To analyze the effect of early physiotherapy (ePHYS) on patients' functional quality of life one year after discharge.
DesignProspective observational study.
SettingsAdult polyvalent ICU.
Patients or participantsPatients with SARS-CoV2 pneumonia requiring invasive mechanical ventilation between March 2020 and July 2022.
InterventionEarly physiotherapy.
Main variables of interestQuality of life measured by CFS, Barthel and SF36 on ICU admission, at hospital discharge and one year after hospital discharge.
ResultsNinety-nine patients included. In the raw data analysis we observed statistically significant differences in SAPS-3 (MdnNo-ePHYS = 59 [53.5–64.5]; MdnYes-ePHYS = 53 [47–58]; P = .001). After propensity score, we did not observe statistically significant differences except for two SF-36 items: social activities (MdnNo-ePHYS = 56.2 [37.5−71.9]; MdnYes-ePHYS = 75 [62.5−97.5]; P = .004; Wilcoxon r effect size = 0.5) and one-year health transition (MdnNo-ePHYS = 50 [50−75]; MdnYes-ePHYS = 75 [50−100]; P = .031; Wilcoxon r effect size = 0.29), where patients who belonged to NO-ePHYS had lower scores than those who did receive ePHYS.
ConclusionsDespite not having found statistically significant differences in most of the items assessed, we should highlight that patients who received ePHYS reported a clear positive influence on their lives.
Analizar el efecto de la fisioterapia precoz (FTSPe) sobre la calidad de vida funcional de los pacientes un año después del alta.
DiseñoEstudio observacional prospectivo.
ÁmbitoUCI polivalente de adultos.
Pacientes o participantesPacientes con neumonía por SARS-CoV2 que requirieron ventilación mecánica invasiva entre marzo de 2020 y julio de 2022.
IntervencionesFisioterapia precoz.
Variables de interés principalesCalidad de vida medida mediante CFS, Barthel y SF36 al ingreso en UCI, al alta hospitalaria y un año después del alta hospitalaria.
ResultadosNinety-nine pacientes incluidos. En el análisis de datos brutos observamos diferencias estadísticamente significativas en SAPS-3 (MdnNo-FTSPe = 59 [53,5–64,5]; MdnSí-FTSPe = 53 [47–58]; P = ,001). Tras la puntuación de propensión, no observamos diferencias estadísticamente significativas excepto en dos ítems del SF-36: actividades sociales (MdnNo-FTSPe = 56,2 [37,5–71,9]; MdnYes-FTSPe = 75 [62,5–97,5]; P = ,004; tamaño del efecto de Wilcoxon r = 0.5) y la transición de salud a un año (MdnNo-FTSPe = 50 [50−75]; MdnSí-FTSPe = 75 [50−100]; P = ,031; tamaño del efecto r de Wilcoxon = 0,29), donde los pacientes que pertenecían a NO-FTSPe tenían puntuaciones más bajas que los que sí recibieron FTSPe.
ConclusionesA pesar de no haber encontrado diferencias estadísticamente significativas en la mayoría de los ítems evaluados, cabe destacar que los pacientes que recibieron ePHYS refirieron una clara influencia positiva en sus vidas.
