Trial design
This randomized controlled and blinded clinical trial was conducted to evaluate the effects of CCM or FCEM or both on the life satisfaction of ARDS survivors. From December 2009 to October 2014, 140 ARDS survivors from mixed medical-surgical ICUs at four academic teaching hospitals in Tehran, Iran, randomly assigned to one of three intervention groups (A, B or C) or a control group (D). Pre- and post-interventions, pulmonary functions and HRQoL status of patients in all groups were collected 48 times through clinical measurements and various questionnaires during 5 years of follow-up. All parts of the study were reviewed according to the Consolidated Standards of Reporting Trials (CONSORT) statement 32. Ethic committees including the institutional investigative review boards at Baqiyatallah University of Medical Sciences, Shahid Beheshti University of Medical Sciences, and Tehran University of Medical Sciences, waived the need for the study protocol approval. The trial was registered with Clinicaltrials.gov (NCT02787720) on 01/06/2016. In addition, the study was conducted according to the Helsinki Declarations guideline. Written consent was obtained directly from the patient on ICU discharge. Informed consent covered both study participation and consent to publish the findings. Surrogate consent from the patient’s legal guardian or healthcare proxy was permitted in cases where the patient did not have decision-making capacity. Patients and their designated family/friend were enrolled as a unit. In addition, the study was conducted according to the Helsinki Declarations guideline 33.
Eligibility criteria of participants
From December 2009 and October 2014, ICU patients with ARDS were screened for eligibility. The diagnosis of ARDS was based on the Berlin definition1, which was as follows; patients with a BMI<40 suffering from ARDS with PaO2/FiO2 < 300 mmHg during mechanically ventilated (MV) with an expected duration of controlled MV of more than 24 hours and ability to tolerate PEEP titration (up to 21 or 15 cmH2O). In addition, patients were eligible to enroll in this study if they met the following criteria: (a) age ≥18 years, (b) able and willing to provide informed consent, (c) willingness of a family member or friend of patients to participate in the study, (d) has basic health literacy and can fill out questionnaire, and (e) full code status. Patients aged less than 18 years and more than 85 years, pregnant women, patients at end-stage medical condition, patients with a history of pulmonary rehabilitation, and patients with a history of neurological or psychiatric disorders were excluded from the study.
Sample size, randomization and blinding
Power calculations determined that 32 patient/family units were needed in each group to achieve a 95% confidence level and a 90% power. Participant enrolment was via convenience sampling. With regard to the sample loss, the sample size was finally calculated to be 35, in each of the intervention and control groups. Eligible patients admitted to ICU were enrolled within 24 h and randomly assigned to one of three intervention groups (A, B or C) or a control group (D). Group (A) who have received both FCEM and CCM programs via trainer. Group (B) received only FCEM via family member/friend as trainer. Group (C) who received only CCM via researcher as trainer, and group (D) who used routine care as a control group. Randomization was achieved with a computer-generated random block design, which was drawn up by an expert statistician who had no clinical involvement in the trial before the beginning of the study. Block randomization was accomplished using Random Allocation Software© (Informer Technologies, Inc., Madrid, Spain) by a computer-generated random. Randomly allocated numbers were placed into sequential containers (i.e. ABCD, containers 1-4; BADC, containers 5-8, etc.) which were kept in a secure location until allocation consignment. The difference between groups was not disclosed, patients consented knowing that they were undergoing pulmonary rehabilitation (PR), but without knowing the details. The assignment was made through confidential communication between the patient's and a third party not involved in the recruitment process. So, patients and their family members, as well as data analyzers were blinded to the assignment group and the differences between the groups.
Intervention
The intervention package had three phases including pre-intervention, intervention, and post-intervention phases (Supplementary file, Figure 1).
Pre-intervention phase: Prior to ICU discharge in pre-intervention phase, life satisfaction in all ARDS survivors was assessed through several multiple questionnaires and tests in terms of physical and mental health as well as quality of life index. Short form health survey of quality of life (SF-36) questionnaire, perceived stress questionnaire (PSQ-14), state/trait anxiety questionnaire, and Kessler Psychological Distress Scale (K10) were used as soft outcomes. Hard outcomes includes Barthel Index (BI) activities of daily living, six-minute walk test (6MWT), and free walking index (WI) test.
