term outcomes after critical illness

Pre-Hospital Exercise

 Arthur et al. [32] (Canada, 2000, N = 249)

Hospital LOS [Median (IQR)]: 6 (5–7) vs. 7 (6–8) days; MD (95% CI) 1.0 (0.0 to 1.0), p = 0.002

Mean (SD) change from baseline SF-36 Physical Component Summary Score during the pre-operative phase: 1.6 (7.5) vs. − 1.5 (7.8); MD (95% CI) 3.0 (1.0 to 5.0)

In-ICU Exercise/Physical Rehabilitation

 Wright et al. [43] (UK, 2018, N = 308)

Mean (SD) SF-36 Physical Component Summary Score at 6 months: 37 (12) vs. 37 (11). MD (95% CI) − 1.1 (− 7.1 to 5.0)

 Morris et al. [51] (USA, 2016, N = 300)

Mean (95% CI) SF-36 Physical Functioning scale score at 6 months: 56 (50–62) vs. 44 (38–50). MD (95% CI) 12 (3.8 to 21), p = 0.001

Mean (95% CI) Functional Performance Inventory score at 6 months: 2.2 (2.1–2.4) vs. 2.0 (1.9–2.2). MD (95% CI) 0.2 (0.04 to 0.4), p = 0.02

Mean (95% CI) Short Physical Performance Battery at 6 months: 9.0 (8.3–9.7) vs. 8.0 (7.2–8.7). MD (95% CI) 1.1 (0.04 to 2.1), p = 0.04

 Hodgson et al. [52] (Australia/New Zealand, 2016, N = 50)

Mean (SD) EQ-5D score at 6 months: 0.63 (0.27) vs. 0.63 (0.33), p = 0.25

Mean (SD) IADL score at 6 months: 7 (2) vs. 7 (1), p = 0.81

Mean (SD) HADS score at 6 months: 12 (9) vs. 11 (7), p = 0.91

 Schaller et al. [53] (Austria/Germany/USA, 2016, N = 200)

Median (IQR) mini-modified FIM at hospital discharge: 8 (4–8) vs. 5 (2–8). MD (95% CI) 3.0 (1.0 to 4.0), p = 0.0002

 Kayambu et al. [54] (Australia, 2015, N = 50)

Mean (SD) SF-36 Physical function score at 6 months: 82 (22) vs. 60 (29), p = 0.04

Mean (SD) SF-36 Physical role score at 6 months: 61 (44) vs. 17 (34), p = 0.005

 Moss et al. [55] (USA, 2016, N = 120)

Mean (SD) Continuous Scale Physical Functional Performance Test short form (CS-PFP-10) at 6 months: 40 (4) vs. 44 (4), p = 0.43

 Denehy et al. [56] (Australia, 2013, N = 150)

Mean (SE) 6MWT (meters) at 12 months: 410 (23) vs. 405 (23). MD (95% CI) 4.7 (− 60 to 69), p = 0.88

Mean (SD) TUG (seconds) at 12 months: 10 (6.2) vs. 14 (25). MD (95% CI) − 7.3 (− 19 to 4.4), p = 0.22)

Mean (SD) Assessment of Quality of Life Measure at 12 months: 0.5 (0.4) vs. 0.5 (0.4). MD (95% CI) 0.0 (− 0.1 to 0.2), p = 0.75

Mean (SD) SF-36 Physical function score at 12 months: 41 (13) vs. 44 (11). MD (95% CI) 1.6 (− 3.7 to 7), p = 0.54

Mean (SD) SF-36 Physical Component Summary score at 12 months: 45 (11) vs. 46 (9). MD (95% CI) 0.3 (− 4.3 to 4.8), p = 0.9

Mean (SD) SF-36 Mental Component Summary score at 12 months: 48 (12) vs. 45 (16). MD (95% CI) 5 (− 1.1 to 11.1), p = 0.12

 Chen et al. [22] (Taiwan, 2010, N = 34)

Median (IQR) Total FIM score at 1 year: 78 (62–126) vs. 31 (21–50), p < 0.001

Post-ICU Exercise/Physical Rehabilitation

 Vitacca et al. [23] (Italy, 2016, N = 48)

Mean (95% CI) change in maximal inspiratory pressure (cmH₂O) at 6 months: 14 (5.8–22) vs. − 0.2 (− 7.8 to 7.4), p = 0.007

Mean (95% CI) change in Basic Activities of Daily Living at 6 months: 1 (0–4) vs. 1 (0–4), p = 0.63

Mean (95% CI) change in EQ-5D score at 6 months: 0.23 (− 0.29 to 0.73) vs. 0.032 (− 0.29 to 0.24), p = 0.04