Despite decreasing ICU and hospital mortality, ICU survivors have a long road to recovery.1,2 A patient who manages to overcome the adversities of critical illness may suffer long-term physical, psychological, and cognitive damage, leading to a decreased quality of life, sometimes without ever fully recovering.1,3
Besides, it is now known that the combination of bed rest, muscle catabolism, and drug side effects accelerate neuromuscular and functional deterioration, referred to as ICU-acquired weakness (ICU-AW).4–7 As the prevalence of ICU-AW has been shown to affect between 25% and 80% of admitted patients, reducing its impact becomes a priority.8 Although difficult to define in clinical practice, early mobilization is accepted as the “intensification and early application of physiotherapy administered to the critically ill patient”.9 Observational studies and randomized controlled trials (RCT) have shown it to be a safe and feasible strategy in the critical care setting10 that can reduce the duration of mechanical ventilation (MV) and improve functional outcomes at hospital discharge.11–13 Two systematic reviews have observed no differences in mortality but lower ICU length of stay, slight improvements in physical function at hospital discharge, and no adverse safety events.14,15
The rationale for mobilization is credible: it has been shown to significantly improve muscle strength and functional capacity, which may counteract ICU-AW development. However, prescribing appropriate mobilization programmes for critically ill patients can be complex. This is partly due to the heterogeneity of the cohort of critically ill patients and is further compounded by a lack of consistency in their application. Not only is there a wide range of exercise types (including passive movement, electrical muscle stimulation, bicycle ergometers, and functional mobilization), but the effect of the “given dose” (defined by the time and intensity of the administered physiotherapy) is not yet apparent.16
We aim to investigate the influence of early physiotherapy (ePHYS) during critical illness on ICU survivors' perceived quality of life one year after discharge. We hypothesize that applying an ePHYS protocol in critically ill patients would improve their health status after critical illness. The ePHYS protocol, carried out jointly with the Rehabilitation and Physiotherapy team, was developed and implemented at the beginning of the COVID-19 pandemic, given the inherent need to initiate such therapy in ICU patients.17
MethodsWe conducted an observational study including all patients admitted to the ICU who required mechanical ventilation for more than 24 h and survived SARS-COV-2 respiratory infection between the 1st of March 2020 and the 1st of February 2022. The period defined between March and April 2020 involves the set of patients who did not receive early physiotherapy (NO-ePHYS). On the other hand, the period specified between May 2020 and February 2022 consists of the patients who received early physiotherapy (YES-ePHYS).
Exclusion criteria for the original study were patients under 18 years old, negative PCR for SARS-COV-2, proven or suspected acute primary brain injury that could lead to global impairment of consciousness or cognition, patients who did not require invasive MV during ICU admission, second or subsequent ICU admissions during their hospital stay, patients with a DNR-order at ICU discharge, patients transferred to another centre and ICU length-of-stay under 48 h.
The research was approved by the Francisco de Vitoria University's Research Ethics Committee, (reference number 58/2021, 3rd November 2021). Procedures were followed in accordance with the ethical standards of the responsible committee on human experimentation (institutional or regional) and with the Helsinki Declaration of 1975. Participation and acceptance of the inclusion of the patient's data were obtained by signing the informed consent document (by the patient or by an authorized surrogate in case the patient could not express their opinion).
We collected relevant demographic and clinical data from every patient. Early physiotherapy protocol involved mobilization and respiratory physiotherapy. The early mobilization protocol was based on the index given by the Medical Research Council (MRC)18–20 which was extrapolated to the Intensive Care Mobility Scale (IMS)21,22 to decide the type of motor physiotherapy that each patient required. The respiratory physiotherapy protocol was based on internationally established decision algorithms once the patient met the minimum clinical stability conditions for a safe intervention for the patient, with the main objectives of increasing lung recruitment and improving secretion management.23
We define early physiotherapy as treatment starting within 24−48 h after ICU admission or as soon as the patient meets the minimum clinical stability criteria to be able to start these sessions (mainly considering cardiac and respiratory parameters). The therapy session was interrupted if the patient developed clinical instability. Patients engaged in progressive mobilization, starting with a range of motion and advancing to bed mobility activities, transferring to an upright position, sitting, standing, marching in place, and walking, as tolerated. Progression of activities was dependent on patient tolerance and stability. Therapy sessions featured one daily session (duration ranged from 20 to 30 min) performed throughout hospital admission until discharge. All rehabilitators and physiotherapists worked primarily in the critical care setting and had previous training and extensive experience in the ICU. Early mobilisation algorithm and mobilisation protocol are available in the Supplementary material (Figs. S1 and S2).