Intervention phase: In the intervention phase, group “A” as the first group of intervention groups received both FCEM and CCM programs with a focus on continuity and support from friends/family. The second group of intervention groups “group B” received only FCEM rehabilitation technique, with additional support from trainer. The third group of intervention groups “group C” received only CCM rehabilitation program, with additional support from researcher. While the control group “group D” received routine care without any external organization. The intervention had four stages; first stage was the awareness and cognition. The patient was evaluated for their insight into their perceived illness severity and perceived sensitivity, or the degree to which they felt threatened by their illness 34. This was performed by means of 3–5 group sessions in the pre-intervention phase. Group sessions included 3–5 patients and lasted for 45–60 min each. Session content included assessments of the participants’ psychological and physical conditions as well as their attitude towards the nature, definition, risk factors, symptoms, medical and nursing care, and complications resulting from ARDS. In a second stage, patients were assessed for their expectations over 3-5 one-hour sessions34. Groups of 3–4 patients shared and learned from each other under the moderation and guidance of the principal researcher. In a third stage, the degree of patient acceptance was assessed using an educational participation method in group discussion. Patients reached practical solutions through using the problem-solving findings of the previous stage. Fourth stage consisted of formative and summative evaluations34. All patients contact their nurse every 2 days to report problems or complications. A multidisciplinary medical team evaluated the patient as a group. Thus, patients were evaluated weekly for 30 days by a pulmonologist, intensivist, internist, psychologist, psychiatrist, physiotherapist and occupational therapist. At each visit, the patient was interviewed, underwent a physical examination, pulmonary-function tests (PFT), posterior-anterior and lateral chest radiography (following discharge), resting oximetry, a standardized 6MWT with continuous oximetry, a free walking index (WI), and laboratory tests. In addition, at any time, if experienced a problem or complication, patients can notify the intensivist, multidisciplinary medical team or their primary care provider. The aim of the formative evaluation was to encourage patients to internalize their locus of control by encouraging personal responsibility about issues of health and seeing his/her self-empowerment. Summative evaluations were performed to evaluate the influence of the intervention on the study's medical outcomes.
Post-intervention phase: All patients were followed for 5-year and their outcomes were assessed 48 times during 5-year or 60 months as follows; in the pre-intervention as baseline, monthly for 42 months continuously post-intervention, in months 45, 48, 51, 54, 57, and 60. During the 60-month follow-up period, patients attended a total of 56 support group webinars addressing topics including returning to work, intimate relationships, nutrition, sleep hygiene, tobacco use, exercise, and leisure activities as well as testing and laboratory issues. Follow-up interviews were conducted during home visits, when available, or with the assistance of telephone, Skype©, Viber©, WhatsApp© and social media options (e.g. Facebook©) according to patient preference.
Role of the Designee
Following informed consent, the designated family member or friend (hereafter called designee) continued through the study with the patient as a ‘unit’. The designee attended the patient’s educational sessions during stages 3 and 4, with stage 2 being optional according to designee preference. Recall that stage 2 deals with patient expectations, stage 3 with patient acceptance and problem-solving, and stage 4 with evaluations and internalizing his/her locus of control. The designee and the patient attended the same sessions, and studied the same learning materials. Up to eight family members were allowed to join in the educational sessions if requested. In stage 3, the designee was charged with learning and reinforcing educational material with the patient. In stage 4, when instructed by study investigators, the designee would administer the KAP assessments to the patient. In addition to scores, the designee would provide additional information on the patient’s home situation and current condition. Patients were assigned a code, and de-identified data were transmitted from the designee to investigators either by encrypted email, telephone, mail or in person.
Rehabilitation plan
All patients had similar inpatient rehabilitation programs. For patients in the three intervention groups, outpatient rehab included daily exercise for 0–2 h/day. Exercise occurred between 8:00 and 10:00, and types included walking, jogging, bicycle, swimming, or other exercise according to patient preference or resource availability and confirmed by multidisciplinary medical team. Daily exercise was supervised by designated family members. Investigators randomly attended sessions in an unannounced fashion. Although not routinely involved, multidisciplinary medical team consultations were available on investigator request. Exercise data were independently collected from the patient and their designated family member on a weekly basis (κ statistic=0.94). To measure the patients’ walking distance, investigators provided them with a Fitbit® (Fitbit, San Francisco, USA). Routine care included education on smoking cessation and education on food selection. Patients were provided printed materials, with dietician assessments available on request. Patients in the control group received the same education and printed materials during their inpatient course. Patients exercised daily, at any time, for ≤2 h according to patient tolerance. Sessions were supervised by family members. Investigators did not attend sessions. Exercise data were independently collected from the patient and their designated family member on a weekly basis (κ statistic=0.22). Walking distance was measured using the Fitbit. Routine care included education on smoking cessation and education on food selection. Patients were provided printed materials, with dietician assessments available on request.