 Brummel et al. [24] (USA, 2013, N = 87)

*3 Groups: Cognitive + Physical Therapy vs. Physical Therapy vs. Usual Care

Median (IQR) Tower test (executive functioning) at 3 months: 10 (8–11) vs. 11 (11–12) vs. 10 (9–12), p = 0.2

Median (IQR) Dysexecutive questionnaire (executive functioning) at 3 months: 9 (2–18) vs. 10 (5–17) vs. 18 (9–29), p = 0.08

Median (IQR) Mini-mental state exam (global cognition) at 3 months: 29 (28–30) vs. 29 (27–30) vs. 28 (27–29), p = 0.64

Median (IQR) TUG (functional mobility) at 3 months: 11 (9–13) vs. 10 (8–13) vs. 8 (8–14), p = 0.79

Median (IQR) Katz ADL at 3 months: 0 (0–2) vs. 0 (0–1) vs. 0 (0–0), p = 0.69

Median (IQR) EQ-5D at 3 months: 75 (54–80) vs. 80 (62–89) vs. 75 (61–86), p = 0.44

 Jones et al. [25] (UK, 2003, N = 126)

SF-36 Physical function scores at 6 months. when controlling for length of ICU stay, were significantly different between group, p = 0.006 (numerical scores not reported)

Number (%) of patients HAD anxiety scale score > 11 at 6 months: 19 (33%) vs. 15 (34%), p = 0.71

 Battle et al. [26] (UK, 2019, N = 60)

Mean (SE) 6MWT at 12 months: 345 (63) vs. 295 (57). MD (95% CI) -50 (-224 to 124), p = 0.37

Mean (SE) HAD-A at 12 months: 4 (1) vs. 9 (1). MD (95% CI) − 4 (1 to 5), p = 0.006

Mean (SE) HAD-D at 12 months: 5 (1) vs. 7 (1). MD (95% CI) − 3 (1 to 3), p = 0.11

 Shelly et al. [27] (India, 2017, N = 35)

Median (IQR) difference in SF-36 Physical component summary at 4 weeks: 10.3 (8.5–14.9) vs. 7.4 (3.7–8.5), p = 0.003

Median (IQR) difference in SF-36 Mental component summary at 4 weeks: 21.8 (15.7–24.1) vs. 14.1 (10.8–19.5), p = 0.006

 McDowell et al. [28] (UK, 2016, N = 60)

Mean (SD) change in SF-36 Role physical score at 6 weeks: 12 (9.8) vs. 5.4 (12). MD (95% CI) 6.6 (0.73 to 12.5), p = 0.03

Mean (SD) change in Incremental Shuttle Walk Test (meters) at 6 weeks: 136 (120) vs. 52 (127). MD (95% CI) 83 (8.3 to 158), p = 0.03

Mean (SD) change in functional limitation profile score at 6 weeks: − 7.8 (7.4) vs. − 3.0 (6.3). MD (95% CI) − 4.8 (− 8.7 to − 0.9), p = 0.02

 McWilliams et al. [29] (UK, 2015, N = 63)

Mean (95% CI) change in SF-36 Physical component summary score at 8–10 weeks: 8.6 (5.4 to 10.6) vs. 3.5 (1.6 to 6.7)

Mean (95% CI) change in SF-36 mental component summary score at 8–10 weeks: 10 (6.9 to 14) vs. 4.3 (0.5 to 7.6)

 Connolly et al. [30] (UK, 2015, N = 20)

Median (IQR) change in Incremental Shuttle Walk Test (meters) at 3 months: 115 (− 3 to 238) vs. 170 (40 to 315)

Median (IQR) change in 6MWT (meters) at 3 months: 140 (36 to 210) vs. 185 (40 to 285)

Median (IQR) change in SF-36 Physical component summary at 3 months: 2 (− 7 to 16) vs. 11 (4 to 28)

Median (IQR) change in SF-36 Mental component summary at 3 months: 14 (− 3 to 27) vs. 11 (− 19 to 19)

Median (IQR) change in HADS at 3 months: − 6 (− 9 to 3) vs. − 5 (− 13 to − 3)

 Batterham et al. [31] (UK, 2014, N = 59)

Mean (SD) anaerobic threshold at week 26: 13 (18) vs. 12 (20). MD (95% CI) of 0.6 (− 1.6–2.8)

Mean (SD) SF-36 Physical function score at week 26: 47 (21) vs. 47 (25). MD (95% CI) of 0.1 (− 6.0 6.2)

Mean (SD) SF-36 Mental health score at week 26: 51 (21) vs. 47 (25). MD (95% CI) 4.4 (− 2.4 to 11.2)

