Family-Based Psychosocial Support and Education as Part of Pulmonary Rehabilitation in COPD A Randomized Controlled Trial

BACKGROUND: Involving family as part of the patient’s rehabilitation plan of care might enhance the management of COPD. Th e primary aim of this study was to investigate the impact of a family-based pulmonary rehabilitation (PR) program on patients and family members’ coping strategies to manage COPD. METHODS: Family dyads (patient and family member) were randomly assigned to family-based (experimental) or conventional (control) PR. Patients from both groups underwent exercise training three times a week and psychosocial support and education once a week, during 12 weeks. Family members of the family-based PR attended the psychosocial support and education sessions together with patients. In the conventional PR, family members did not participate. Family coping and psychosocial adjustment to illness were assessed in patients and family members of both groups. Patients’ exercise tolerance, functional balance, muscle strength, and health-related quality of life were also measured. All measures were collected pre/post-program. RESULTS: Forty-two dyads participated (patients: FEV 1 , 70.4% 22.1% predicted). Patients ( P 5 .048) and family members ( P 5 .004) in the family-based PR had signifi cantly greater improvements in family coping than the control group. Family members of the family-based PR had signifi cantly greater changes in sexual relationships ( P 5 .026) and in psychologic distress ( P 5 .033) compared with the control group. Patients from both groups experienced signifi cant improvements in exercise tolerance, functional balance, knee extensors strength, and health-related quality of life aft er intervention ( P , .001). CONCLUSIONS: Th is research supports family-based PR programs to enhance coping and psychosocial adjustment to illness of the family system. TRIAL REGISTRY: ClinicalTrials.gov; No.: NCT02048306; URL: www.clinicaltrials.gov CHEST 2015; 147 ( 3 ): 662 672 [ Original Research COPD ]

Pulmonary rehabilitation (PR) has been demonstrated to be eff ective for patients with COPD during stable periods or shortly aft er an exacerbation. 1 Th is intervention has been also acknowledged as an important component of integrated care to manage COPD. 1 However, successful integrated care interventions demand the involvement of both patients and family members in care planning, implementation, and oversight. [2][3][4] Th e impact and challenges of living with a patient with COPD at all grades are well described, [5][6][7][8][9] including physical and emotional burden and distressing symptoms (eg, anxiety and depression). 5,10,11 Moreover, in some research, families have expressed the need for more information about disease management and for emotional support (eg, how to handle breathlessness, exacerbations, and anxiety symptoms). 8,[10][11][12] Attending to patients' and family members' needs, preferences, and expectations might have potential to promote a more integrated and collaborative approach to care in COPD. 13,14 Family interventions have been shown to improve family coping in chronic diseases such as diabetes, 15 cardiovascular disease, 16 and breast cancer, 17 but their impact has received limited investigation in COPD. Furthermore, the use of more positive coping and problem-solving strategies has been associated with better health outcomes, namely less depression and anxiety [18][19][20] and improved exercise tolerance 20 and quality of life 19 in patients and better self-rated physical and mental health in family members 5 living with COPD. However, only one study was identifi ed that tested benefi ts of including family members in a multidisciplinary PR program. 21 Th erefore, the primary aim of this study was to investigate the impact of a family-based PR program on patients and family members' coping strategies to manage COPD. It was hypothesized that participation in a family-based PR program would improve coping strategies of the family system without interfering with patients' benefi t obtained from a conventional PR program. Th e secondary aims were to explore its impact on family psychosocial adjustment to illness and patients' exercise tolerance and health-related quality of life.

Study Design
Th is was a single-blinded, randomized controlled trial. Family dyads (ie, patient with COPD and family member) were randomly assigned to family-based PR (experimental) or conventional PR (control) and were unaware of group allocation. Participants were only told that they were entering a PR program that involved the family and that, depending on group allocation, the involvement of the family member would diff er.
Th e outcome measures were collected from patients and family members 3 days before and aft er the PR program. Th e family-based PR was conducted at a diff erent time than the conventional PR. Randomization was performed by a computer-generated schedule in random blocks of three. Th e allocation sequence was kept in sealed opaque envelopes by a researcher who was not involved in data collection. Th is researcher drew the envelope and scheduled dyads of both groups. Approval for this study was obtained from the Center Health Regional Administration (2011-02-28) and national data protection committee (8940/2012). Written informed consent was obtained from each participant. Th is study was reported according to CONSORT (Consolidated Standards of Reporting Trials) recommendations. 22

Participants
Consecutive patients with stable COPD were recruited from three primary care centers. Patients were considered eligible for the study if they (1) were diagnosed with COPD according to the GOLD (Global Initiative for Chronic Obstructive Lung Disease) criteria; (2) had a family member Ն 18 years old who provided physical and/or supportive care, without receiving any payment; and (3) were able to provide informed consent to participate in the study. Patients were excluded if they had exacerbations or hospital admissions 1 month prior to the study, severe neurologic/musculoskeletal conditions, and/or unstable cardiovascular disease. Dyads were excluded if one of them presented severe psychiat-ric conditions or inability to understand and cooperate or if one of them refused to participate.