Patients completed telephone interviews one year after ICU discharge and provided their current health status. Perceived quality of life was assessed using the following tools: Barthel Index (BI; a generic measure of physical or functional disability, which evaluates the patient's level of independence concerning basic activities of daily living24), Clinical Frailty Scale (CFS; assessment of frailty25,26 and the 36-Item Short Form Health Survey (SF-36; composed of 36 questions (items) that assess both positive and negative states of health27). We collected CFS before ICU admission and CFS one year after hospital discharge (hDC); BI at ICU discharge, at hDC, and one year after hDC; and SF-36 one year after hDC. One year after hospital discharge, the perceived quality of life was assessed by telephone.
The primary outcome was a change in health status from pre-admission to one year after hDC, measured by the Clinical Frailty Scale (CFS). The secondary outcome was the Barthel Index (BI) change and the observed SF-36 health rating scale across the same timeframe.
Statistical analysisCategorical or discrete variables are expressed as numbers and percentages, while continuous variables are expressed as medians and interquartile ranges. The assumption of normal distribution for each scale and continuous variable was rejected (P < .001) using the Anderson-Darling and Shapiro-Wilk tests. The association between having received physiotherapy and other categorical variables was performed using Pearson's Chi-square test or Fisher's exact test. The intensity of the association was analyzed by Cramér's V (no association between the variables if Cramér's V < 0.2, a small association for values of 0.2, a moderate association for values 0.2−0.6, and a strong association for values >0.6).28 Differences in scale scores and other quantitative variables between the NO-ePHYS and YES-ePHYS groups were analyzed using Mann Whitney U test before matching both groups using the propensity score technique, and Wilcoxon paired test after the matching process (generalized linear model with logit function was used to estimate the propensity score and optimal method was used to perform the matching by SAPS-3 scores, age, and sex). The effect size, and thus the magnitude or intensity of the differences found, was assessed using Wilcoxon r effect size (small effect from 0.1 to 0.4, medium 0.4−0.6, and large ≥ 0.6).29
P values <.05 were considered statistically significant. Statistical analysis was conducted using R version 3.6.3 (R Core Team, 2020).
A sensitivity analysis was also performed with other strategies based on the propensity score methodology, first estimating the matching using methods other than the optimal one available in the R software packages “ipw” (version 1.2.1) and “MatchIt” (version 4.7.1) (version 3.6.3).30–32 Finally, propensity score weighting was performed using the Inverse Probability of Treatment Weighting, running robust linear multiple regression models (R package MASS, version 7.3–60.2).33
ResultsUnpaired analysisOf 1057 patients, 99 were eligible for inclusion: 35 patients from the NO-ePHYS and 64 from the YES-ePHYS. Baseline characteristics, demographic data, and hospital outcomes are reported in the Supplementary material (Table S1). In our cohort, patients within both groups shared similar demographic data. However, differences were observed for SAPS-3 on admission: (MdnNo-ePHYS = 59 [53.5–64.5]; MdnYes-ePHYS = 53 [47–58]; P = .001). Regarding clinical data, we did not observe differences when analyzing days under invasive mechanical ventilation (MdnNo-ePHYS = 11 [9−16]; MdnYes-ePHYS = 15 [9−25]; P = .2), although more patients in the NO-ePHYS group required non-invasive mechanical ventilation after extubation (82.9% in NO-ePHYS; 30.8% in YES-ePHYS; P < .001). Time from intubation to the first rehabilitation session was 3 [1.5–6] days in the YES-ePHYS group and 9 [6.5–13] in the NO-ePHYS group (P = .002). No significant differences were observed in length of stay and mortality during the study period.
Of the 99 patients, 76 answered the telephone surveys (77.6% response rate). Five patients in the NO-ePHYS group (14.3%) and 17 patients in the YES-ePHYS group (27%) did not answer, with no statistically significant differences between the two groups (P = .15). Regarding the telephone survey recipient, two surrogates within the NO-ePHYS group (6,7%) and 11 surrogates within the YES-ePHYS group (23,9%) answered the questions (P = ,065). 82,9% of the surveys were patient-self-completed.
The sensitivity analysis performed with other strategies based on propensity score methodology is shown in Supplementary material (Section S2).