 Jackson et al. [33] (USA, 2012, N = 21

Median (IQR) Tower test at 3 months: 13 (12–14) vs. 7.5 (4.0–8.5), adjusted p < 0.01

Median (IQR) TUG at 3 months: 9 (9–12) vs. 10 (9–12), adjusted p = 0.51

Median (IQR) Functional Activities Questionnaire Score at 3 months: 1.0 (0.0–3.0) vs. 8.0 (6.0–12), adjusted p = 0.04

Median (IQR) Dysexecutive questionnaire (executive functioning) at 3 months: 8 (6–14) vs. 16 (8–19), adjusted p = 0.74

Median (IQR) Mini-mental state exam (global cognition) at 3 months: 30 (29–30) vs. 27 (25–29), adjusted p = 0.25

 Elliott et al. [34] (Australia, 2011, N = 195)

Mean (95% CI) SF-36 Physical function score at 26 weeks: 15 (12–18) vs. 14 (11–16). MD (95% CI) 1 (− 3 to 5), effect size 0.08

Mean (95% CI) 6MWT at 26 weeks: 126 (99–153) vs. 116 (86–147). MD (95% CI) 9.6 (− 31 to 51), effect size 0.08

Mean (95% CI) SF-36 Physical component summary score at 26 weeks: 11 (8–14) vs. 11 (8–13). MD (95% CI) 0 (− 3 to 4), effect size 0.03

Mean (95% CI) SF-36 Mental component summary score at 26 weeks: 10 (6–13) vs. 8 (5–11). MD (95% CI) 2 (− 3 to 6), effect size 0.10

Follow-up services

 Jonasdottir et al. [35] (Iceland, 2018, N = 168)

Mean (SD) HADS-A total score at 12 months: 4 (3) vs. 2.5 (2.8), p = 0.005

Mean (SD) HADS-D total score at 12 months: 4 (3) vs. 4 (4), p = 0.895

Mean (SD) IES-R Score at 12 months: 20 (17) vs. 14 (15), p = 0.066

 Jensen et al. [36] (Denmark, 2016, N = 386)

MD (95% CI) SF-36 Physical component summary score at 12 months: 1.4 (− 1.5 to 4.4), p = 0.35

MD (95% CI) SF-36 Mental component summary score at 12 months: 1.9 (− 1.1 to 4.9), p = 0.21

MD (95% CI) HADS-A at 12 months: − 0.21 (− 1.22 to 0.80), p = 0.68

MD (95% CI) HADS-D at 12 months: − 0.20 (− 1.12 to 0.72), p = 0.67

MD (95% CI) HTQ-IV score (PTSD severity) at 12 months: − 1.42 (− 3.94 to 1.11), p = 0.27

 Schmidt et al. [37] (Germany, 2016, N = 143)

MD (SD) difference in SF-36 Physical component summary score at 12 months: 10 (12) vs. 8 (14). MD (95% CI) 1.1 (− 2.7 to 4.9), p = 0.56

MD (SD) difference in SF-36 Mental component summary score at 12 months: 4 (13) vs. 2 (13). MD (95% CI) 1.4 (− 2.4 to 5.2), p = 0.47

 Cuthbertso et al. [38] (UK, 2009, N = 286)

Mean (SD) SF-36 Physical component summary score at 12 months: 42 (11) vs. 41 (12). MD (95% CI) 1.1 (− 1.9 to 4.2), p = 0.46

Mean (SD) SF-36 Mental component summary score at 12 months: 47 (13) vs. 47 (12). MD (95% CI) 0.4 (− 3.0 to 3.7), p = 0.83

Mean (SD) EQ-5D Utility score at 12 months: 0.58 (0.37) vs. 0.60 (0.30). MD (95% CI) 0 (− 0.1 to 0.1), p = 0.57

Mean (SD) HADS-A score at 12 months: 6 (5) vs. 6 (4). MD (95% CI) − 0.8 (− 1.9 to 0.4), p = 0.57

 Douglas et al. [39] (USA, 2007, N = 335)

No difference in SF-8 physical scores at 2 months, p = 0.40, controlling for baseline scores and APACHE III

No difference in SF-8 mental scores at 2 months, p = 0.22, controlling for baseline scores and APACHE III

Psychosocial Programs

 Cox et al. [40] (USA, 2018, N = 80)

Mean (95% CI) change from baseline for PHQ-9 at 3 months: Telephone group (TG) − 3.9 (− 5.6 to − 2.2), Mobile group (MG) − 4.8 (− 6.6 to − 2.9), Education group (EG) − 3.0 (− 5.3 to 0.8). MD (95% CI) TG vs. EG − 0.9 (− 3.7 to 2.0), p = 0.41. MD (95% CI) MG vs. EG − 1.7 (− 4.7 to 1.2), p = 0.25