Intervention
In both groups, patients underwent 12 weeks of PR composed of exercise training and psychosocial support and education, conducted in primary care centers. Family members assigned to the family-based PR participated in the psychosocial support and education component together with patients. Family members randomized to conventional PR did not attend the sessions with patients, with the exception of sessions used to obtain baseline and post-intervention assessment data.
Exercise Training: Training frequency was three sessions per week. Sessions lasted 60 min and were delivered by the same physiotherapists in both groups, ensuring a consistent and uniform training among all patients. Th is component is described elsewhere. 23 Psychosocial Support and Education: Sessions were designed based on a comprehensive literature review on COPD rehabilitation, 24-26 needs of families living with COPD , 7,27,28 and interventions for families living with other chronic diseases. 29,30 Education aimed to provide information about COPD, increase the skills of the family to adjust to and manage the disease, and promote adherence to therapy and healthy lifestyles. Psychosocial support intended to help the family to manage the emotional demands of living with COPD, facilitate the communication within the family and with health/social services, and develop a sense of family identity, enhancing its cohesion.
Weekly sessions, lasting approximately 90 min, were conducted by a multidisciplinary team (physiotherapist, gerontologist, psychologist, nurse, and clinician). Th ese professionals assumed the role of facilitators by supporting participants in their doubts, encouraging them to share experiences, normalizing emotions, and assuming an empathic attitude. Several didactic methods were used during the sessions, such as group discussions, home tasks, role playing, and brainstorming.   Table 1 . At each session, a handout was provided to participants. Content presented to both groups was similar; however, in the control group family members did not participate, and, therefore, the content relating to psychosocial and educational topics solely focused on the patient's perspective.

Outcome Measures
Descriptive Characteristics: Sociodemographic information (age, sex, education, marital status, and current occupation) was collected from patients and family members. BMI, activities limitation resulting from dyspnea (assessed with the Modifi ed British Medical Research Council questionnaire 31 ), and lung function, 32 assessed with a portable spirometer (MicroLab 3500, CareFusion Corporation ), were collected from patients. Data on the kin relationship with the patient and the caregiving duration was obtained from family members. Patients and family members fi lled in the Family Crisis Oriented Personal Scales (F-COPES) (the main outcome measure) 33 and the Psychosocial Adjustment To Illness Scale-Self Report (PAIS-SR). 34 Family Coping: Th e F-COPES identifi es family problem-solving and behavioral strategies used by families in crisis situations and has been used to assess the impact of interventions in the family. 17,35 F-COPES focuses on two levels of interaction: from the individual to the family system (the way in which the family manages crises and problems internally) and from the family to the social environment (the way in which the family manages problems outside its boundaries). 17 . Th e acquiring social support subscale measures a family's ability to acquire support from friends, relatives, neighbors, and extended family. Th e reframing subscale assesses the family's ability to redefine stressful events to help them be manageable by the family. Th e seeking spiritual support subscale examines the family's ability to acquire spiritual support. Th e mobilizing family to acquire and accept help subscale measures the family's ability to seek community resources and accept help from others. The passive appraisal subscale assesses the family's ability to accept diffi cult issues, minimizing reactivity. F-COPES describes a variety of coping behaviors, and items are rated using a 5-point Likert scale, ranging from "strongly disagree" (1) to "strongly agree" (5). Th e total score ranges from 29 to 145, with higher scores indicating more positive coping and problem-solving strategies.
Psychosocial Adjustment: Th e PAIS-SR has been used to assess the impact of rehabilitation programs on psychosocial adjustment to the disease. 36 . Th e participant selects the statement that best describes his/her personal experience. Th e total score ranges from 0 to 138, and higher scores indicate poorer adjustment.
Th e following outcome measures were collected only from patients .
Exercise Tolerance: Exercise tolerance was measured using the 6-min walk test. Th e measurement properties of this test are well established in COPD. 39 Two tests were performed according to standardized guidelines. 40 Functional Balance: Th e Timed Up-and-Go test was used to assess functional balance. 41 Patients were instructed to walk quickly, but as safely as possible. Two tests were performed and the best performance considered.
Muscle Strength: Knee extensors strength of the dominant limb was assessed using the 10-repetition maximum (10-RM) with ankle weights. 42 In patients with COPD, the completion of 1-RM testing may not be safe 43 ; thus, multiple RM, such as 10-RM, were used.