Propensity score matchingGiven the possibility of bias when assessing the results obtained in the surveys related to the quality of life, since statistically significant differences were found in the severity score on admission, we performed a propensity score matching according to SAPS-3 values, sex, and age (Fig. 1A). As shown in Fig. 1B, the standardized mean difference between both groups is reduced for the overall propensity score (PS) and each of the variables introduced in the PS model. As shown in Fig. 1C, the severity score of admission is significantly the same in both groups after the matching process (P = .11).
Paired analysis
After the matching process by propensity score, no differences were observed in most demographic and clinical variables (Table 1). We did observe statistically significant differences in respiratory comorbidity (8.6% in NO-ePHYS, 31.4% in YES-ePHYS; P = .03; Cramér's V 0.25), the use of non-invasive mechanical ventilation (17.1% in NO-ePHYS, 68.6% in YES-ePHYS; P < .001; Cramér's V 0.491) and the hospital length-of-stay after ICU discharge (MdnNo-ePHYS = 15 [10−28]; MdnYes-ePHYS = 22.5 [16.8−37]; P = .036; Wilcoxon r effect size = 0.281). No apparent difference was observed when assessing MRC scoring. Patients who belonged to the YES-ePHYS showed a lower MRC at the beginning of rehabilitation (MdnNo-ePHYS = 38 [34−45]; MdnYes-ePHYS = 36 [0−42]; P = .026; Wilcoxon r effect size = 0.214), but the MRC score on hospital discharge had the same value for both groups (MdnNo-ePHYS = 48 [46−50]; MdnYes-ePHYS = 48 [46−50]; P = .04; Wilcoxon r effect size = 0.082).
Demographic and clinical data after propensity score matching.
| NO-ePHYS (N = 35) | YES-ePHYS (N = 35) | P-value | r Wilcoxon size effect 4 | Cramer V 5 | ||
|---|---|---|---|---|---|---|
| Age, yr, median (IQR) | 65.0 (59.5−69.5) | 66.0 (56.5−74.5) | >,9 1 | 0.05 | – | |
| Sex, n (%) | Male | 20 (57.1%) | 19 (54.3%) | >.9 2 | – | 0.00 |
| Female | 15 (42.9%) | 16 (45.7%) | ||||
| Comorbidities, n (%) | Cardiovascular | 13 (37.1%) | 19 (54.3%) | .2 2 | – | 0.143 |
| Respiratory | 3 (8.6%) | 11 (31.4%) | .034 3 | – | 0.25 | |
| Renal | 13 (37.1%) | 19 (54.3%) | .2 2 | – | 0.143 | |
| Hepatic | 2 (5.7%) | 2 (5.7%) | >.9 3 | – | 0 | |
| Cancer disease | 6 (17.1%) | 6 (17.1%) | >.9 2 | – | 0 | |
| Endocrine | 16 (45.7%) | 13 (37.1%) | .6 2 | – | 0.058 | |
| Mechanical ventilation data | Days under invasive MV, days (IQR) | 11 (9−16) | 15 (9−25) | .2 1 | 0.15 | – |
| Reintubation, n (%) | 6 (17.1%) | 2 (5.7%) | .3 3 | – | 0.135 | |
| Non-invasive MV, n (%) | 6 (17.1%) | 24 (68.6%) | <.001 2 | – | 0.491 | |
| Need of prone position, n (%) | 17 (48.6%) | 24 (68.6%) | .14 2 | – | 0.174 | |
| Neuromuscular blockade, n (%) | 27 (77.1%) | 19 (54.3%) | .077 2 | – | 0.211 | |
| Organ failure, n (%) | Cardiovascular | 31 (88.6%) | 27 (77.1%) | .3 2 | – | 0.114 |
| Respiratory | 33 (94.3%) | 35 (100.0%) | .5 2 | – | 0.086 | |
| Renal | 12 (34.3%) | 9 (25.7%) | .6 2 | – | 0.062 | |
| Hepatic | 2 (5.7%) | 0 (0.0%) | .5 3 | – | 0.086 | |
| Hematologic | 6 (17.1%) | 4 (11.4%) | .7 3 | – | 0.041 | |
| MRC, median (IQR) | At the first rehabilitation session (MRCstart) | 38 (34−45) | 36 (0−42) | .026 1 | 0.214 | – |
| On hospital discharge (MRCend) | 48 (46−50) | 48 (46−50) | .4 1 | 0.082 | – | |
| Difference MRCend-MRCstart | 10 (6.5−12) | 12 (6−31) | .041 1 | 0.104 | – | |
| ICU-LOS, days, median (IQR) | 14 (10−30.5) | 22.5 (16−33.2) | .051 1 | 0.24 | – | |
| Hospital-LOS after ICU discharge, days, median (IQR) | 15 (10−28) | 22.5 (16.8−37) | .036 1 | 0.281 | – | |
Yr = years; IQR = interquartile range; MV = mechanical ventilation; therapy; MRC = Medical Research Council; LOS = length-of-stay.