Mean (95% CI) change from baseline for GAD-7 at 3 months: TG − 1.6 (− 3.0 to − 0.1), MG − 2.1 (− 3.7 to − 0.5), EG − 0.6 (− 2.5 to 1.3). MD (95% CI) TG vs. EG − 1.0 (− 3.3 to 1.4), p = 0.43. MD (95% CI) MG vs. EG − 1.5 (− 3.9 to 1.0), p = 0.24

Mean (95% CI) change from baseline for PTSD at 3 months: TG − 2.2 (− 5.6 to 1.2), MG − 2.6 (− 6.3 to 1.2), EG − 3.5 (− 8.0 to 1.0). MD (95% CI) TG vs. EG 1.3 (− 4.4 to 7.0), p = 0.65. MD (95% CI) MG vs. EG − 0.9 (− 4.9 to 6.8), p = 0.75

Mean (95%) change from baseline for PHQ-15 at 3 months: TG − 3.7 (− 5.2 to − 2.2), MG − 5.3 (− 7.0 to − 3.7), EG − 4.8 (− 6.8 to 2.7). MD (95% CI) TG vs. EG 1.1 (− 1.5 to 3.6), p = 0.41. MD (95% CI) MG vs. EG − 0.6 (− 3.2 to 2.0), p = 0.52

 Cox et al. [41] (USA, 2017, N = 175)

Mean (SE) HADS at 6 months: 16 (1) vs. 16 (1). MD (95% CI) − 0.3 (− 2.7 to 2.0), p = 0.78

Mean (SE) IES-R at 6 months: 29 (3) vs. 26 (3). MD (95% CI) 3.6 (− 2.7 to 10.0), p = 0.26

Mean (SE) EQ-5D at 6 months: 61 (3) vs. 61 (3). MD (95% CI) 0.3 (− 5.9 to 6.6), p = 0.92

 Agren et al. [42] (Sweden, 2014, N = 84)

No difference in SF-36 between groups at 12 months (numeric data and p-value not reported)

Diaries

 Garrouste et al. [44] (France, 2012, N = 216)

Mean (SD) of IES-R data score at 12 months pre-diary 35 (16), diary 21 (12), post diary 30 (15), p = 0.03

 Jones et al. [45] (6 European countries, 2010, N = 352)

Incidence of PTSD 5% vs 13%, p = 0.02

 Knowles et al. [8] (UK, 2009, N = 36)

Number (%) of patients with HADS-anxiety score ≥ 8 at ~ 7 weeks: 2 (11) vs. 7 (39), p < 0.05

Number (%) of patients with HADS-depression score ≥ 8 at ~ 7 weeks: 3 (17) vs. 8 (44), p < 0.07

Information & Education Programs

 Demircelik et al. [46] (Turkey, 2016, N = 100)

Mean (SD) in change in HADS (anxiety) from ICU to 1 week discharge: 4.2 (0.58) vs. 0.6 (0.42), p < 0.01

Mean (SD) in change in HADS (depression) from ICU to 1 week discharge: 3.5 (0.53) vs. 0.3 (0.46), p < 0.01

 Fleisher et al. [47] (Germany, 2014, N = 211)

Mean (SD) SF-12 Physical component summary score at 3 months: 41 (9) vs. 40 (10). MD (95% CI) 0.3 (− 3.1 to 3.6), p = 0.87

Mean (SD) SF-12 Mental component summary score at 3 months: 47 (11) vs. 48 (11). MD (95% CI) − 1.3 (− 5.3 to 2.6), p = 0.5

ABCDE Bundle

 Sosnowski et al. [48] (Australia, 2018, N = 30)

Mean (SD) SF-36 Physical component summary score at 90 days: 44 (12) vs. 38 (11)

Mean (SD) SF-36 Mental component summary score at 90 days: 47 (16) vs. 40 (16)

Earplugs and eye mask during ICU

 Demoule et al. [49] (France, 2017, N = 64)

Median (IQR) HADS-A at 90 days: 8 (4–11) vs. 6 (4–12), p = 0.69

Median (IQR) HADS-D at 90 days: 6 (3–12) vs. 6 (2–9), p = 0.63

Median IES-R at 90 days: 11 (5–18) vs. 16 (9–27), p = 0.15

Structured mirrors during ICU

 Giraud et al. [50] (UK, 2016, N = 223)

Mean (SD) EQ-5D at 12 weeks: 73 (19) vs. 77 (15), p = 0.13