Data Analysis
Using F-COPES data from a previous pilot study (not published), two sample size estimations (for patients and family members) with 95% power at a significant level of .05 were performed. These analyses determined that a statistically signifi cant diff erence in F-COPES total score would be detected with 42 patients (partial h 2 5 0.078) and with 30 family members ( h 2 5 0.110). As PR programs have considerable dropouts, varying between 20% and 40%, 45,46 56 family dyads (28 per group) were recruited. Th ese power analyses were performed using the G*Power 3 soft ware (University Düsseldorf).
Descriptive statistics were used to describe the sample. For each measure, the normality of data was investigated with Shapiro-Wilk tests. Independent t tests for normally distributed data and Mann-Whitney U tests for ordinal/nonnormally distributed data were used to compare baseline measures between groups. x 2 tests were used for categorical data. Two-way analysis of variance with repeated measures was used to establish the signifi cant eff ects for time, group, and group 3 time interaction. Th e level of signifi cance was set at 0.05. Statistical analysis was completed with the estimation of eff ect sizes for each outcome measure to evaluate the magnitude of treatment eff ect. 47 Th e eff ect size was computed via partial h 2 , as it is the index more commonly reported for two-way analysis of variance with repeated measures. 48 Partial h 2 was interpreted as a small ( h 2 Ն 0.01), medium ( h 2 Ն 0.06), or large ( h 2 Ն 0.14) eff ect. 49 Data analyses were performed using SPSS Statistics version 20.0 (IBM).

Results
Participants' Characteristics Figure 1 shows the CONSORT fl ow diagram of the trial.
Of the 69 dyads screened for this study, 13 were excluded. Eight did not meet inclusion criteria, and fi ve declined to participate. Therefore, 56 dyads were allocated to the experimental (n 5 28) or control (n 5 28) group. Forty-two dyads completed the intervention and post-test assessments and were included in the analysis. There were no significant differences between completers and dropouts regarding any of the sociodemographic, clinical, or psychologic baseline characteristics ( P . .05).
Baseline sociodemographic characteristics of patients and family members of both groups are provided in Table 2 . No signifi cant diff erences between groups were noted in baseline characteristics, with the exception of family members' marital status ( P 5 .037).

Adherence
Patients and family members in the experimental group attended a mean of 11.1 Ϯ 0.9 psychosocial support and education sessions, achieving an overall adherence rate of 92% Ϯ 8.7%. In the control group, patients' adherence to psychosocial support and education component was 90.8% Ϯ 7.1% (mean of 10.9 Ϯ 0.9 sessions, P 5 .626). Attendance to exercise training sessions was similar in both groups, with rates of 82.1% Ϯ 15.3% and 83.4% Ϯ 12% ( P 5 .755). Figure 2 shows the results on family coping in patients and family members of the experimental and control groups. Th e magnitude of improvement in family coping in patients ( P 5 .048, h 2 5 0.091) and family members ( P 5 .004, h 2 5 0.226) of the experimental group exceeded the improvement of the control group ( Fig 2 ).

Family Coping
Aft er the intervention, patients ( P 5 .017) and family members ( P 5 .047) of both groups reported the use of more strategies of acquiring social support ( Table 3 ) . Th e coping strategies of reframing, seeking spiritual support, and mobilizing to acquire and accept help were more frequent in family members of the experimental group than in those of the control group ( P , .05, h 2 from 0.149 to 0.255) ( Table 3 ). The strategy mobilizing to acquire and accept help was also more used by patients of the experimental group than by those of the control group ( P 5 .028, h 2 5 0.117) ( Table 3 ).

Psychosocial Adjustment to Illness
Th e results of the psychosocial adjustment to illness are presented in Table 4 . Patients and family members from both the experimental and control groups experienced improvements in psychosocial adjustment to COPD ( P 5 .003 and P 5 .001), with no diff erences between groups ( P 5 .454 and P 5 .252). Family members of the experimental group had signifi cant changes in sexual relationships ( P 5 .026, h 2 5 0.151) and in psychologic distress ( P 5 .033, h 2 5 0.123) compared with family members of the control group.

Patients' Outcome Measures
Both the experimental and control groups experienced signifi cant improvements in exercise tolerance, functional  balance, knee extensors strength, and health-related quality of life aft er the intervention ( P , .001; h 2 from 0.228 to 0.622), with no differences between groups ( P . .05) ( Table 5 ).