Table 2 shows the results of the surveys carried out during the study period after propensity score matching. We did not observe statistically significant differences between groups except for two SF-36 items: social activities (MdnNo-ePHYS = 56.2 [37.5−71.9]; MdnYes-ePHYS = 75 [62.5−97.5]; P = .004) and one-year health transition (MdnNo-ePHYS = 50 [50−75]; MdnYes-ePHYS = 75 [50−100]; P = .031), where patients who belonged to NO-ePHYS had lower scores than those who did receive early physiotherapy (YES-ePHYS). Regarding size effect, we observed a moderate-large effect size in social activities (Wilcoxon r effect size = 0.5) and a small-medium size effect in one-year health transition items (Wilcoxon r effect size = 0.29).
Quality of life questionnaires during the study period for propensity score matching population.
| Survey | NO-ePHYS (n = 35) | YES-ePHYS (n = 35) | P-value 1 | r Wilcoxon size effect 2 | |
|---|---|---|---|---|---|
| Barthel Index, median (IQR) | Upon ICU discharge | 40 (25−70) | 40 (25−60) | >.9 | 0.013 |
| Upon hDC | 75 (55−90) | 60 (45−92.5) | .3 | 0.109 | |
| 1 yr after hDC | 90 (80−95) | 100 (80−100) | .4 | 0.262 | |
| CFS, median (IQR) | On ICU admission | 3 (2−3) | 3 (3−3) | .067 | 0.234 |
| 1 yr after hDC | 4 (3−4) | 3 (3−4) | >.9 | 0.169 | |
| SF36 performed 1 yr after hDC, median, (IQR) | Limitations of activities | 55 (35−72.5) | 75 (45−85) | .3 | 0.287 |
| Physical health problems | 37.5 (0−68.8) | 75 (0−100) | .11 | 0.222 | |
| Emotional health problems | 66.7 (33.3−100) | 66.7 (33.3−100) | .9 | 0.079 | |
| Energy | 55 (41.2−65) | 70 (50−80) | .12 | 0.304 | |
| Mental health | 64 (56−75) | 72 (56−80) | .5 | 0.116 | |
| Social activities | 56.2 (37.5−71.9) | 75 (62.5−97.5) | .004 | 0.502 | |
| Pain | 55 (45−77.5) | 67.5 (55−77.5) | .4 | 0.190 | |
| General Health | 47.5 (31.2−55) | 55 (45−70) | .2 | 0.264 | |
| One-year health transition | 50 (50−75) | 75 (50−100) | .031 | 0.289 | |
Yr = years; IQR = interquartile range; hDC = hospital discharge.
Finally, sensitivity analysis with other strategies based on propensity score methodology and propensity score weighting through Inverse Probability of Treatment Weighting (using robust linear multiple regression models) are shown in the Supplementary material (Table S3). Having received or not rehabilitation presented significant weighting (P < .05) in Barthel Index at hospital discharge, CFS at ICU admission, and SF36 (health transition and social function).
DiscussionThis prospective cohort study examined the association between early physiotherapy performed in mechanically ventilated patients and the perceived quality of life. Even though we did not observe differences in both groups when analyzing functional items related to the quality of life, patients who received ePHYS expressed better social life and perceived health improvement (measured by a one-year health transition item).