Data collection
Age, gender, body mass index (BMI), marital status, family number, household size, urban home location status, full- or part-time employment status, cause of ARDS, ICU and hospital length of stay were collected as socio-demographic and clinical characteristics of patients. Initial illness severity were compared between groups using validated scales including Simplified Acute Physiology Score (SAPS)-III35, 36, Sequential Organ Failure Assessment (SOFA)-II 37, 38, Acute Physiology and Chronic Health Evaluation (APACHE)-IV39, Lung Injury Score (LIS), and Multi Organ Dysfunction Score (MODS) at first day of ICU admission40. Moreover, select treatment requirements were includes renal replacement therapy, paralytic treatment (>1 bolus dose or continuous infusion), systemic stress-dose glucocorticoid treatment (e.g. Hydrocortisone 200mg IVP x 1 then 100mg IV three times daily for 5 days), tracheostomy placement, or use of high-frequency oxygenation technique including high frequency oscillation ventilation (HFOV) and high frequency percussive ventilation (HFPV). The high frequency techniques were specifically recorded as they are not part of routine ICU ventilator care unlike Bi-level ventilation or airway pressure-release ventilation (APRV). Prone positioning is not routinely used in Iranian ICUs.
Research instruments
Life satisfaction in all ARDS survivors was assessed through several questionnaires and tests in terms of physical and mental health as well as quality of life index. Data collection tools consisted of hard outcomes and soft outcomes. Hard outcomes included the Barthel Index (BI) activities of daily living index, six-minute walk test (6MWT), free walking index (WI) and pulmonary function tests (PFT). Soft outcomes included the short form health survey of quality of life (SF-36) questionnaire, the perceived stress (PSQ-14) questionnaire, state/trait anxiety, and the Kessler Psychological Distress Scale (K10). Both hard and soft outcomes, except the PFT, were evaluated 48 times over 5-year or 60 months as follows; in the pre-intervention as baseline, monthly for 42 months continuously after the intervention, in the months 45, 48, 51, 54, 57 and 60.
SF-36 questionnaire: The questionnaire consists of eight domains including physical functioning, physical role limitation, social functioning, bodily pain, mental health, mental role limitation, vitality and general health. Scoring of each domain was calculated independently and scores ranged from 0 (the worst) to 100 (the best)41, 42. The SF-36 questionnaire is a validated tool, whose reliability in this study was assessed by test–retest and Cronbach’s alpha, 0.93 and 0.94, respectively 43, 44.
The 14-item Perceived Stress Questionnaire (PSQ-14): Overall stress was measured using the validated 14-item perceived stress questionnaire (PSQ-14)45.PSQ-14 scores are obtained by reversing the scores on the seven positive items, including 0=4, 1=3, 2=2, 3=1 and 4=0, then, summing across all 14 items. Scores range from 14 to 70 46.The reliability of thePSQ-14 was assessed in this study by test-retest and Cronbach’s alpha, 0.95 and 0.96, respectively 47.
The State/trait Anxiety questionnaire: Patient anxiety was assessed through the State-Trait Anxiety questionnaire48. This validated tool has 20 items for assessing trait anxiety and state anxiety, respectively. All items are rated on a four-point scale ranging from very low (1 point) to very high (4 points). Higher scores indicate greater anxiety 34. The reliability of the State-Trait Anxiety questionnaire was assessed in this study by test-retest and Cronbach’s alpha, 0.92 and 0.94, respectively 49.
Barthel Index (BI) of Activities of Daily Living: The BI is an ordinal scale used to measure ADL’s. It uses 10 variables describing ADL and mobility 50. Scores range from 0 to 2 or 3 for each item. Total possible scores range from 0–20, with lower scores indicating increased disability. If used to measure improvement after rehabilitation, changes of more than two points in the total score reflect a probable genuine change, and change on one item from fully dependent to independent is also likely to be reliable. The reliability of the BI questionnaire was assessed in this study by test-retest and Cronbach’s alpha, 0.91 and 0.92 respectively 51.