Discussion
To our knowledge, this is the first randomized controlled trial to investigate the impact of family-based PR on patients and family members' coping strategies. Th e main fi ndings indicate that integrating the family member in PR contributed to improve the coping strategies of the family to manage the disease, with further improvement in family members' sexual functioning and psychologic distress. In addition, patients from both groups experienced signifi cant improvements in exercise toler-ance, functional balance, knee extensors strength, and health-related quality of life.
Living with COPD has been described as a psychologic distressing experience, 50 which involves different coping eff orts and aff ects relational dynamics. 10,51 Th is is explained by the incapacitating nature of the disease characterized by stable periods alternated with periods of exacerbations, which leads to family having to deal with the uncertainty of exacerbation occurrence and with specifi c demands, such as monitoring health status and adherence to treatments. Although these impacts are greater as the disease progresses, families have expressed the need for more information about the disease and strategies for its management. 5,8,[10][11][12] However, this has been poorly valued by health professionals and researchers. Th is study has contributed to the current body of knowledge by showing that a family-based PR is eff ective in enhancing the coping strategies of all of those living with COPD. Specifically, patients and family members of the family-based PR made greater use of community resources to cope with their problems (external coping). 35 Moreover, the improvement in family coping was more pronounced in family members than patients, namely in the strategies of reframing (internal coping) and seeking spiritual support (external coping). Th e ability to manage stressful events by redefi ning the event in more helpful terms and to obtain Data are presented as mean Ϯ SD. PAIS-SR 5 Psychosocial Adjustment to Illness Scale-Self Report. See Table 3 legend for expansion of other abbreviation. a Samples of the experimental and control groups were unbalanced in the vocational environment domain, and, thus, results of this domain were not analyzed, nevertheless, they have been accounted for the global score. spiritual support are frequently endorsed by families living with chronic diseases and have been associated with lower stress levels. [52][53][54] Th erefore, including the broader relational context in which COPD is experienced, is benefi cial to the family and seems to be a more integrated care model of delivering PR.
Patients and family members from both groups experienced improvements in their psychosocial adjustment to the disease. Improvements in psychosocial morbidity among patients with COPD aft er PR have been previously reported. 55 However, the present study also demonstrated that psychosocial support and education for the family contributed to improved psychologic adjustment to the disease and sexual functioning of the family member. Th ese are important results, since these family members tend to lose intimacy and caring feelings for their partner, which are replaced by feelings of duty (because of marriage vows and societal expectations), 8,56 thereby increasing their psychologic distress. 11,50 Few studies have developed and evaluated interventions involving family members of patients with COPD, 21,57,58 and only one has reported the experience of family members aft er participating in a multidisciplinary PR program. Positive results on understanding the disease, enhancing the relationship, and their coping strategies were reported up to 2 years aft er the program. 21 However, family members were invited to participate in just one session and considered it somewhat insuffi cient to their needs. Participating in psychosocial support and education interventions has been found to increase the well-being of the family in other populations, such as cancer, 30,59 schizophrenia, 60 and psychosis. 61 Th is study is innovative, as it extends these fi ndings to the COPD population.
Although a greater improvement in patients' functioning of the experimental group compared with those from the control group could be believed to be more compelling, diff erences between groups were not found. Th is was not unexpected, as similar exercise training was provided to both groups of patients, and family members from the experimental group were never directly encouraged to be facilitators of patients' functioning. Future studies should explore whether other levels of family engagement in PR aff ect patients' functioning, for example, by encouraging patient's physical activities.
Some limitations need to be acknowledged. Th e main fi ndings of this study were based on self-report instruments and may not represent actual changes in patients' or family members' behavior. Furthermore, two F-COPES subscales, namely mobilizing family to acquire and seek help ( a f 5 0.402) and passive appraisal ( a p 5 0.430), had slight low internal consistency, which may have interfered with the results. Future studies could use other self-reported instruments combined with qualitative methods. Th is randomized controlled trial was conducted with a small sample of each COPD grade; therefore, it was not possible to determine whether the severity of disease impacted on the outcome. It was also not possible to blind the outcome assessor, which could have influenced the results. Finally, long-term follow-up was not collected, which would strengthen these results. Th erefore, it is currently unknown if these eff ects were sustained. Further research with longer follow-ups and with larger samples is necessary to investigate the short-and long-term eff ects of family-based PR on each COPD grade.

Conclusions
Family-based PR benefi ts the family by improving the coping strategies and the psychosocial adjustment to illness. To contribute to integrated care toward managing COPD, PR programs should consider actively involving the family system within the care delivery.

Acknowledgments
Author contributions: A. M. is the guarantor of the study. A. M. and D. F. contributed to study conception and design and obtaining funding; D. B. contributed as a consultant and provided advice during the conception and design of the project; C. J., J. C., and R. G. contributed to data collection and analysis; and A. M. contributed to draft ing the manuscript. All authors critically revised the paper for important intellectual content.
Financial/nonfi nancial disclosures: Th e authors have reported to CHEST that no potential confl icts of interest exist with any companies/organizations whose products or services may be discussed in this article .

Role of sponsors:
Th e sponsors had no role in the design of the study, the collection and analysis of the data, or the preparation of the manuscript.
Other contributions: We thank all institutions, patients, and family members involved for their participation in this research.