Multidisciplinary management of critically ill patients, including the approach to different aspects of the post-ICU syndrome,34 has become increasingly relevant in recent years. This has led to a growing literature on the advantages of early rehabilitation in critically ill patients. Most clinical trials aim to reduce ICU-AW incidence, associated with poor short-term outcomes (including decreased ventilator-free days and discharged-to-home rate), poor long-term survival, physical functioning, and quality of life.14,19,35,36 All this mounting evidence has led to the publication of international guidelines recommending early rehabilitation.37
A recent study52 showed that early rehabilitation in the ventilated patient was not associated with a better prognosis and is even related to fewer adverse events. However, we believe the differences observed in the rehabilitation time dedicated to each of the two groups (20.8 ± 14.6 min vs. 8.8 ± 9 min) can explain some of the published findings. Furthermore, the implementation of rehabilitation requires consensus between different team members. There is also no clear-cut definition of physiotherapy's optimal duration and intensity, as it depends on many factors.37 What seems quite clear is that delayed mobilization of ICU patients imparts no discernible improvement in any important outcomes,38 even though the adequate timing to introduce mobilization therapy during ICU stays to maximize progress in patient outcomes remains controversial.39 Regarding dose or intensity, Paton et al.40 observed improved health status six months after ICU admission when they achieved higher levels of mobilization but did not increase the number of active mobilizations. In our study, we performed one session per day of 20−30 min.
The Barthel Index, Clinical Frailty Scale, and SF-36 measured health-related quality of life. Previous studies41,42 show no differences between the two groups regarding SF-36 physical component score, SF-36 mental component score, or functional independence measures. However, when analyzing time-to-start therapy, most patients were enrolled after a median of 4 days of ventilation and received physical rehabilitation three days afterwards, suggesting a 7-day therapy delay. Morris et al.43 managed to start standardized rehabilitation therapy (SRT) with a median day-to-first therapy of 1 (0–2) compared with the day-first therapy exercise for the usual care group (7-IQR, 4–10-). At six months, the SRT group had higher scores for the SF-36 physical function scale score (difference, 12.2 [95% CI, 3.8–20.7], P = .001), with no effect on SF-36 mental health score (P = .19) or SF-36 physical health score (P = .05). Furthermore, Patel et al.44 observed higher physical component scores on quality-of-life testing (measured with SF-36 physical component score: 41,1 [IQR 31,8–49,4] vs 52 [45,3–56,8]; P < ,0001) and improved long-term cognitive impairment (measured with MoCA score: 23 [21–26] vs. 26 [IQR 24–28]; P = .0001) in the intervention group compared with the usual care group. Likewise, and despite efforts to promote the start of rehabilitation in critically ill patients, patients in the YES-ePHYS group had a median of 3 (1.5–6) days between the day of intubation and the first day of physical therapy. This shows the inherent difficulty in initiating early mobilization protocols, which is very much present in the COVID-19 pandemic.
Our study has several potential limitations. First, although a single-centre study with a relatively limited number of patients limits generalizability, we believe it may benefit data homogeneity and management consistency. Secondly, we have faced the challenges of conducting long-term outcome studies in critical care.45,46 Though social science literature suggests a minimum rate of 70%–80%,47 no such guidelines exist for ICU survival studies. Low follow-up rates could be a significant limitation, as they may affect the study's validity. Other treatments administered after hospital discharge, such as continued rehabilitation services, cognitive therapy, psychotherapy, or receipt of psychoactive medications, were not measured and could modify these findings.