Kessler Psychological Distress Scale (K10): The questionnaire is a 10-item assessment intended to yield a global measure of distress based on questions about anxiety and depressive symptoms that a person has experienced in the most recent four-week period 52. All items are rated on a 5-point scale ranging from “none” (1 point) to “all” of the time (5 points). The maximum score is 50 (severe distress) and the minimum is 10 (no distress) 53. The reliability of the K10 questionnaire was assessed in this study by test–retest and Cronbach’s alpha, 0.93 and 0.92 respectively 54.
Pulmonary function tests (PFT)
PFT was conducted twice for all patients, once before the intervention and once after the intervention. PFT was performed by spirometry (which assesses static and dynamic pulmonary volumes) and diffusion capacity (which assesses the capacity for gas exchange through the alveolar barrier) techniques to assess pulmonary function in this study. All tests were proceeded according to American Thoracic Society guidelines, standardization of Spirometry 2019 update 55. The expiratory volume in the first second (FEV1) and vital capacity (FVC) during forced expiration were recorded. Total lung capacity (TLC) and residual volume (RV) were measured using the plethysmography method, and diffusing capacity of the lung for carbon monoxide (DLCO) was measured using the single breath- hold method 56.
Statistical analysis
Assuming that the measure of the effect of FCEM and CCM on the parental burden of care (medium effect size) is at least d=10, the sample size in each group was determined to be 32, at a confidence level of 95%, and a test power of 80%. With regard to the sample loss, the sample size was finally calculated to be 35, in each of the intervention and control groups. In total, 140 patients were enrolled with the aim to manage the dropouts in this study. Statistical analysis was performed using IBM® SPSS® Statistics 21.0 (IBM Corp., Armonk, NY) and IBM® SPSS® AMOS™ 21.0 (IBM Corp., Armonk, NY). Discrete variables are expressed as counts and percentages. Power calculations determined that 32 patient/family units were needed in each group to achieve a 95% confidence level and a 90% power. Continuous variables are expressed as means and standard deviations. Analysis of variance (ANOVA) and Chi-Square analyses were used to compare numeric and discrete variables. Generalized Estimation Equations (GEE) and Panel Analysis were performed on longitudinal data, and the results were expressed as odds ratios (ORs). Multiple Cox proportional hazards regression was used to assess the effect of treatment groups on Barthel time and KPDS time. Structural Equation Modeling (SEM) was performed to examine the effects of clinical factors on outcomes of acute respiratory distress syndrome, and the model was evaluated using the root mean square error of approximation (RMSEA), the normed fit index (NFI), and the goodness-of-fit index (GFI). Statistical significance was defined as p-value <0.05.
Structural equation modeling (SEM) was used to assess for direct and indirect inter-variable associations (walking index that was checked every month) through mediators. Pulmonary factors including total lung capacity (TLC), diffusing capacity of the lungs for carbon monoxide (DLCO), residual volume (RV), forced vital capacity (FVC), forced expiratory volume in 1 second (FEV1), Kessler psychological distress scale (K10), and Barthel Index (BI) were added in the independent variable list. Descriptive statistics were used to summarize the data. Pearson correlation was employed to investigate the association between independent variables and other outcomes. SEM was used to find direct and indirect associations between independent variables and other outcomes. Since outcomes were not normally distributed (Kolmogorov-Smirnov P<0.05), the SEM model was estimated using Asymptotically Distribution-Free estimates (ADF). To assess model fit, indices including (normed fit index (NFI), incremental fit index (IFI), Tucker-Lewis index (TLI), and comparative fit index (CFI), root mean square error of approximation (RMSEA), and chi T square/DF ratio (χ2/DF) were calculated as goodness-of-fit index(GFI). Direct association was reported in standardized parameters, whereas indirect association was reported through the mediator’s effects. If both direct and indirect effects were significant for a proposed covariate, the total association was estimated by sum-of-two effects. Modeling and statistics were performed using IBM® SPSS® AMOS™ 21 (IBM Corp., Armonk, NY) 57.