Third, regarding the MRC score and questionnaire responses, we could only perform a more exhaustive analysis of the time effect for the BI, as we obtained results at the three previously defined times. MRC at baseline was somewhat worse in the YES-ePHYS group (2 points less), with no differences at hospital discharge. This slight difference at baseline could explain the statistically significant difference when assessing the temporal progression of MRC. It should be noted that both groups, after propensity score, are homogeneous, making it somewhat difficult to find a plausible explanation for why the YES-ePHYS group had a slightly lower MRCstart value than the NO-ePHYS group. The SF-36 survey was only used one year after hospital discharge, precluding a temporal analysis of the results. Moreover, due to our small sample size, we performed a propensity score matching only with the variables suspected to be confounders and measured before the exposure occurred. With such a small sample size, the model could not support the inclusion of more variables.48
Fourth, we did not assess our patients' cognitive impairment adequately. We considered cognitive assessment during the study design process but eventually discarded it, given that an adequate evaluation of this aspect requires a healthcare visit (risking the total number of patients who completed the follow-up). Additionally, all outcome measures were collected through self-administered questionnaires via telephone interviews, although some responses were provided by caregivers. This approach may have introduced recall bias, as patients were asked to retrospectively assess their health status and quality of life one year after hospital discharge, potentially leading to inaccuracies in memory or perception. Furthermore, social desirability bias may have influenced responses, with participants possibly overstating positive outcomes or minimizing difficulties to align with perceived expectations from healthcare professionals.
Finally, we must acknowledge the temporal distribution of the patient groups, which may have introduced a time-related bias. Patients in the control group were admitted during the first wave of the pandemic (March–April 2020), a period marked by an overwhelming healthcare burden, limited understanding of COVID-19, and rapidly evolving treatment protocols. In contrast, patients in the intervention group were admitted from May 2020 onwards, when clinical knowledge had improved and additional healthcare staff had been hired, reducing the workload. Consequently, improvements in clinical outcomes could partially reflect broader changes in care over time, and not solely the effect of the physiotherapy intervention.
The present study also has several strengths. Considering our relatively small ICU capacity (between 8 and 10 available ICU beds), we included a high sample size. The findings of our study reinforce our commitment to continue working on our multidisciplinary protocol for the management of post-ICU syndrome (coordinating both the hospital team and the Primary Care health centres attached to the hospital area to which we belong). Our post-discharge follow-up programme ensures a successful handover with the inpatient ward team and discusses the next steps with the patient and family. We also offer a support programme for relatives and caregivers, considering the patient's values and wishes in the shared decision-making process. Despite being a young programme, so far, we have achieved encouraging results, observing improvement in components of mental health (fear, self-esteem, coping, sleep disorders), in the patient's ability to perform basic activities of daily living and in the perceived caregiver overload.49–51
Implementation of ICU mobilization and rehabilitation is highly variable. Reasons found to explain why early physiotherapy does not occur routinely are a lack of dedicated ICU physical and occupational therapists, lack of awareness and uptake by clinicians, concerns regarding the safety of mobilization, and excessive sedation.52,53 Hodgson et al.54 provide ten tips or strategies to help ICU clinicians optimize early mobilization and rehabilitation.
This study brings the holistic management of critically ill patients to the table. We should once and for all encourage comprehensive care, focusing our attention on both the physical and psychological issues of our patients.
ConclusionsPatients who received early physiotherapy reported better outcomes in specific aspects of quality of life, namely social functioning and perceived health transition, one year after hospital discharge. However, no significant differences were observed in key functional or mental health scales, suggesting a limited but potentially meaningful clinical impact of the intervention in selected domains.
CRediT authorship contribution statementBLV: investigation, writing original draft, editing. LMC: investigation, writing original draft, editing, and reviewing. ARA: investigation, editing, and reviewing. RCV: software and formal analysis, editing, and reviewing. RML: methodology, reviewing and editing. AAA: reviewing and editing. PBM: reviewing and editing. JDJ: reviewing and supervising. FG: methodology, project administration, supervision, and reviewing.
Ethics approval and consent to participateThe research was approved by the Francisco de Vitoria University's Research Ethics Committee. Participation and acceptance of the inclusion of the patient's data were obtained by signing the informed consent document.
Declaration of Generative AI and AI-assisted technologies in the writing processManuscript was written without use of any form of AI.
Funding statementThis research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Data availabilityData are available upon request.
The authors declare they have no competing interests.
The following is the supplementary data to this article:
