The Effects of Exercise on Physical and Psychological Outcomes in Cancer Caregivers: Results From the RECHARGE Randomized Controlled Trial

The Effects of Exercise on Physical and Psychological Outcomes in Cancer Caregivers: Results From... Abstract Background Family caregivers to cancer patients are at risk for physical and psychological morbidity as a direct result of being in the caregiver role. Exercise as an intervention to support caregiver health and well-being has been tested in other caregiver populations; however, no randomized controlled trials testing the effects of an aerobic and resistance training program have been conducted where cancer caregivers only have participated. Purpose To examine the effects of a 12-week exercise program on quality of life (Short Form 36 v2), psychological outcomes, physical activity levels, and physical fitness in caregivers to cancer patients. Methods Seventy-seven caregivers were randomized to either an exercise intervention or wait-list control. Two participants in the exercise arm dropped out immediately after their baseline assessment. Outcome data for n = 77 participants were analyzed using intention-to-treat analysis with linear mixed models. Results Significant group by time interaction effects were detected for exercise on the mental health component score of quality of life, self-reported weekly exercise levels, and the six-minute walk test. Small to medium effect sizes were detected on several outcomes including physical health component of quality of life (d = 0.36), mental health component of quality of life (d = 0.74), depression (d = 0.41), anxiety (d = 0.35), physical activity levels (d = 0.80), and the six-minute walk test (d = 0.21). Conclusions The findings suggest that exercise may improve health and well-being and could be considered as an additional type of support offered to cancer caregivers. Clinical trial registration no NCT02580461. Exercise, Randomized controlled trial, Cancer caregivers, Physical health • Psychological health Background Family caregivers (hereafter referred to as caregivers) to cancer patients comprise a significant proportion of Canadian and U.S. populations. Most recent statistics indicate that there are approximately 900,000 cancer caregivers in Canada [1] and 4.6 million in the USA [2]. Current models of cancer care delivery are reliant on caregivers to provide up to 50% of the care and support required by patients as they go through treatment and into recovery [3], but this does not come without a cost. Specifically, caregivers may be overwhelmed [4], suffer from physical and psychological health problems [4, 5], and neglect their own health behaviors [6, 7]. Recognizing the importance of caregivers to care delivery and the negative impacts that may occur while in the caregiver role, supporting caregiver health and well-being has become an important health system and health policy agenda [8]. Interventions to support caregiver health and well-being have traditionally focused on psychological support through support groups or individual counselling, education, or respite [3]. In the last 15 years, researchers have also investigated the role of physical activity interventions in supporting or improving caregiver’s psychological and physical health. Many of these interventions have been done with caregivers to Alzheimer/dementia patients and have shown positive effects of physical activity on decreasing caregiver burden [9–12], improving quality of life [13], improving sleep quality [10, 11], increasing activity levels [9–12], and decreasing depression [13–15]. More recently, physical activity interventions with caregivers to cancer patients have been investigated [16–19], with some positive effects on physical and psychological outcomes, including quality of life [17], improved mood [17], and increased sleep quality [18]. The current research on physical activity interventions in cancer caregivers however is lacking in several keys areas. Of the cancer caregiver physical activity interventions published to date, two of the four [17, 18] were not randomized controlled trials and tested the effects of a yoga program. One of the cancer caregiver studies included caregivers only [17], while the other three included patients and caregivers together [16, 18, 19]. Caregiver only interventions have been shown to improve caregivers’ outlook on their caregiving situation [4]. In addition, lack of reporting of factors related to internal validity in these studies makes it difficult to determine the effectiveness of exercise on the outcomes measured [20]. Thus, there is a lack of randomized controlled trial evidence on the effects of aerobic and resistance exercise (i.e., general exercise recommendations) where cancer caregivers only participated. To address the gap in the current literature, we designed a randomized controlled trial exercise intervention for cancer caregivers [21]. This trial is registered with the national clinical trials registry NCT02580461 (clinicaltrials.gov). No changes were made to the design of the trial after the protocol was published; however, the data analysis plan changed slightly (as outlined in the statistical analysis section). The aim of the study was to examine quantitative and qualitative outcomes using a mixed methods design. The quantitative results of the 12-week structured exercise program on self-reported physical health (primary outcome), psychological well-being, physical fitness levels, and physical activity levels (secondary outcomes) in caregivers caring for adult cancer patients are reported here. The qualitative results are reported elsewhere [22]. Objectives The objectives of the study were the following: (a) to examine whether a 12-week exercise program increases the physical health component score on the Short Form 36 v2 [23] measure of quality of life in those caregivers randomized to the exercise intervention versus control condition; (b) to examine whether a 12-week exercise program increases self-reported physical activity levels, physical fitness levels (cardiorespiratory, muscular, flexibility, and balance), overall self-reported quality of life, self-reported sleep quality, and perceived social support, and decreases self-reported depression and anxiety in those caregivers randomized to the exercise intervention versus control condition; and(c) to document adherence rates to exercise in those randomized to the intervention group. Materials and Methods Trial Design and Participants We conducted a randomized controlled trial with an exercise intervention group and a wait-list control group using a 1:1 allocation ratio. Participants in the wait-list control group provided baseline and 12-week assessment data only; they did not complete any other individual or group activities and did not track their physical activity. Participants in the wait-list control group were enrolled in the exercise intervention after completing 12 weeks of control. All participants (intervention and control group) were invited to participate in a 12-week maintenance exercise program once they had completed the 12-week exercise intervention. Individuals who were providing care to an adult patient were eligible to participate. We defined caregiving as providing physical or psychological support to a loved one living with cancer. If participants asked whether they were considered a caregiver, we provided them with a list of caregiving roles as outlined by the National Cancer Institute [24] for clarification. Other eligibility criteria included: (a) not planning to move in the next 12 months; (b) willing to attend twice weekly exercise classes; (c) being free from any medical conditions that would prevent participation in an exercise program; (d) not achieving recommended levels of physical activity as outlined by the Canadian physical activity guidelines [25]; and (e) able to speak and read English. Exclusion criteria were (a) providing care to a child with cancer; (b) having a health condition that limits participation in a physical activity program (e.g., unstable angina, uncontrolled hypertension, or diabetes); (c) being a bereaved caregiver (previously caring for a cancer patient who has now died); (d) meeting Canadian physical activity guidelines for the past three months; and (e) unable to attend the group exercise classes. Participants were recruited from a previously conducted caregiver survey [26] as well as through poster advertisement at a number of hospital and community facilities. Prior to enrollment, participants were screened by the research assistant or student co-investigator to determine their eligibility. Standardized screening protocols were used to determine ability to participate in an exercise program [27]. The trial was conducted at a cancer and exercise-specific research lab (Health and Wellness lab) within a University in a large city in western Canada that provides exercise space with aerobic and resistance training equipment. Approval from the local ethics board was obtained prior to study initiation (HREBA.CC‐16‐0801). Intervention We used the Canadian physical activity guidelines for adults to determine the type and amount of exercise that participants should be targeting (i.e., the goal was to achieve 150 min of aerobic exercise per week in bouts of 10 minor more, and 2 days per week of resistance exercises for each of the major muscle groups) [25]. For this study, exercise was defined as “physical activity that is planned, structured and repetitive and that has as a final or intermediate objective the improvement or maintenance of physical fitness” [28]. The 12-week exercise intervention consisted of (a) twice weekly resistance exercise group classes at the Health and Wellness lab; (b) additional weekly independent aerobic exercise sessions to reach 150 min per week; and (c) seven group-based education sessions during the 12-week intervention. The purpose of the education sessions was to address common barriers to physical activity and to reinforce principles of behavior change. This strategy has been effectively employed in previous exercise for caregiver interventions [9, 11] and is in keeping with theoretically based behavioral interventions. Prior to randomization, a standardized fitness battery (outlined in the physical fitness outcomes section) was performed by a certified exercise physiologist. Based on the results of the fitness testing and any physical limitations to aerobic or resistance exercise noted during the assessment, the certified exercise physiologist designed an individualized resistance and aerobic exercise prescription for the participant. A written copy of the prescription was provided to each participant. The twice weekly 1-hr center-based group exercise classes focused on resistance training wherein participants completed their prescribed resistance exercises. The class was overseen by a certified personal trainer and a kinesiology student volunteer. In general, participants performed one to two sets of 10–12 repetitions of resistance exercises that incorporated the major muscle groups of the body (10–12 exercises), core, and balance exercises. Adherence to the exercise class was tracked via recording attendance at each class. For the aerobic component, participants were instructed to participate in at least 10 min of aerobic exercise of their choice, at a perceived level of exertion of at least 5–6 [29], and then to progress over the course of the intervention to achieving 150 min of moderate aerobic activity per week. Perceived level of exertion is a consistently used method to assess subjective perception of exercise intensity, has been validated against physiologic measures of exercise intensity, and is practical, affordable, and feasible to use in a variety of exercise situations. An introduction to aerobic exercise including ideas for types of aerobic exercise, goal setting, warm up and cool down, measuring perceived exertion, and tracking progress was provided in the first education session along with written materials. Adherence to the aerobic prescription was monitored via weekly self-report physical activity logs. Progression through the aerobic and resistance training programs followed a pre-set schedule devised by the certified personal trainer and then tailored to each individual participant based on their fitness level and resistance training exercise experience. In general, for resistance training, in the first 1 to 3 weeks, participants used body weight or very light weight to familiarize themselves with the proper technique. In Weeks 2–6, resistance was gradually increased to the point where participants could perform 2 sets of 10–12 repetitions of each resistance exercise. In the final 6 weeks, the goal was for participants to increase their resistance and perform 2 sets of 8–10 repetitions. Each resistance training session was tracked by the participant while they were in the class (number of sets, weight, and repetition) for each exercise and then reviewed weekly by the certified personal trainer. For the independent aerobic sessions, depending on fitness level, participants began with a minimum of 10 min per session at a perceived level of exertion between 2 and 5 [29]. Based on feedback from the certified personal training and exercise physiologist, progression occurred at a rate of no more than 10% per week for duration of activity. Intensity was slowly increased so that participants were consistently achieving a perceived level of exertion of 5–6 per session. There were seven education sessions offered to participants. The sessions occurred biweekly, immediately after the group resistance exercise class, and were taught by the certified personal trainer or the student co-investigator. The education sessions included written and verbal information about healthy behaviors (sleep hygiene, benefits of physical activity and physical activity guidelines, stress management, maintaining exercise behaviors, and self-compassion) and strategies for successful behavior change (goal setting and increasing self-efficacy). Theoretical foundations for behavior change were based on Social Cognitive Theory [30] and the Health Action Process Approach [31]. We did not record attendance at the education sessions. All staff involved in the trial (research assistant, CPTs, CEPs, and principle investigator) had previous experience running exercise trials for cancer populations through the Health and Wellness lab and were trained about the specifics of this trial during study meetings held prior to participant enrollment. Outcomes A variety of patient reported, and physical fitness outcomes were collected. All outcomes were assessed at baseline, post exercise intervention (at 12 weeks), and post maintenance (at 24 weeks). Baseline and post-intervention data are reported here. Assessments were conducted by a certified exercise physiologist blinded to treatment allocation and took approximately 1.5 hr to complete. Research participants, the research assistant, and certified personal trainers leading the exercise classes were aware of treatment allocation. Demographic data Self-reported demographic data included age, gender, marital status, level of education, employment status, and income. Caregiving information included number of months in the caregiver’s role and number of hours of care provided per week (e.g., transportation, meal preparation, medication management, side effect management, and personal care). Care recipient (patient) information was also collected, as reported by the caregiver, including diagnosis, time since diagnosis, type of treatment currently on, relationship to the caregiver, and age of the patient. Psychosocial outcomes The Pittsburgh sleep quality index, a validated self-report questionnaire [32], was used to assess sleep quality. This 19-item questionnaire has been used in previous caregiver research [33] and has consistent internal reliability (α ≥ 0.80), and constructs validity upon comparison to four other sleep rating instruments [34]. A total score of 5 or greater is indicative of poor sleep quality. The Centers for Epidemiological Studies Depression scale [35], a 20-item questionnaire, was used to measure depression. A score of 16 or greater may indicate clinical depression. The Centers for Epidemiological Studies Depression scale has good sensitivity and specificity and high internal consistency (α ≥ 0.80), has been found to detect differences between caregivers and noncaregivers [36], and has been used to measure changes in caregiver depressive symptoms after interventions [37]. Quality of life was measured using the Short Form 36 v2 [23]. This self-report instrument consists of 36 items and measures 2 domains (mental health component and physical health component) of quality of life. The Short Form 36 has been widely used in many populations, has good internal consistency (α ≥ 0.8), and has well-documented content and construct validity [38]. Normative data have been published on Canadian populations [39]. The State Trait Anxiety index [40] was used to assess anxiety. This questionnaire contains 20 items and has been used in clinical settings to diagnose anxiety, and has good internal consistency (α ≥ 0.85) and test–retest reliability [40]. Normative age group data have been published, and generally a score of 40 or greater is associated with clinical anxiety [40]. Social support was measured using the Multidimensional Scale of Perceived Social Support, which is a 12-item questionnaire with high internal consistency (α ≥ 0.90) and good construct validity [41]. Total scores of 5 or greater on this measure are indicative of good social support. Exercise behaviors Physical activity behaviors were measured using self-report and class attendance. The Godin leisure time exercise questionnaire [42] was used to determine changes to weekly level of exercise over time. This self-report three-item instrument collects information on weekly time spent in strenuous, moderate, and mild physical activity. The Godin leisure time exercise questionnaire is widely used in many populations and has shown consistent test–retest validity (correlations between r = .24 for light activity and r = .84 for vigorous activity) compared with other self-report physical activity questionnaires [43]. Self-report weekly physical activity logs (type of exercise, time participating, and perceived level of exertion) and documented attendance at twice weekly exercise classes were also recorded. Physical fitness assessments Measurements of physical fitness included body composition, cardiorespiratory fitness, musculoskeletal fitness, flexibility, and balance. All fitness measures were conducted by a certified exercise physiologist using standardized protocols as outlined by the Canadian Society of Exercise Physiology physical activity training for health [44] or the Seniors fitness test [45]. The same exercise physiologist completed all of the assessments required on each individual participant. Measurements of physical fitness were chosen based on several factors including our outcome assessors familiarity with these tests, the potential of aged participants with low physical activity levels and experience, and possible comorbidities. Weight and height were used to calculate body mass index score. Waist and hip circumferences were used to calculate hip to waist ratio. Cardiorespiratory fitness was measured using a 6-min walk test (6MWT), which is a valid functional test of aerobic fitness in individuals who may be sedentary or have comorbidities [45, 46]. The 6MWT has been found to have high retest reliability (0.95) and to have both convergent and discriminant validity when used to assess aerobic fitness in older adults [45]. Muscular strength tests included grip strength, which is widely used as an indicator of overall body strength, is predictive of future functional limitations [47], and has normative age values [44]. In addition, a 30-s sit-to-stand test was used to assess lower body muscular endurance [45]. Flexibility was assessed using the sit-and-reach test [48, 49]. Balance was measured using a one-leg stance test [44]. Sample size The sample size calculation, as described in the published protocol [21], was based on the physical health component score of the Short Form 36 as the primary outcome variable. As changes to Short Form 36 scores after exercise interventions have ranged between 3.0 to 17.75 points, and standard deviations have ranged from 3 to 26 [13, 33, 50–54], we chose a change of 10 points and a standard deviation of 15 for the sample size calculation. A difference in means sample size calculation was used [55]. The sample size from this calculation, which sets power to detect a difference between groups and over time on the physical health component score of the Short Form 36 at 0.80, was 72 participants (i.e., 36 in each group). Randomization After written informed consent was obtained and baseline testing completed, participants were randomized. To protect against bias, participants were randomly assigned to one of the two different intervention arms (exercise intervention or control) using Research Randomizer [56], which is a web-based randomization generator. To avoid imbalance in the number of participants assigned to each group [57], block randomization was used and included 10 sets of 8 nonunique numbers per set. The student co-investigator generated the randomization assignment and prepared individual opaque randomization envelopes prior to any participants being enrolled. The contents of the randomization envelopes were known only to the student co-investigator. The research assistant accessed a new randomization envelope after the baseline assessment for each participant and then contacted the participant to inform them of their allocation. Statistical Methods Statistical analyses of data were performed using IBM Statistical Package for Social Sciences software version 22.0. An intention–to-treat analysis was carried out using linear mixed models (LMMs), so that all participants who provided baseline data could be included in the analysis. As the data set contained repeated-measures, linear mixed modeling is considered a more robust test of significance compared with repeated-measures ANOVA [58, 59] for several reasons. Missing data, a common occurrence in repeated measures, do not need to be discounted. Also, the assumption of sphericity (equal co-variances among groups) does not need to be met, variables measured at different levels (e.g., multilevel sampling) can be incorporated, and the assumption of independence is not violated by using LMMs [59]. Descriptive statistics were used to characterize the data and examine differences between baseline characteristics of the intervention and control groups. The t-test, chi-squared test, or Fisher’s Exact Test was used to examine group differences at baseline. Descriptive statistics were used to characterize adherence rates to the exercise intervention as the mean class attendance over 12 weeks, as well as mean level of aerobic, resistance, and total minutes of exercise per week. Fitness testing, including 6MWT, 30-s sit to stand, trunk forward flexion sit and reach, and balance, was characterized using measures of central tendency (mean and standard deviation) at baseline and Time 1. To determine group by time interaction effects of the 12-week exercise intervention on the Short Form 36 physical health component score, we used linear mixed modeling (primary outcome). The fixed effects were the intercept, group, time, and group by time interaction. Time was also set as a repeated effect. We tested autoregressive and unstructured covariance structures and chose the autoregressive covariance as the −2 restricted log likelihood was smaller. As several participants’ caregiving situations changed throughout the course of the trial (e.g., four participants’ care recipient died, three participants’ care recipient had a recurrence, and two care recipients completed treatment), we controlled for caregiving hours per week by including this as a covariate in our model. All assumptions of covariance were met. The estimated marginal mean change in physical component score for exercise and control group was calculated using LMM. Cohen’s d effect sizes were calculated using the unstandardized mean change scores (12 weeks – baseline) and the pooled standard deviations [60]. To determine group by time interaction effects of the 12-week exercise intervention on the secondary outcomes of Short Form 36 mental health component score, Pittsburgh sleep quality index, Centers for Epidemiological Studies Depression scale, Multidimensional scale of perceived social support, State trait anxiety inventory, total Godin leisure time exercise questionnaire score, fitness measures, and BMI, we used the same linear mixed modeling procedure. For the group by time interaction effect of exercise on the Godin leisure time exercise questionnaire, the covariate caregiving hours were not included in the LMM as the assumption of homogeneity of regression slopes was violated. Maas and Hox [61] suggested that 30 data points at 2 different levels provide reasonable unbiased assessment of regression coefficients and variance; therefore, we determined that our sample size was adequate to run this secondary analysis. Results Recruitment for the trial started in May 2015 and was ongoing until February 2016. During that time, a total of 230 participants were assessed for eligibility, and 153 participants were excluded (n = 4 could not attend the location for the classes, n = 21 were already exercising at recommended levels, n = 120 were not interested in an exercise program, and n = 8 were too overwhelmed with their caregiving responsibilities). A Consolidated Standards of Reporting Trials [62] diagram of participant flow through the trial is depicted in Fig. 1. A total of 77 eligible participants were enrolled in the study and were randomized to either the exercise intervention or control group. Two participants who were randomized to the exercise intervention group dropped out immediately after their baseline assessment (e.g., did not start the intervention) because of changes to their care recipient status. Of the 39 participants who were in the wait-list control group, 6 dropped out of the study after completing their 12-week assessment and did not start the intervention. The 12-week exercise intervention started in June 2015 and continued until June 2016 at which point all participants had been enrolled and completed their treatment allocation. We used a rolling recruitment and enrollment strategy whereby after four participants had been randomized to the exercise group, we started the group-based exercises classes. As more participants were recruited and enrolled, they joined the existing classes. In total, 14 groups were run sequentially, with each group having between 8 and 10 participants. We ran either daytime or evening groups to accommodate caregivers’ schedules. Once enrolled in a group, the caregivers remained in the same group. All participants followed the same protocol and were taught by one of the two certified personal trainers. All participants progressed through the exercise program as anticipated. Most participants chose walking as their aerobic exercise. Across 12 weeks, 51% of participants were meeting the physical activity guidelines of 150 min of moderate to vigorous aerobic activity per week and two sessions per week of resistance training. Fig. 1. View largeDownload slide Consolidated Standards of Reporting Trials (CONSORT) [62] diagram of flow of participants through the RECHARGE trial. ITT intention to treat; LMM linear mixed models. Fig. 1. View largeDownload slide Consolidated Standards of Reporting Trials (CONSORT) [62] diagram of flow of participants through the RECHARGE trial. ITT intention to treat; LMM linear mixed models. Baseline Data The majority of participants were female (61%), the average age was 53 years, and participants generally had a high socioeconomic status. Most participants were either retired or not working outside of the home (41.6%), and were providing care to their spouse (79.2%). Caregivers were providing care to patients with a variety of cancer types, most commonly breast (22.1%), prostate (19.5%), colorectal (11.7%), and hematological malignancies (16.9%). Most patients (87%) were on some type of treatment for their cancer. Time in the caregiver role was on average 30 months, and average time spent providing care per week was 19 hr. Baseline analysis indicated no differences between the exercise and control group on either demographic or caregiving variables; therefore, data presented in Table 1 are combined for all participants. The only statistical difference found between groups was that more caregivers in the control group were living with their care recipient than in the exercise intervention group (p = .034). Table 1 Demographic and caregiving variables for cancer family caregivers in RECHARGE study, N = 77 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 SD standard deviation. aWhere assumptions of chi-squared test were violated, Fisher’s Exact Test was used. View Large Table 1 Demographic and caregiving variables for cancer family caregivers in RECHARGE study, N = 77 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 SD standard deviation. aWhere assumptions of chi-squared test were violated, Fisher’s Exact Test was used. View Large Baseline measures of physical and psychological health, quality of life, physical activity levels, and fitness testing for the exercise intervention and control group are shown in Table 2. There were no statistically significant differences between groups on these baseline measures. Participants reported high levels of state and trait anxiety (M = 40.20 and 42.5, respectively), depression (M = 18.04), and poor sleep quality (M = 8.34). Participants rated their perceived social support as moderate (M = 5.04). The physical health component of quality of life (M = 52.58) was, on average, higher in our participants than standardized norms for the Canadian population; however, the mental health component of quality of life (M = 39.64) was lower than standardized norms [39]. Total number of health conditions (M = 1.4) was comparable to health conditions of similar aged population [63]. Self-report of physical activity (M = 23.58) showed that most participants were considered moderately active [64]. Anthropomorphic and physical fitness measures showed the majority of participants were overweight (mean BMI = 8.26), however had aerobic fitness (mean 6MWT = 639.59 m) within healthy adult range [65]. Measures of strength showed that the majority of participants had fair grip strength (M = 74.46) [66]. No statistically significant differences were found between groups on measures of physical fitness. Table 2 Baseline physical and psychological health, quality of life, PA levels, and fitness of cancer caregivers in RECHARGE trial (N = 77) Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 SD standard deviation; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; BMI body mass index; PA physical activity; QOL quality of life. a PA levels using Godin Leisure Time Exercise questionnaire. Strenuous activity = times per week participants engaged in strenuous activity for 15 min. Moderate activity = times per week participants engaged in moderate activity for 15 min. Mild activity = times per week participants engaged in mild activity for 15 min. Total weekly score = (9 * strenuous) + (5 * moderate) + (3 * mild). View Large Table 2 Baseline physical and psychological health, quality of life, PA levels, and fitness of cancer caregivers in RECHARGE trial (N = 77) Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 SD standard deviation; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; BMI body mass index; PA physical activity; QOL quality of life. a PA levels using Godin Leisure Time Exercise questionnaire. Strenuous activity = times per week participants engaged in strenuous activity for 15 min. Moderate activity = times per week participants engaged in moderate activity for 15 min. Mild activity = times per week participants engaged in mild activity for 15 min. Total weekly score = (9 * strenuous) + (5 * moderate) + (3 * mild). View Large Primary Outcome LMM analysis revealed that there was no significant group by time interaction effect on the physical health component of quality of life across 12 weeks (F [1, 57.35] = 1.04; p = .31) after controlling for the covariate care hours per week. The covariate was not significantly related to the physical component score. In addition, no significant main effects of time or group assignment were found (Table 3). Estimated marginal means analysis showed that the exercise group physical component score increased by 0.30 over 12 weeks, while the control group physical component score decreased by −1.24 (Table 4). Despite the nonsignificant effects, a small effect size for the exercise intervention on physical health component score (d = .36) was found (Table 5). Table 3 Linear mixed model analysis assessing outcome for intention-to-treat exercise and control group from baseline to 12 weeks Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aCovariate care hours/week was not included as preliminary analysis revealed that the assumption of homogeneity of regression slopes was violated. *Statistically significant at p < .05 level. View Large Table 3 Linear mixed model analysis assessing outcome for intention-to-treat exercise and control group from baseline to 12 weeks Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aCovariate care hours/week was not included as preliminary analysis revealed that the assumption of homogeneity of regression slopes was violated. *Statistically significant at p < .05 level. View Large Table 4 Estimated marginal group mean for outcome measures at baseline and 12 weeks Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aEstimated marginal mean evaluated with covariate care hours = 14.95. View Large Table 4 Estimated marginal group mean for outcome measures at baseline and 12 weeks Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aEstimated marginal mean evaluated with covariate care hours = 14.95. View Large Table 5 Unadjusted means between exercise and control group for baseline and 12 weeks, and effect sizes for outcome variables Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 SD standard deviation; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; 6MWT 6-min walk test. aCohen d formula used = difference between two mean changes between groups (T2–T1) divided by pooled SDs. View Large Table 5 Unadjusted means between exercise and control group for baseline and 12 weeks, and effect sizes for outcome variables Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 SD standard deviation; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; 6MWT 6-min walk test. aCohen d formula used = difference between two mean changes between groups (T2–T1) divided by pooled SDs. View Large Secondary Outcome Results Results of the LMM analysis of secondary outcomes including Short Form 36 mental health component score, Pittsburgh sleep quality index, Centers for Epidemiological Studies Depression scale, State trait anxiety inventory, Godin leisure time exercise questionnaire, 6MWT, 30-s sit-to-stand test, and grip strength are shown in Table 4. Quality of life Short Form 36 mental health component score There was a statistically significant group by time interaction effect for the Short Form 36 mental health component score (F [1, 61.52] = 7.02; p = .01) after controlling for the effect of care hours per week (Fig. 2), which indicated that the group effect varied with time. Simple effects testing indicated that for the exercise group, there was a statistically significant increase in their mental component scores across the 12 weeks (F [1, 62.23] = 29.27; p < .001). The mental health component score increased in the exercise group by 8.34 points and in the control group by 2.66 points based on estimated marginal means analysis (Table 5). The effect size of the intervention on the mental health component score was medium (d = 0.74; Table 5). Fig. 2. View largeDownload slide Estimated marginal mean SF-36 MCS score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. SF-36 MCS Short Form 36 v2 mental component score. Fig. 2. View largeDownload slide Estimated marginal mean SF-36 MCS score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. SF-36 MCS Short Form 36 v2 mental component score. Psychological health Depression Results of the LMM analysis for depression showed no significant group by time interaction effects; however, main effects for time were detected (F [1, 60.02] = 12.12; p = .001). Regardless of group assignment, both groups showed improved depression scores (i.e., less depressed) over 12 weeks. The covariate care hours per week was not significantly related to depression scores. In addition, the intervention effect to decrease depression scores (i.e., less depressed) was small (d = −0.41; Table 5). Anxiety Main effects for time were found for both state anxiety (F [1, 64.71] = 6.46; p = .01) and trait anxiety (F [1, 60.50] = 12.77; p = .001). Regardless of group assignment, both groups had decreased levels of anxiety over 12 weeks (Table 5). No group by time interactions were detected. Effect sizes for the exercise intervention on anxiety scores were small (Table 5). Sleep quality Sleep quality was significantly related to the covariate care hours (F [1, 97.58] = 4.07; p = .05). No group by time interaction effects or main interactions were found for sleep quality. Small effect sizes for the intervention on sleep quality were detected (d = −0.35; Table 5). Social support No differences on social support were detected between the groups or over time. Small effect sizes for exercise on social support were demonstrated (d = 0.22) Physical activity levels and physical fitness There were significant group by time interaction effects on the Godin leisure time exercise questionnaire scores (F [1, 69.26] = 6.84; p = .011) and the 6MWT (F [1, 56.03] = 6.90; p = .01; Figs. 3 and 4), indicating that the group effect varied with time. For the 6MWT, simple effect testing revealed that the exercise group and the control group both had a statistically significant increase in the number of meters they walked comparing baseline with 12 weeks. At 12 weeks, the difference between the exercise group and the control group in the number of meters they could walk in 6 min approached statistical significance (F [1, 76.04] = 3.93; p = .051). This suggests that the exercise group had greater improvements in walk test scores over 12 weeks compared with the control group. For the Godin leisure time exercise questionnaire scores, simple effects testing indicated that for the exercise group, there was a statistically significant increase in their weekly exercise levels across the 12 weeks (F [1, 70.96] = 10.78; p = .002). In addition, at 12 weeks, there was a statistically significant difference between the exercise and the control group with respect to weekly exercise levels (F [1, 129.15] = 7.37; p = .008). Analysis using estimated marginal means showed the exercise group’s Godin leisure time exercise questionnaire score increased by 13.48, while the control group’s score decreased by 1.27 (Table 5). Effect sizes for the exercise intervention on the Godin leisure time exercise questionnaire scores were large (d = 0.80) and for the 6MWT were small (d = 0.21; Table 5). Fig. 3. View largeDownload slide Estimated marginal mean Godin Leisure Time Exercise Questionnaire (GLTEQ) score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Fig. 3. View largeDownload slide Estimated marginal mean Godin Leisure Time Exercise Questionnaire (GLTEQ) score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Fig. 4. View largeDownload slide Estimated marginal mean 6-min walk test (6MWT) distance across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Fig. 4. View largeDownload slide Estimated marginal mean 6-min walk test (6MWT) distance across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Other measures of physical fitness showed no group by time interaction effects. There was a main effect of time effect on the sit-to-stand test (F [1, 58.69] = 17.37; p = .001), with both groups showing improvements over time. Adherence to Intervention Physical activity levels (as determined from physical activity logs and class attendance) throughout the intervention showed the majority of participants achieved recommended levels of weekly aerobic activity. In addition, the majority were participating in resistance training exercises at least twice per week. The mean attendance (calculated as total number of classes attended out of a possible 24) across 12 weeks was 69%. Protocol adherence was monitored by the student co-investigator regularly attending exercise classes, through biweekly study meetings and through notification to the study team of potential protocol deviations. Harms There were no serious adverse events recorded for this trial. One participant had an exacerbation of a hiatus hernia. However, this participant was able to continue in the exercise class after clearance from his surgeon and modification to his strength training routine. Another participant had an exacerbation of a previous hip strain. After consultation with a physiotherapist, this participant decided to withdraw from the study. Discussion We conducted a randomized controlled trial intervention with cancer caregivers to determine whether exercise would improve self-reported quality of life, psychological health, measures of physical fitness, and physical activity levels. As there is limited research evaluating this type of intervention (i.e., a group-based structured aerobic and resistance training program) in cancer caregivers to date, our findings add some important new information to the current literature. For our primary outcome, the physical health component of quality of life, we found no differences between the exercise and the control group after 12 weeks. Several factors could account for this null finding. The physical health component of quality of life score is a composite of measures of physical functioning, impediments to one’s role based on poor health, bodily pain, and overall general health. The participants in our study had baseline physical health component scores that were higher than the expected population norms [39], had few medical morbidities, and had good overall health, meaning that further improvements to their physical component score would be difficult to detect (i.e., a ceiling effect). Despite the lack of statistically significant difference between the groups across 12 weeks, a small effect size of the intervention on the physical health component score was demonstrated. Northouse et al. [4] conducted a meta-analysis in 2010 and determined that caregiver interventions ranged between small and medium effect sizes. They argued that small effect sizes are in keeping with many interventions (physical activity and other types of support) tested in cancer patients and other chronic illness populations, and therefore, this effect should not be discounted as unimportant. Northouse et al. [4] also found that caregiver interventions delivered over longer periods of time generally showed greater effects. As we used perceived exertion versus objective measures (i.e., heart rate monitor) to measure exercise intensity, it could be that participants were not achieving high enough intensity levels of exercise to improve the physical health component. Finally, a larger sample size might have provided greater power to detect effects of exercise on the physical component score of quality of life. Contrary to the null findings on the physical health component score of quality of life, the exercise intervention improved caregiver’s mental health component of quality of life score. This is a noteworthy finding, given that the caregivers in our study had low baseline mental health component scores compared with population norms. Our findings are similar to previous research on physical activity interventions in various caregiver populations, where changes to mental health component scores have ranged between 2 and 8 points [13, 17–19]. Based on estimated marginal means, the difference in mental health component scores between the exercise and intervention group was approximately 6 points which is beyond the suggested threshold for minimally important difference of one half a standard deviation [67] (4.1 points) for the MCS score [39]. The medium effect size of the intervention on the mental health component scores is also an important indicator that the caregivers who participated in regular physical activity noticed meaningful improvements in the mental health component of their quality of life. Further substantiating this effect, our follow-up interviews with caregivers [22] highlighted that caregivers perceived these improvements to their mental well-being as an important benefit of physical activity. The caregivers in our study had poor psychological health at baseline as evidenced by high levels of depression and anxiety, and poor sleep quality. This adds further evidence to the growing body of literature attesting to the negative impacts of caregiving on psychological health. Irrespective of group assignment, we found significant time effects on measure of depression, state anxiety, and trait anxiety. In other words, time alone seemed to improve caregiver’s anxiety and depression. Badger et al. [16] reported similar “time-only” effects on depression in their exercise intervention with caregivers and cancer patients. It could be that simply being enrolled in an intervention for caregivers provided a sense of being supported and taking control, which in turn improved aspects related to mood. From the interviews we conducted with caregivers after they completed the exercise intervention [22], the recognition of caregivers needing help and their ability to take control were important components of improving their overall feelings of well-being. The intervention was effective at improving physical activity levels and physical fitness measures. This was demonstrated through self-reported physical activity levels and measures of aerobic physical fitness. Similar improvements in physical activity levels and fitness measures have been reported in other caregiver exercise trials [9, 11, 13]. Although the improvements to aerobic fitness likely do not represent a clinically meaning difference, we believe this is an important outcome given that there was a small effect on aerobic fitness after just 12 weeks of exercise. In addition to improvements in physical activity levels and physical fitness, the caregivers in our study also achieved recommended levels of physical activity on a weekly basis and attended classes regularly. These findings demonstrate that caregivers are able to participate in regular physical activity with the support of a group-based exercise class, and the benefits of this regular activity include improvements to overall level of weekly physical activity and aerobic fitness. This is an important finding considering the reported barriers caregivers face in participating in regular physical activity [6, 7] and their increased risk for morbidities that could be mitigated through regular exercise. Improvements to physical activity levels could also stem from the social support gained from the group-based format of the class. It is well documented that social support is an important component of physical activity behavior [68]. In our qualitative interviews with caregivers, the group environment was consistently noted as an important component of the intervention [22]. Despite improvements to physical activity levels and aerobic fitness, there were no improvements to strength after 12 weeks of regular resistance training. This null finding could be explained by the following factors. As the participants had average strength at baseline, gains to their strength would be slower than if their strength at baseline would have been low [66]. In addition, many of the participants had no history of performing the strength training exercises, which meant that during the first 3–4 weeks of the program very little resistance was used. Gains in muscle strength are dependent on the overload principle whereby muscles need to be exercised at a workload that is greater than normal [66]. The portion of the time that participants spent overloading their muscles (i.e., 7–8 weeks) was likely not long enough to see strength gains. Limitations To our knowledge, this is the first randomized controlled trial in cancer caregivers to test the effects of a group-based aerobic and resistance training exercise program on physical and psychological outcomes. Although our findings add new information to the literature, some limitations must be considered. Common to most behavioral interventions, participants were not blinded to their treatment allocation that may lead to performance bias. To minimize other sources of bias, including detection bias, we blinded the fitness assessors. Our sample of caregivers had high socioeconomic status, and the participants were generally physically healthy. This may not be representative of the population of caregivers and therefore limits the generalizability of our findings. We used self-report physical activity logs to collect weekly physical activity in addition to class attendance. Future research on exercise interventions in caregivers could include accelerometer data to provide more accurate collection of physical activity levels. Only 56% of the participants returned 12 weeks of completed physical activity logs, while the remaining participants only partially completed their logs. Thus, our estimation of weekly physical activity levels is conceivably an overestimation, as those participants who completed all the logs were also likely to be highly adherent to the intervention. The participants who dropped out of the study did so predominantly because of being overwhelmed with their caregiving responsibilities. Finally, the broad age range of participants might have made it more difficult to detect an effect on some of the outcomes including the primary outcomes and social support. A limitation of this study was our inability to adequately capture the extent of the caregiving experience with respect to the nature and intensity of caregiving. This stems from lack of validated instruments to comprehensively measure this construct, our efforts to try to balance participant burden in the number of questionnaires they were completing, and the recognition that the caregiving experience is largely dependent on many personal, social, and situational factors [69]. We measured caregiving hours per week as a crude estimate of total caregiving responsibilities. We then controlled for caregiving hours per week as a crude estimate of changes that may have occurred to the caregiving status throughout the study. Other factors related to the caregiving situation, such as stage of illness and health of the care recipient or functional status (e.g., Eastern Cooperative Oncology Group performance status), an inventory of caregiving responsibilities, and other dependents, might also be important factors to measure and control for in future caregiver studies. In addition, although we controlled for caregiving hours per week, it should be noted that the caregivers in our study could have been less overwhelmed than the general population of caregivers. It should be noted that the results from this trial are specific to the design of the intervention, which included a group-based exercise and education program. It is difficult to attribute the results from the trial to exercise alone. We included a self-reported social support outcome measure in an attempt to try to examine whether there was a social component of the group-based program. We did not find any changes to social support, however, other measures of the group effect (e.g., social cohesion, role modelling, and sense of belonging) may also have been contributory. Implications for Clinical Practice and Future Research Directions Findings from this study suggest that a group-based exercise program for cancer caregivers may be effective at improving aspects of quality of life, physical fitness, and physical activity levels. The addition of physical activity programs to the range of supportive services offered to caregivers should be considered as an important additional mechanism in supporting health and well-being in this invaluable and vulnerable group. Our qualitative analysis of the impact of this intervention [22] also highlighted the perceived value of exercise as an intervention for caregivers and the need for more choice in the type of supports that are currently offered to caregivers. The time-only effects that were observed in many of the outcomes suggest that recognition of and attention to caregivers was an important component of the study (i.e., Hawthorne effect). Indeed, the caregivers we interviewed all thanked the student co-investigator for designing and conducting a study to support caregivers and suggested that recognition of caregiving by health care providers and researchers is an important supportive mechanism. Building on this work and previous research evaluating exercise in caregivers, the following future research directions could be considered. The baseline physical health status of our participants was better than we expected. Health status and functional level at baseline may be an important indicator of the type of participant who would self-select to be in an exercise study. Those caregivers with poor overall health, multiple morbidities, or poor functional level may not volunteer to be in an exercise study, which paradoxically is the type of participant that such an intervention should be targeting. Although we had the capacity to accommodate those caregivers who were older and in poorer health, we did not generally attract these types of caregivers to our study. Recruitment efforts to target caregivers in poorer health, those with lower socioeconomic status, or minority populations, and exercise interventions designed to be flexible to accommodate lower baseline functional levels would be important considerations for future research. We found the mental health component of quality of life improved; however, other measures of mental health such as depression and anxiety did not improve significantly. Future research could evaluate the impact of psychological support and exercise combined (e.g., walk and talk, or combined exercise and support group). Given the high psychological morbidity in our sample, exercise in addition to psychological support may be more effective at improving overall mental well-being. In the meta-analysis by Northouse et al. [4], caregivers interventions that had a combination of the types of support offered showed greater effect sizes than single types of support. In addition, our qualitative findings [22] suggested that other indicators of caregiver well-being, such as positivity, life satisfaction, and feelings of control, may be important outcomes to measure to determine the impact of exercise on caregivers’ psychological well-being. Determination of sustainable exercise programming, such as at community centers or cancer-specific agencies, would also be an important next step. Because this work was part of a doctoral research program that had limited funding and time constraints, the length of our intervention was likely too short. To provide more robust estimation of the effects of exercise and to address questions about maintenance of exercise behaviors, long-term programs should be the goal. Researchers partnering with community agencies where exercise programs for cancer patients are already being offered, or partnering with caregivers using participatory action research, would be a logical next step. Finally, we intentionally designed a group-based exercise and education program for caregivers based on theories of behavior change and the well-documented effect of social support in exercise adherence. Given caregivers overwhelming time demands and potential inability to leave their care recipient, it would be important in future work to examine home-based versus group-based exercise program to determine whether a balance between feasibility and outcomes could be achieved. Acknowledgements This study was funded by The Canadian Institutes of Health Research (CIHR) Doctoral research award, The Izaak Walton Killam scholarship, and the Psychosocial Oncology Research Training program. Compliance with Ethical Standards Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards Authors Colleen A. Cuthbert, Kathryn King-Shier,Dean Ruether, Dianne Tapp, Kathryn Wytsma-Fisher, Tak S. Fung, and S. Nicole Culos-Reed declare that they have no conflict of interest. All procedures, including the informed consent process, were conducted in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Ethical Approval This research involved human participants. Ethics approval from the local governing ethics review board was obtained prior to any contact with participants. Informed Consent All participants signed an informed consent form. References 1. Sinha M . Spotlight on Canadians, Results from the General Social Survey: Portrait of Caregivers . Statistics Canada; 2012 . Available at http://www.statcan.gc.ca/pub/89-652-x/89-652-x2013001-eng.pdf. Accessibility verified July 2016 . 2. National Alliance for Caregiving/AARP . 2015 Report: Caregiving in the US . Available at http://www.caregiving.org/caregiving2015/. Accessibility verified July 25, 2016. 3. Northouse L , Williams AL , Given B , McCorkle R . Psychosocial care for family caregivers of patients with cancer . J Clin Oncol . 2012 ; 30 ( 11 ): 1227 – 1234 . Google Scholar CrossRef Search ADS PubMed 4. Northouse LL , Katapodi MC , Song L , Zhang L , Mood DW . Interventions with family caregivers of cancer patients: Meta-analysis of randomized trials . CA Cancer J Clin . 2010 ; 60 ( 5 ): 317 – 339 . Google Scholar PubMed 5. Stenberg U , Ruland CM , Miaskowski C . Review of the literature on the effects of caring for a patient with cancer . Psychooncology . 2010 ; 19 ( 10 ): 1013 – 1025 . Google Scholar CrossRef Search ADS PubMed 6. Beesley VL , Price MA , Webb PM ; Australian Ovarian Cancer Study Group; Australian Ovarian Cancer Study—Quality of Life Study Investigators . Loss of lifestyle: health behaviour and weight changes after becoming a caregiver of a family member diagnosed with ovarian cancer . Support Care Cancer . 2011 ; 19 ( 12 ): 1949 – 1956 . Google Scholar CrossRef Search ADS PubMed 7. Rha SY , Park , Y , Song SK , Lee CE , Lee J . Caregiver burden and health-promoting behaviors among the family caregivers of cancer patients . Eur J Oncol Nurs . 2015 ; 19 : 174 – 181 . Google Scholar CrossRef Search ADS PubMed 8. Kent EE , Rowland JH , Northouse L et al. Caring for caregivers and patients: Research and clinical priorities for informal cancer caregiving . Cancer . 2016 ; 122 ( 13 ): 1987 – 1995 . Google Scholar CrossRef Search ADS PubMed 9. Connell CM , Janevic MR . Effects of a telephone-based exercise intervention for dementia caregiving wives: A randomized controlled trial . J Appl Gerontol . 2009 ; 28 ( 2 ): 171 – 194 . Google Scholar CrossRef Search ADS PubMed 10. Hirano A , Suzuki Y , Kuzuya M , Onishi J , Ban N , Umegaki H . Influence of regular exercise on subjective sense of burden and physical symptoms in community-dwelling caregivers of dementia patients: A randomized controlled trial . Arch Gerontol Geriatr . 2011 ; 53 ( 2 ): e158e163 . Google Scholar CrossRef Search ADS 11. King AC , Baumann K , O’Sullivan P , Wilcox S , Castro C . Effects of moderate-intensity exercise on physiological, behavioral, and emotional responses to family caregiving: A randomized controlled trial . J Gerontol A Biol Sci Med Sci . 2002 ; 57 ( 1 ): M26 – M36 . Google Scholar CrossRef Search ADS PubMed 12. Lowery D , Cerga-Pashoja A , Iliffe S et al. The effect of exercise on behavioural and psychological symptoms of dementia: The EVIDEM-E randomised controlled clinical trial . Int J Geriatr Psychiatry . 2014 ; 29 ( 8 ): 819 – 827 . Google Scholar CrossRef Search ADS PubMed 13. Hill K , Smith R , Fearn M , Rydberg M , Oliphant R . Physical and psychological outcomes of a supported physical activity program for older carers . J Aging Phys Act . 2007 ; 15 ( 3 ): 257 – 271 . Google Scholar CrossRef Search ADS PubMed 14. Danuculov MAD , Kozasa EH , Ribas KT , Galduroz JCF , Garcia MC , Verreschi ITN . A yoga and compassion meditation program reduce stress in familial caregivers of Alzheimer’s disease patients . eCAM . 2013 ; doi: 10.1155/2013/513149 . 15. Waelde LC , Thompson L , Gallagher-Thompson D . A pilot study of a yoga and meditation intervention for dementia caregiver stress . J Clin Psychol . 2004 ; 60 ( 6 ): 677 – 687 . Google Scholar CrossRef Search ADS PubMed 16. Badger T , Segrin C , Dorros SM , Meek P , Lopez AM . Depression and anxiety in women with breast cancer and their partners . Nurs Res . 2007 ; 56 ( 1 ): 44 – 53 . Google Scholar CrossRef Search ADS PubMed 17. Martin AC , Keats MR . The impact of yoga on quality of life and psychological distress in caregivers for patients with cancer . Oncol Nurs Forum . 2014 ; 41 ( 3 ): 257 – 264 . Google Scholar CrossRef Search ADS PubMed 18. Milbury K , Mallaiah S , Lopez G et al. Vivekananda yoga program for patients with advanced lung cancer and their family caregivers . Integr Cancer Ther . 2015 ; 14 ( 5 ): 446 – 451 . Google Scholar CrossRef Search ADS PubMed 19. Winters-Stone KM , Lyons KS , Dobek J et al. Benefits of partnered strength training for prostate cancer survivors and spouses: Results from a randomized controlled trial of the Exercising Together project . J Cancer Surviv . 2016 ; 10 ( 4 ): 633 – 644 . Google Scholar CrossRef Search ADS PubMed 20. Cuthbert CA , King-Shier K , Ruether D , Tapp DM , Culos-Reed SN . What is the state of the science on physical activity interventions for family caregivers? A systematic review and RE-AIM evaluation . J Phys Act Health . 2017 ; 14 ( 7 ): 578 – 595 . Google Scholar CrossRef Search ADS PubMed 21. Cuthbert CA , King-Shier K , Tapp DM , Ruether D , Jackson C , Culos-Reed SN . Renewing caregiver health and wellbeing through exercise (RECHARGE): A randomized controlled trial . Contemp Clin Trials . 2016 ; 50 : 273 – 283 . Google Scholar CrossRef Search ADS PubMed 22. Cuthbert CA , Culos-Reed SN , King-Shier K , Ruether D , Bischoff MB , Tapp DM . Creating an upward spiral: A qualitative study of caregivers’ experience of participating in a structured physical activity program . Eur J Cancer Care . 2017 ;26(6): e12684 . 23. Ware JE Jr , Sherbourne CD . The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection . Med Care . 1992 ; 30 ( 6 ): 473 – 483 . Google Scholar CrossRef Search ADS PubMed 24. National Cancer Institute . Coping with cancer for family and friends . Available at www.cancer.gov/about-cancer/coping/family-friends. Accessibility verified July 2016 . 25. Canadian Society for Exercise Physiology . Canadian physical activity guidelines 2012 . Available at www.csep.ca. Accessibility verified July 28, 2015 . http://dx.doi.org/10.1249/MSS.0b013e318213fefb. 26. Cuthbert CA , King-Shier K , Tapp D , Ruether D , Culos-Reed SN . Exploring gender differences in self-reported physical activity and health among older caregivers . Oncol Nurs Forum . 2017 ; 44 ( 4 ): 435 – 445 . Google Scholar CrossRef Search ADS PubMed 27. Warburton DER , Jamnik VK , Bredin SSD , Gledhill N . On behalf of the PAR-Q+ Collaboration . The physical activity readiness questionnaire for everyone (PARQ+) and electronic physical activity readiness medical examination (ePARmed-X+) . Health Fitness J Canada . 2011 ; 4 ( 2 ): 3 – 23 . 28. Garber CE , Blissmer B , Deschenes R et al. American College of Sports Medicine Position Stand . Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise . Med Sci Sports Exerc . 2011; 43 ( 7 ): 1334 – 1359 . CrossRef Search ADS PubMed 29. Borg GA . Psychophysical bases of perceived exertion . Med Sci Sports Exerc . 1982 ; 14 ( 5 ): 377 – 381 . Google Scholar CrossRef Search ADS PubMed 30. Bandura A . The primacy of self-regulation in health promotion . Appl Psychol . 2005 ; 54 ( 2 ): 245 – 254 . Google Scholar CrossRef Search ADS 31. Schwarzer R . Social-cognitive factors in changing health-related behaviors . Curr Dir Psychol Sci . 2001 ; 10 ( 2 ): 47 – 51 . Google Scholar CrossRef Search ADS 32. Buysse DJ , Reynolds CF III , Monk TH , Berman SR , Kupfer DJ . The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research . Psychiatry Res . 1989 ; 28 ( 2 ): 193 – 213 . Google Scholar CrossRef Search ADS PubMed 33. King MB , Whipple RH , Gruman CA , Judge JO , Schmidt JA , Wolfson LI . The Performance Enhancement Project: Improving physical performance in older persons . Arch Phys Med Rehabil . 2002 ; 83 ( 8 ): 1060 – 1069 . Google Scholar CrossRef Search ADS PubMed 34. Carpenter JS , Andrykowski MA . Psychometric evaluation of the Pittsburgh Sleep Quality Index . J Psychosom Res . 1998 ; 45 ( 1 ): 5 – 13 . Google Scholar CrossRef Search ADS PubMed 35. Radloff LS . The CES-D scale: A self report depression scale for research in the general population . Appl Psychol Meas . 1977 ; 1 : 385 – 401 . Google Scholar CrossRef Search ADS 36. Pinquart M , Sorensen S . Differences between caregivers and non-caregivers in psychological health and physical health: A meta-analysis . Psychol Aging . 2003 ; 18 ( 2 ): 250 – 267 . Google Scholar CrossRef Search ADS PubMed 37. Pinquart M , Sörensen S . Helping caregivers of persons with dementia: Which interventions work and how large are their effects ? Int Psychogeriatr . 2006 ; 18 ( 4 ): 577 – 595 . Google Scholar CrossRef Search ADS PubMed 38. Turner-Bowker DM , Bartley PJ , Ware JE. SF-36 Health Survey & SF Bibliography: (1988–2000) . Lincoln, RI : QualityMetric Inc .; 2002 . 39. Hopman WM , Towheed T , Anastassiades T et al. Canadian normative data for the SF-36 health survey . Can Med Assoc J . 2000 ; 162 ( 3 ): 265 – 271 . 40. Spielberger CD. State-Trait Anxiety Inventory: Bibliography . 2nd ed . Palo Alto, CA : Consulting Psychologists Press ; 1989 . 41. Zimet GD , Powell SS , Farley GK , Werkman S , Berkoff KA . Psychometric characteristics of the Multidimensional Scale of Perceived Social Support . J Pers Assess . 1990 ; 55 ( 3–4 ): 610 – 617 . Google Scholar CrossRef Search ADS PubMed 42. Godin G , Shephard RJ . A simple method to assess exercise behavior in the community . Can J Appl Sport Sci . 1985 ; 10 ( 3 ): 141 – 146 . Google Scholar PubMed 43. Jacobs DR Jr , Ainsworth BE , Hartman TJ , Leon AS . A simultaneous evaluation of 10 commonly used physical activity questionnaires . Med Sci Sports Exerc . 1993 ; 25 ( 1 ): 81 – 91 . Google Scholar CrossRef Search ADS PubMed 44. Canadian Society for Exercise Physiology . Physical Activity Training for Health (PATH): CSEP − Health and Fitness Program’s Health-Related Appraisal and Counselling Strategy . Ottawa, Ontario, Canada: Canadian Society for Exercise Physiology ; 2013 . 45. Rikli R , Jones JC. Senior Fitness Test Manual . 2nd ed . Windsor, Ontario, Canada: Human Kinetics Inc.; 2002 . 46. Burr JF , Bredin SS , Faktor MD , Warburton DE . The 6-minute walk test as a predictor of objectively measured aerobic fitness in healthy working-aged adults . Phys Sportsmed . 2011 ; 39 ( 2 ): 133 – 139 . Google Scholar CrossRef Search ADS PubMed 47. Rantanen T , Guralnik JM , Foley D et al. Midlife hand grip strength as a predictor of old age disability . JAMA . 1999 ; 281 ( 6 ): 558 – 560 . Google Scholar CrossRef Search ADS PubMed 48. Jackson A , Langford NJ . The criterion-related validity of the sit and reach test: Replication and extension of previous findings . Res Q Exerc Sport . 1989 ; 60 ( 4 ): 384 – 387 . Google Scholar CrossRef Search ADS PubMed 49. Lemmink KA , Kemper HC , de Greef MH , Rispens P , Stevens M . The validity of the sit-and-reach test and the modified sit-and-reach test in middle-aged to older men and women . Res Q Exerc Sport . 2003 ; 74 ( 3 ): 331 – 336 . Google Scholar CrossRef Search ADS PubMed 50. Eyigor S , Karapolat H , Durmaz B . Effects of a group-based exercise program on the physical performance, muscle strength and quality of life in older women . Arch Gerontol Geriatr . 2007 ; 45 ( 3 ): 259 – 271 . Google Scholar CrossRef Search ADS PubMed 51. Hand BD , Cavanaugh S , Forbes W , Govern J , Cress E . Changes in health related quality of life and functional fitness with exercise training in older adults who attend senior centers . Activit Adapt Aging . 2012 ; 36 : 29 – 54 . 52. Iversen MD , Fossel AH , Katz JN . Enhancing function in older adults with chronic low back pain: A pilot study of endurance training . Arch Phys Med Rehabil . 2003 ; 84 ( 9 ): 1324 – 1331 . Google Scholar CrossRef Search ADS PubMed 53. Li WC , Chen YC , Yang RS , Tsauo JY . Effects of exercise programmes on quality of life in osteoporotic and osteopenic postmenopausal women: A systematic review and meta-analysis . Clin Rehabil . 2009 ; 23 ( 10 ): 888 – 896 . Google Scholar CrossRef Search ADS PubMed 54. Whitehurst MA , Johnson BL , Parker CM , Brown LE , Ford AM . The benefits of a functional exercise circuit for older adults . J Strength Cond Res . 2005 ; 19 ( 3 ): 647 – 651 . Google Scholar PubMed 55. Brant R . Inference for means: comparing two independent samples . Available at http://stat.ubc.ca/~rollin/. Accessibility verified July 25, 2016 . 56. Research randomizer . Available at www.randomizer.org. Accessibility verified April 30, 2015. 57. Freidman LM , Furberg CD , DeMets DL. Fundamentals of Clinical Trials . 3rd ed . New York, NY : Springer ; 1998 . Google Scholar CrossRef Search ADS 58. Quene H , van der Bergh H . On multi-level modeling of data from repeated measures designs: A tutorial . Speech Commun . 2004 ; 43 : 103 – 121 . Google Scholar CrossRef Search ADS 59. West BT . Analyzing longitudinal data with the linear mixed models procedure in SPSS . Eval Health Prof . 2009 ; 32 ( 3 ): 207 – 228 . Google Scholar CrossRef Search ADS PubMed 60. Cohen J. Statistical Power Analysis for the Behavioral Sciences . 2nd ed . New York, NY: Lawrence Erlbaum Associates ; 1988 . 61. Maas CJM , Hox JJ . Sufficient sample sizes for multilevel modeling . Methodology . 2005 ; 1 : 85 – 92 . Google Scholar CrossRef Search ADS 62. Schulz KF , Altman DG , Moher D ; CONSORT Group . CONSORT 2010 statement: Updated guidelines for reporting parallel group randomized trials . Ann Intern Med . 2010 ; 152 ( 11 ): 726 – 732 . Google Scholar CrossRef Search ADS PubMed 63. Public Health Agency of Canada . Fact sheet: Government of Canada chronic disease initiatives . Ottawa, ON ; 2011 . Available at http://www.phac-aspc.gc.ca. Accessibility verified July 2016 . PubMed PubMed 64. Godin G . The Godin – Shepard leisure time physical activity questionnaire . Health Fitness J Canada . 2011 ; 4 ( 1 ): 18 – 22 . 65. Camarri B , Eastwood PR , Cecins NM , Thompson PJ , Jenkins S . Six minute walk distance in healthy subjects aged 55 – 75 years . Respir Med . 2006 ; 100 ( 4 ): 658 – 665 . Google Scholar CrossRef Search ADS PubMed 66. Heyward VH. Advanced Fitness Assessment and Exercise Prescription . 6th ed . Champaign, IL : Human Kinetics ; 2010 . 67. Norman GR , Sloan JA , Wyrwich KW . Interpretation of changes in health-related quality of life: The remarkable universality of half a standard deviation . Med Care . 2003 ; 41 ( 5 ): 582 – 592 . Google Scholar PubMed 68. Bandura A . Health promotion by social cognitive means . Health Educ Behav . 2004 ; 31 ( 2 ): 143 – 164 . Google Scholar CrossRef Search ADS PubMed 69. Fletcher BS , Miaskowski C , Given B , Schumacher K . The cancer family caregiving experience: An updated and expanded conceptual model . Eur J Oncol Nurs . 2012 ; 16 ( 4 ): 387 – 398 . Google Scholar CrossRef Search ADS PubMed © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Annals of Behavioral Medicine Oxford University Press

The Effects of Exercise on Physical and Psychological Outcomes in Cancer Caregivers: Results From the RECHARGE Randomized Controlled Trial

Loading next page...
 
/lp/ou_press/the-effects-of-exercise-on-physical-and-psychological-outcomes-in-CjMZfAsAEj
Copyright
© Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0883-6612
eISSN
1532-4796
D.O.I.
10.1093/abm/kax040
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Family caregivers to cancer patients are at risk for physical and psychological morbidity as a direct result of being in the caregiver role. Exercise as an intervention to support caregiver health and well-being has been tested in other caregiver populations; however, no randomized controlled trials testing the effects of an aerobic and resistance training program have been conducted where cancer caregivers only have participated. Purpose To examine the effects of a 12-week exercise program on quality of life (Short Form 36 v2), psychological outcomes, physical activity levels, and physical fitness in caregivers to cancer patients. Methods Seventy-seven caregivers were randomized to either an exercise intervention or wait-list control. Two participants in the exercise arm dropped out immediately after their baseline assessment. Outcome data for n = 77 participants were analyzed using intention-to-treat analysis with linear mixed models. Results Significant group by time interaction effects were detected for exercise on the mental health component score of quality of life, self-reported weekly exercise levels, and the six-minute walk test. Small to medium effect sizes were detected on several outcomes including physical health component of quality of life (d = 0.36), mental health component of quality of life (d = 0.74), depression (d = 0.41), anxiety (d = 0.35), physical activity levels (d = 0.80), and the six-minute walk test (d = 0.21). Conclusions The findings suggest that exercise may improve health and well-being and could be considered as an additional type of support offered to cancer caregivers. Clinical trial registration no NCT02580461. Exercise, Randomized controlled trial, Cancer caregivers, Physical health • Psychological health Background Family caregivers (hereafter referred to as caregivers) to cancer patients comprise a significant proportion of Canadian and U.S. populations. Most recent statistics indicate that there are approximately 900,000 cancer caregivers in Canada [1] and 4.6 million in the USA [2]. Current models of cancer care delivery are reliant on caregivers to provide up to 50% of the care and support required by patients as they go through treatment and into recovery [3], but this does not come without a cost. Specifically, caregivers may be overwhelmed [4], suffer from physical and psychological health problems [4, 5], and neglect their own health behaviors [6, 7]. Recognizing the importance of caregivers to care delivery and the negative impacts that may occur while in the caregiver role, supporting caregiver health and well-being has become an important health system and health policy agenda [8]. Interventions to support caregiver health and well-being have traditionally focused on psychological support through support groups or individual counselling, education, or respite [3]. In the last 15 years, researchers have also investigated the role of physical activity interventions in supporting or improving caregiver’s psychological and physical health. Many of these interventions have been done with caregivers to Alzheimer/dementia patients and have shown positive effects of physical activity on decreasing caregiver burden [9–12], improving quality of life [13], improving sleep quality [10, 11], increasing activity levels [9–12], and decreasing depression [13–15]. More recently, physical activity interventions with caregivers to cancer patients have been investigated [16–19], with some positive effects on physical and psychological outcomes, including quality of life [17], improved mood [17], and increased sleep quality [18]. The current research on physical activity interventions in cancer caregivers however is lacking in several keys areas. Of the cancer caregiver physical activity interventions published to date, two of the four [17, 18] were not randomized controlled trials and tested the effects of a yoga program. One of the cancer caregiver studies included caregivers only [17], while the other three included patients and caregivers together [16, 18, 19]. Caregiver only interventions have been shown to improve caregivers’ outlook on their caregiving situation [4]. In addition, lack of reporting of factors related to internal validity in these studies makes it difficult to determine the effectiveness of exercise on the outcomes measured [20]. Thus, there is a lack of randomized controlled trial evidence on the effects of aerobic and resistance exercise (i.e., general exercise recommendations) where cancer caregivers only participated. To address the gap in the current literature, we designed a randomized controlled trial exercise intervention for cancer caregivers [21]. This trial is registered with the national clinical trials registry NCT02580461 (clinicaltrials.gov). No changes were made to the design of the trial after the protocol was published; however, the data analysis plan changed slightly (as outlined in the statistical analysis section). The aim of the study was to examine quantitative and qualitative outcomes using a mixed methods design. The quantitative results of the 12-week structured exercise program on self-reported physical health (primary outcome), psychological well-being, physical fitness levels, and physical activity levels (secondary outcomes) in caregivers caring for adult cancer patients are reported here. The qualitative results are reported elsewhere [22]. Objectives The objectives of the study were the following: (a) to examine whether a 12-week exercise program increases the physical health component score on the Short Form 36 v2 [23] measure of quality of life in those caregivers randomized to the exercise intervention versus control condition; (b) to examine whether a 12-week exercise program increases self-reported physical activity levels, physical fitness levels (cardiorespiratory, muscular, flexibility, and balance), overall self-reported quality of life, self-reported sleep quality, and perceived social support, and decreases self-reported depression and anxiety in those caregivers randomized to the exercise intervention versus control condition; and(c) to document adherence rates to exercise in those randomized to the intervention group. Materials and Methods Trial Design and Participants We conducted a randomized controlled trial with an exercise intervention group and a wait-list control group using a 1:1 allocation ratio. Participants in the wait-list control group provided baseline and 12-week assessment data only; they did not complete any other individual or group activities and did not track their physical activity. Participants in the wait-list control group were enrolled in the exercise intervention after completing 12 weeks of control. All participants (intervention and control group) were invited to participate in a 12-week maintenance exercise program once they had completed the 12-week exercise intervention. Individuals who were providing care to an adult patient were eligible to participate. We defined caregiving as providing physical or psychological support to a loved one living with cancer. If participants asked whether they were considered a caregiver, we provided them with a list of caregiving roles as outlined by the National Cancer Institute [24] for clarification. Other eligibility criteria included: (a) not planning to move in the next 12 months; (b) willing to attend twice weekly exercise classes; (c) being free from any medical conditions that would prevent participation in an exercise program; (d) not achieving recommended levels of physical activity as outlined by the Canadian physical activity guidelines [25]; and (e) able to speak and read English. Exclusion criteria were (a) providing care to a child with cancer; (b) having a health condition that limits participation in a physical activity program (e.g., unstable angina, uncontrolled hypertension, or diabetes); (c) being a bereaved caregiver (previously caring for a cancer patient who has now died); (d) meeting Canadian physical activity guidelines for the past three months; and (e) unable to attend the group exercise classes. Participants were recruited from a previously conducted caregiver survey [26] as well as through poster advertisement at a number of hospital and community facilities. Prior to enrollment, participants were screened by the research assistant or student co-investigator to determine their eligibility. Standardized screening protocols were used to determine ability to participate in an exercise program [27]. The trial was conducted at a cancer and exercise-specific research lab (Health and Wellness lab) within a University in a large city in western Canada that provides exercise space with aerobic and resistance training equipment. Approval from the local ethics board was obtained prior to study initiation (HREBA.CC‐16‐0801). Intervention We used the Canadian physical activity guidelines for adults to determine the type and amount of exercise that participants should be targeting (i.e., the goal was to achieve 150 min of aerobic exercise per week in bouts of 10 minor more, and 2 days per week of resistance exercises for each of the major muscle groups) [25]. For this study, exercise was defined as “physical activity that is planned, structured and repetitive and that has as a final or intermediate objective the improvement or maintenance of physical fitness” [28]. The 12-week exercise intervention consisted of (a) twice weekly resistance exercise group classes at the Health and Wellness lab; (b) additional weekly independent aerobic exercise sessions to reach 150 min per week; and (c) seven group-based education sessions during the 12-week intervention. The purpose of the education sessions was to address common barriers to physical activity and to reinforce principles of behavior change. This strategy has been effectively employed in previous exercise for caregiver interventions [9, 11] and is in keeping with theoretically based behavioral interventions. Prior to randomization, a standardized fitness battery (outlined in the physical fitness outcomes section) was performed by a certified exercise physiologist. Based on the results of the fitness testing and any physical limitations to aerobic or resistance exercise noted during the assessment, the certified exercise physiologist designed an individualized resistance and aerobic exercise prescription for the participant. A written copy of the prescription was provided to each participant. The twice weekly 1-hr center-based group exercise classes focused on resistance training wherein participants completed their prescribed resistance exercises. The class was overseen by a certified personal trainer and a kinesiology student volunteer. In general, participants performed one to two sets of 10–12 repetitions of resistance exercises that incorporated the major muscle groups of the body (10–12 exercises), core, and balance exercises. Adherence to the exercise class was tracked via recording attendance at each class. For the aerobic component, participants were instructed to participate in at least 10 min of aerobic exercise of their choice, at a perceived level of exertion of at least 5–6 [29], and then to progress over the course of the intervention to achieving 150 min of moderate aerobic activity per week. Perceived level of exertion is a consistently used method to assess subjective perception of exercise intensity, has been validated against physiologic measures of exercise intensity, and is practical, affordable, and feasible to use in a variety of exercise situations. An introduction to aerobic exercise including ideas for types of aerobic exercise, goal setting, warm up and cool down, measuring perceived exertion, and tracking progress was provided in the first education session along with written materials. Adherence to the aerobic prescription was monitored via weekly self-report physical activity logs. Progression through the aerobic and resistance training programs followed a pre-set schedule devised by the certified personal trainer and then tailored to each individual participant based on their fitness level and resistance training exercise experience. In general, for resistance training, in the first 1 to 3 weeks, participants used body weight or very light weight to familiarize themselves with the proper technique. In Weeks 2–6, resistance was gradually increased to the point where participants could perform 2 sets of 10–12 repetitions of each resistance exercise. In the final 6 weeks, the goal was for participants to increase their resistance and perform 2 sets of 8–10 repetitions. Each resistance training session was tracked by the participant while they were in the class (number of sets, weight, and repetition) for each exercise and then reviewed weekly by the certified personal trainer. For the independent aerobic sessions, depending on fitness level, participants began with a minimum of 10 min per session at a perceived level of exertion between 2 and 5 [29]. Based on feedback from the certified personal training and exercise physiologist, progression occurred at a rate of no more than 10% per week for duration of activity. Intensity was slowly increased so that participants were consistently achieving a perceived level of exertion of 5–6 per session. There were seven education sessions offered to participants. The sessions occurred biweekly, immediately after the group resistance exercise class, and were taught by the certified personal trainer or the student co-investigator. The education sessions included written and verbal information about healthy behaviors (sleep hygiene, benefits of physical activity and physical activity guidelines, stress management, maintaining exercise behaviors, and self-compassion) and strategies for successful behavior change (goal setting and increasing self-efficacy). Theoretical foundations for behavior change were based on Social Cognitive Theory [30] and the Health Action Process Approach [31]. We did not record attendance at the education sessions. All staff involved in the trial (research assistant, CPTs, CEPs, and principle investigator) had previous experience running exercise trials for cancer populations through the Health and Wellness lab and were trained about the specifics of this trial during study meetings held prior to participant enrollment. Outcomes A variety of patient reported, and physical fitness outcomes were collected. All outcomes were assessed at baseline, post exercise intervention (at 12 weeks), and post maintenance (at 24 weeks). Baseline and post-intervention data are reported here. Assessments were conducted by a certified exercise physiologist blinded to treatment allocation and took approximately 1.5 hr to complete. Research participants, the research assistant, and certified personal trainers leading the exercise classes were aware of treatment allocation. Demographic data Self-reported demographic data included age, gender, marital status, level of education, employment status, and income. Caregiving information included number of months in the caregiver’s role and number of hours of care provided per week (e.g., transportation, meal preparation, medication management, side effect management, and personal care). Care recipient (patient) information was also collected, as reported by the caregiver, including diagnosis, time since diagnosis, type of treatment currently on, relationship to the caregiver, and age of the patient. Psychosocial outcomes The Pittsburgh sleep quality index, a validated self-report questionnaire [32], was used to assess sleep quality. This 19-item questionnaire has been used in previous caregiver research [33] and has consistent internal reliability (α ≥ 0.80), and constructs validity upon comparison to four other sleep rating instruments [34]. A total score of 5 or greater is indicative of poor sleep quality. The Centers for Epidemiological Studies Depression scale [35], a 20-item questionnaire, was used to measure depression. A score of 16 or greater may indicate clinical depression. The Centers for Epidemiological Studies Depression scale has good sensitivity and specificity and high internal consistency (α ≥ 0.80), has been found to detect differences between caregivers and noncaregivers [36], and has been used to measure changes in caregiver depressive symptoms after interventions [37]. Quality of life was measured using the Short Form 36 v2 [23]. This self-report instrument consists of 36 items and measures 2 domains (mental health component and physical health component) of quality of life. The Short Form 36 has been widely used in many populations, has good internal consistency (α ≥ 0.8), and has well-documented content and construct validity [38]. Normative data have been published on Canadian populations [39]. The State Trait Anxiety index [40] was used to assess anxiety. This questionnaire contains 20 items and has been used in clinical settings to diagnose anxiety, and has good internal consistency (α ≥ 0.85) and test–retest reliability [40]. Normative age group data have been published, and generally a score of 40 or greater is associated with clinical anxiety [40]. Social support was measured using the Multidimensional Scale of Perceived Social Support, which is a 12-item questionnaire with high internal consistency (α ≥ 0.90) and good construct validity [41]. Total scores of 5 or greater on this measure are indicative of good social support. Exercise behaviors Physical activity behaviors were measured using self-report and class attendance. The Godin leisure time exercise questionnaire [42] was used to determine changes to weekly level of exercise over time. This self-report three-item instrument collects information on weekly time spent in strenuous, moderate, and mild physical activity. The Godin leisure time exercise questionnaire is widely used in many populations and has shown consistent test–retest validity (correlations between r = .24 for light activity and r = .84 for vigorous activity) compared with other self-report physical activity questionnaires [43]. Self-report weekly physical activity logs (type of exercise, time participating, and perceived level of exertion) and documented attendance at twice weekly exercise classes were also recorded. Physical fitness assessments Measurements of physical fitness included body composition, cardiorespiratory fitness, musculoskeletal fitness, flexibility, and balance. All fitness measures were conducted by a certified exercise physiologist using standardized protocols as outlined by the Canadian Society of Exercise Physiology physical activity training for health [44] or the Seniors fitness test [45]. The same exercise physiologist completed all of the assessments required on each individual participant. Measurements of physical fitness were chosen based on several factors including our outcome assessors familiarity with these tests, the potential of aged participants with low physical activity levels and experience, and possible comorbidities. Weight and height were used to calculate body mass index score. Waist and hip circumferences were used to calculate hip to waist ratio. Cardiorespiratory fitness was measured using a 6-min walk test (6MWT), which is a valid functional test of aerobic fitness in individuals who may be sedentary or have comorbidities [45, 46]. The 6MWT has been found to have high retest reliability (0.95) and to have both convergent and discriminant validity when used to assess aerobic fitness in older adults [45]. Muscular strength tests included grip strength, which is widely used as an indicator of overall body strength, is predictive of future functional limitations [47], and has normative age values [44]. In addition, a 30-s sit-to-stand test was used to assess lower body muscular endurance [45]. Flexibility was assessed using the sit-and-reach test [48, 49]. Balance was measured using a one-leg stance test [44]. Sample size The sample size calculation, as described in the published protocol [21], was based on the physical health component score of the Short Form 36 as the primary outcome variable. As changes to Short Form 36 scores after exercise interventions have ranged between 3.0 to 17.75 points, and standard deviations have ranged from 3 to 26 [13, 33, 50–54], we chose a change of 10 points and a standard deviation of 15 for the sample size calculation. A difference in means sample size calculation was used [55]. The sample size from this calculation, which sets power to detect a difference between groups and over time on the physical health component score of the Short Form 36 at 0.80, was 72 participants (i.e., 36 in each group). Randomization After written informed consent was obtained and baseline testing completed, participants were randomized. To protect against bias, participants were randomly assigned to one of the two different intervention arms (exercise intervention or control) using Research Randomizer [56], which is a web-based randomization generator. To avoid imbalance in the number of participants assigned to each group [57], block randomization was used and included 10 sets of 8 nonunique numbers per set. The student co-investigator generated the randomization assignment and prepared individual opaque randomization envelopes prior to any participants being enrolled. The contents of the randomization envelopes were known only to the student co-investigator. The research assistant accessed a new randomization envelope after the baseline assessment for each participant and then contacted the participant to inform them of their allocation. Statistical Methods Statistical analyses of data were performed using IBM Statistical Package for Social Sciences software version 22.0. An intention–to-treat analysis was carried out using linear mixed models (LMMs), so that all participants who provided baseline data could be included in the analysis. As the data set contained repeated-measures, linear mixed modeling is considered a more robust test of significance compared with repeated-measures ANOVA [58, 59] for several reasons. Missing data, a common occurrence in repeated measures, do not need to be discounted. Also, the assumption of sphericity (equal co-variances among groups) does not need to be met, variables measured at different levels (e.g., multilevel sampling) can be incorporated, and the assumption of independence is not violated by using LMMs [59]. Descriptive statistics were used to characterize the data and examine differences between baseline characteristics of the intervention and control groups. The t-test, chi-squared test, or Fisher’s Exact Test was used to examine group differences at baseline. Descriptive statistics were used to characterize adherence rates to the exercise intervention as the mean class attendance over 12 weeks, as well as mean level of aerobic, resistance, and total minutes of exercise per week. Fitness testing, including 6MWT, 30-s sit to stand, trunk forward flexion sit and reach, and balance, was characterized using measures of central tendency (mean and standard deviation) at baseline and Time 1. To determine group by time interaction effects of the 12-week exercise intervention on the Short Form 36 physical health component score, we used linear mixed modeling (primary outcome). The fixed effects were the intercept, group, time, and group by time interaction. Time was also set as a repeated effect. We tested autoregressive and unstructured covariance structures and chose the autoregressive covariance as the −2 restricted log likelihood was smaller. As several participants’ caregiving situations changed throughout the course of the trial (e.g., four participants’ care recipient died, three participants’ care recipient had a recurrence, and two care recipients completed treatment), we controlled for caregiving hours per week by including this as a covariate in our model. All assumptions of covariance were met. The estimated marginal mean change in physical component score for exercise and control group was calculated using LMM. Cohen’s d effect sizes were calculated using the unstandardized mean change scores (12 weeks – baseline) and the pooled standard deviations [60]. To determine group by time interaction effects of the 12-week exercise intervention on the secondary outcomes of Short Form 36 mental health component score, Pittsburgh sleep quality index, Centers for Epidemiological Studies Depression scale, Multidimensional scale of perceived social support, State trait anxiety inventory, total Godin leisure time exercise questionnaire score, fitness measures, and BMI, we used the same linear mixed modeling procedure. For the group by time interaction effect of exercise on the Godin leisure time exercise questionnaire, the covariate caregiving hours were not included in the LMM as the assumption of homogeneity of regression slopes was violated. Maas and Hox [61] suggested that 30 data points at 2 different levels provide reasonable unbiased assessment of regression coefficients and variance; therefore, we determined that our sample size was adequate to run this secondary analysis. Results Recruitment for the trial started in May 2015 and was ongoing until February 2016. During that time, a total of 230 participants were assessed for eligibility, and 153 participants were excluded (n = 4 could not attend the location for the classes, n = 21 were already exercising at recommended levels, n = 120 were not interested in an exercise program, and n = 8 were too overwhelmed with their caregiving responsibilities). A Consolidated Standards of Reporting Trials [62] diagram of participant flow through the trial is depicted in Fig. 1. A total of 77 eligible participants were enrolled in the study and were randomized to either the exercise intervention or control group. Two participants who were randomized to the exercise intervention group dropped out immediately after their baseline assessment (e.g., did not start the intervention) because of changes to their care recipient status. Of the 39 participants who were in the wait-list control group, 6 dropped out of the study after completing their 12-week assessment and did not start the intervention. The 12-week exercise intervention started in June 2015 and continued until June 2016 at which point all participants had been enrolled and completed their treatment allocation. We used a rolling recruitment and enrollment strategy whereby after four participants had been randomized to the exercise group, we started the group-based exercises classes. As more participants were recruited and enrolled, they joined the existing classes. In total, 14 groups were run sequentially, with each group having between 8 and 10 participants. We ran either daytime or evening groups to accommodate caregivers’ schedules. Once enrolled in a group, the caregivers remained in the same group. All participants followed the same protocol and were taught by one of the two certified personal trainers. All participants progressed through the exercise program as anticipated. Most participants chose walking as their aerobic exercise. Across 12 weeks, 51% of participants were meeting the physical activity guidelines of 150 min of moderate to vigorous aerobic activity per week and two sessions per week of resistance training. Fig. 1. View largeDownload slide Consolidated Standards of Reporting Trials (CONSORT) [62] diagram of flow of participants through the RECHARGE trial. ITT intention to treat; LMM linear mixed models. Fig. 1. View largeDownload slide Consolidated Standards of Reporting Trials (CONSORT) [62] diagram of flow of participants through the RECHARGE trial. ITT intention to treat; LMM linear mixed models. Baseline Data The majority of participants were female (61%), the average age was 53 years, and participants generally had a high socioeconomic status. Most participants were either retired or not working outside of the home (41.6%), and were providing care to their spouse (79.2%). Caregivers were providing care to patients with a variety of cancer types, most commonly breast (22.1%), prostate (19.5%), colorectal (11.7%), and hematological malignancies (16.9%). Most patients (87%) were on some type of treatment for their cancer. Time in the caregiver role was on average 30 months, and average time spent providing care per week was 19 hr. Baseline analysis indicated no differences between the exercise and control group on either demographic or caregiving variables; therefore, data presented in Table 1 are combined for all participants. The only statistical difference found between groups was that more caregivers in the control group were living with their care recipient than in the exercise intervention group (p = .034). Table 1 Demographic and caregiving variables for cancer family caregivers in RECHARGE study, N = 77 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 SD standard deviation. aWhere assumptions of chi-squared test were violated, Fisher’s Exact Test was used. View Large Table 1 Demographic and caregiving variables for cancer family caregivers in RECHARGE study, N = 77 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 Variables All (N = 77) Exercise (n = 38) Control (n = 39) t-Test/chi- squared test Significance Mean (SD/range) Mean (SD/range) Mean (SD/range) Age 53.25 (12.25/22–79) 51.42 (12.6/22–75) 55.03 (11.76/30–79) −1.3 0.199 Age of patient/care recipient 57.46 (12.25/22–79) 57.11 (13.5/27–82) 57.79 (12.6/21–78) −0.229 0.82 Caregiving, hr/week 19.16 (23.2/2–168) 15.09 (15.2/2–84) 23.12 (28.9/2–168) −1.515 0.134 Caregiving, time in months 29.96 (40.69/2–240) 29.0 (47.9/2–240) 30.91 (32.75/3–144) −0.204 0.839 % % % Gender 0.02 0.887  Men 39 36.8 41  Women 61 63.2 59 Marital status 0.002a 1  Married/Common law 89.6 89.5 89.7  Not married 10.4 10.5 10.3 Education level 4.368a 0.228  Some high school 5.2 2.6 7.7  High school 10.4 13.2 7.7  Post secondary 84.4 84.3 84.7 Employment status 0.078 0.962  Work full time 44.2 44.7 43.6  Work part time 14.3 13.2 15.4  Retired/Do not work outside home 41.6 42.1 41 Income 3.304a 0.347  Less than or equal to $50 000/year 13 15.8 10.3  $50,001–100,000/year 44.2 34.2 53.8  More than or equal to $100,000/year 32.5 39.5 25.6  Do not wish to answer 10.4 10.5 10.3 Living with patient 4.5 0.034  Yes 81.8 71.1 92.3  No 18.2 28.9 7.7 Relationship to patient 2.173a 0.367  Spouse 79.2 73.7 84.6  Daughter/Son 14.3 15.8 12.8  Other 6.5 10.5 2.6 Type of cancer 4.181a 0.392  Breast 22.1 18.4 25.6  Prostate 19.5 26.3 12.8  Colorectal 11.7 15.8 7.7  Hematological 16.9 13.2 20.5  Other 29.9 26.3 33.3 Patient on treatment 0.002a 1  Yes 87 86.8 87.2  No 13 13.2 12.8 SD standard deviation. aWhere assumptions of chi-squared test were violated, Fisher’s Exact Test was used. View Large Baseline measures of physical and psychological health, quality of life, physical activity levels, and fitness testing for the exercise intervention and control group are shown in Table 2. There were no statistically significant differences between groups on these baseline measures. Participants reported high levels of state and trait anxiety (M = 40.20 and 42.5, respectively), depression (M = 18.04), and poor sleep quality (M = 8.34). Participants rated their perceived social support as moderate (M = 5.04). The physical health component of quality of life (M = 52.58) was, on average, higher in our participants than standardized norms for the Canadian population; however, the mental health component of quality of life (M = 39.64) was lower than standardized norms [39]. Total number of health conditions (M = 1.4) was comparable to health conditions of similar aged population [63]. Self-report of physical activity (M = 23.58) showed that most participants were considered moderately active [64]. Anthropomorphic and physical fitness measures showed the majority of participants were overweight (mean BMI = 8.26), however had aerobic fitness (mean 6MWT = 639.59 m) within healthy adult range [65]. Measures of strength showed that the majority of participants had fair grip strength (M = 74.46) [66]. No statistically significant differences were found between groups on measures of physical fitness. Table 2 Baseline physical and psychological health, quality of life, PA levels, and fitness of cancer caregivers in RECHARGE trial (N = 77) Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 SD standard deviation; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; BMI body mass index; PA physical activity; QOL quality of life. a PA levels using Godin Leisure Time Exercise questionnaire. Strenuous activity = times per week participants engaged in strenuous activity for 15 min. Moderate activity = times per week participants engaged in moderate activity for 15 min. Mild activity = times per week participants engaged in mild activity for 15 min. Total weekly score = (9 * strenuous) + (5 * moderate) + (3 * mild). View Large Table 2 Baseline physical and psychological health, quality of life, PA levels, and fitness of cancer caregivers in RECHARGE trial (N = 77) Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 Exercise group (n = 38) Control group (n = 39) t-Test/χ2 test Significance Mean (SD/range) Mean (SD/range) Psychological health  CES-D 17.03 (11.31/0–45) 19.03 (11.48/4–49) −0.77 0.444  STAI-S 38.71 (13.03/20–66) 41.68 (13.55/21–74) −0.975 0.333  STAI-T 41.66 (13.80/21–72) 42.44 (12.38/25–71) −0.261 0.795  MSPSS 5.06 (1.56/1.17–7) 5.03 (1.42/1.08–7) 0.099 0.921 Physical health  Number of health conditions 1.16 (1.22/0–5) 0.90 (0.88/0–3) 1.071 0.288  Number of medications 1.53 (1.89/0–8) 1.59 (1.87/0–6) −0.148 0.883 Smoking status 0.374 0.615  Nonsmoker (%) 94.7 97.4  Smoker (%) 5.3 2.6 PSQI 8.46 (3.88/3–17) 8.41 (4.16/2–16) 0.053 0.958 QOL  SF-36 PCS 53.44 (7.99/31.61–67.56) 51.75 (7.57/32.36–68.43) 0.953 0.343  SF-36 MCS 38.54 (10.86/15.06–59.58) 40.71 (11.94/17.96–62.36) −0.836 0.406 Physical activity levelsa  Strenuous activity 0.53 (1.2/0–5) 0.59 (1.83/0–10) −0.179 0.858  Moderate activity 1.76 (1.98/0–6) 1.95 (2.48/0–7) −0.362 0.718  Mild activity 3.24 (3.08/0–15) 2.95 (2.45/0–7) 0.455 0.65  Total weekly score 23.26 (18.96/0–96) 23.90 (25.21/0–122) −0.124 0.901 Physical fitness measures  BMI, kg/m2 27.37 (5.19/17.5–41.1) 29.12 (6.80/18.7–53.7) −1.264 0.21  Waist-to-hip ratio 0.94 (0.06/0.81–1.04) 0.95 (0.08/0.74–1.15) −0.992 0.324  Balance—eyes open, s 33.57 (14.48/1.84–45) 31.19 (15.96/1.63–45) 0.306 0.76  Balance—eyes closed, s 9.59 (10.39/1.0–45) 7.32 (7.19/1–33.96) 0.625 0.534  Sit to Stand, no. of stands 15.97 (4.52/8–36) 16.72 (5.96/8–33) 0.618 0.539  Sit and Reach, cm 24.74 (10.42/1.80–43) 23.74 (11.68/1.9–41.3) 0.389 0.698  Grip strength, kg 73.21 (20.73/31.8–111) 75.69 (17.3/38–114) −0.572 0.569  6-Min walk test, m 651.64 (74.36/450–800) 627.84 (101.70/359–840) 1.17 0.246 SD standard deviation; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; BMI body mass index; PA physical activity; QOL quality of life. a PA levels using Godin Leisure Time Exercise questionnaire. Strenuous activity = times per week participants engaged in strenuous activity for 15 min. Moderate activity = times per week participants engaged in moderate activity for 15 min. Mild activity = times per week participants engaged in mild activity for 15 min. Total weekly score = (9 * strenuous) + (5 * moderate) + (3 * mild). View Large Primary Outcome LMM analysis revealed that there was no significant group by time interaction effect on the physical health component of quality of life across 12 weeks (F [1, 57.35] = 1.04; p = .31) after controlling for the covariate care hours per week. The covariate was not significantly related to the physical component score. In addition, no significant main effects of time or group assignment were found (Table 3). Estimated marginal means analysis showed that the exercise group physical component score increased by 0.30 over 12 weeks, while the control group physical component score decreased by −1.24 (Table 4). Despite the nonsignificant effects, a small effect size for the exercise intervention on physical health component score (d = .36) was found (Table 5). Table 3 Linear mixed model analysis assessing outcome for intention-to-treat exercise and control group from baseline to 12 weeks Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aCovariate care hours/week was not included as preliminary analysis revealed that the assumption of homogeneity of regression slopes was violated. *Statistically significant at p < .05 level. View Large Table 3 Linear mixed model analysis assessing outcome for intention-to-treat exercise and control group from baseline to 12 weeks Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 Outcomes Group effect Time effect Group × time effect Covariate care hours/week F (df1, df2) p F (df1, df2) p F(df1, df2) p F (df1, df2) p Quality of life  SF-36 PCS 2.28 (1, 71.68) .14 0.37 (1, 58.12) .54 1.04 (1, 57.35) .31 .05 (1, 97.73) .82  SF-36 MCS .08 (1, 75.88) .78 25.01 (1, 62.28) <.001* 7.02 (1, 61.52) .01* 3.23 (1, 100.16) .08 Psychological health  CES-D .73 (1, 74.13) .40 12.12 (1, 60.02) .001* 1.01 (1, 59.33) .32 .57 (1, 94.50) .45  PSQI .00 (1, 70.89) .99 2.17 (1, 57.02) .15 0.55 (1, 56.29) .46 4.07 (1, 97.58) .05  STAI-S .90 (1, 75.11) .35 6.46 (1, 64.71) .01* .35 (1, 63.86) .56 1.93 (1, 117.20) .17  STAI-T .04 (1, 59.86) .85 12.77 (1, 60.50) .001* .35 (1, 59.86) .56 1.68 (1, 91.65) .20  MSPSS .11 (1, 74.68) .75 2.78 (1, 61.04) .10 .00 (1, 60.27) .99 1.24 (1, 105.18) .27 Physical activity  GLTEQ 2.64 (1, 74.79) .11 4.70 (1, 69.26) .03* 6.85 (1, 69.26) .01* not includeda Physical fitness  BMI, kg/m2 1.87 (1, 78.42) .18 2.16 (1, 54.64) .15 .06 (1, 54.37) .81 .63 (1, 56.25) .43  Balance—eyes open, s 1.46 (1, 102.2) .23 .014 (1, 59.75) .91 4.49 (1, 56.92) .04* .00 (1, 77.39) .99  Balance—eyes closed, s 1.6 (1, 108.63) .21 .47 (1, 64.16) .50 2.57 (1, 58.21) .11 1.28 (1, 95.01) .26  Sit to Stand, no. of stands .17 (1, 76.29) .68 17.37 (1, 58.69) <.001* .91 (1, 58.24) .35 .10 (1, 76.46) .75  Sit and Reach, cm .39 (1, 90.70) .53 2.44 (1, 53.95) .12 1.16 (1, 52.70) .29 .19 (1, 61.73) .66  Grip strength, kg .29 (1, 74.69) .59 1.25 (1, 57.38) .27 .05 (1, 57.12) .83 .74 (1, 67.63) .39  6MWT, m 2.51 (1, 73.80) .12 33.46 (1, 56.16) <.001* 6.90 (1, 56.03) .01* .48 (1, 61.76) .49 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aCovariate care hours/week was not included as preliminary analysis revealed that the assumption of homogeneity of regression slopes was violated. *Statistically significant at p < .05 level. View Large Table 4 Estimated marginal group mean for outcome measures at baseline and 12 weeks Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aEstimated marginal mean evaluated with covariate care hours = 14.95. View Large Table 4 Estimated marginal group mean for outcome measures at baseline and 12 weeks Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 Outcome Group Baseline meana (SE) 12 weeks meana (SE) Mean difference Quality of life  SF-36 PCS Exercise 53.44 (1.27) 53.74 (1.37) 0.30 Control 51.64 (1.28) 50.39 (1.35) −1.24  SF-36 MCS Exercise 38.56 (1.83) 46.90 (1.96) 8.34 Control 40.70 (1.84) 43.36 (1.94) 2.66 Psychological health  CES-D Exercise 17.02 (1.76) 12.55 (1.88) −4.47 Control 18.07 (1.77) 15.57 (1.86) −2.51  PSQI Exercise 8.33 (0.65) 7.44 (0.70) −0.89 Control 8.04 (0.66) 7.73 (0.69) −0.31  STAI-S Exercise 38.70 (2.05) 33.91 (2.26) −4.79 Control 40.34 (2.11) 37.34 (2.24) −3.01  STAI-T Exercise 41.64 (1.99) 37.16 (2.11) −4.48 Control 41.53 (2.01) 38.29 (2.10) −3.24  MSPSS Exercise 5.07 (0.24) 5.32 (0.26) 0.26 Control 5.17 (0.24) 5.43 (0.25) 0.26 Physical activity  GLTEQ Exercise 22.67 (3.42) 35.55 (3.77) 12.89 Control 23.89 (3.42) 22.64 (3.58) −1.26 Physical fitness  BMI, kg/m2 Exercise 27.37 (0.97) 27.36 (0.98) 0.016 Control 29.17 (0.97) 29.04 (0.98) 0.085  Balance—eyes open, s Exercise 33.57 (2.27) 38.75 (2.43) 5.18 Control 31.52 (2.3) 32.23 (2.36) 0.70  Balance—eyes closed, s Exercise 9.77 (1.3) 8.62 (1.44) −1.15 Control 7.13 (1.3) 8.7 (1.4) 1.58  Sit to Stand, no. of stands Exercise 15.97 (0.89) 17.92 (0.94) 1.94 Control 16.84 (0.89) 18.08 (0.92) 1.24  Sit and Reach, cm Exercise 24.13 (1.9) 27.66 (1.9) 3.53 Control 23.46 (1.85) 25.73 (1.9) 2.27  Grip strength, kg Exercise 73.20 (3.06) 74.51 (3.14) 1.31 Control 75.70 (3.07) 76.59 (3.12) 0.89  6MWT, m Exercise 651.63 (14.22) 680.09 (14.43) 28.46 Control 628.68 (14.24) 639.69 (14.38) 11.10 SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; BMI body mass index; 6MWT 6-min walk test. aEstimated marginal mean evaluated with covariate care hours = 14.95. View Large Table 5 Unadjusted means between exercise and control group for baseline and 12 weeks, and effect sizes for outcome variables Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 SD standard deviation; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; 6MWT 6-min walk test. aCohen d formula used = difference between two mean changes between groups (T2–T1) divided by pooled SDs. View Large Table 5 Unadjusted means between exercise and control group for baseline and 12 weeks, and effect sizes for outcome variables Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 Outcome Time Exercise group (n = 38) Control group (n = 39) Cohen da Mean (SD) Mean (SD) Quality of life  SF-36 PCS Baseline 53.44 (7.99) 51.75 (7.57) 12 weeks 54.70 (7.79) 50.17 (7.96) 0.36  SF-36 MCS Baseline 38.54 (10.86) 40.71 (11.94) 12 weeks 48.81 (11.98) 42.56 (11.3) 0.74 Psychological health  CES-D Baseline 17.03 (11.31) 19.03 (11.48) 12 weeks 10.42 (10.86) 16.89 (11.2) −0.41  PSQI Baseline 8.46 (3.88) 8.41 (4.16) 12 weeks 6.37 (4.08) 7.80 (4.43) −0.35  STAI-S Baseline 38.71 (13.03) 41.68 (13.55) 12 weeks 32.67 (12.65) 38.11 (13.17) −0.19  STAI-T Baseline 41.66 (13.80) 42.44 (12.38) 12 weeks 34.16 (14.02) 39.57 (11.36) −0.34  MSPSS Baseline 5.06 (1.56) 5.03 (1.42) 12 weeks 5.46 (1.73) 5.09 (1.55) 0.22 Physical activity  GLTEQ Baseline 23.26 (18.96) 23.90 (25.21) 12 weeks 38.5 (21.1) 22.37 (18.37) 0.8 Physical fitness  BMI, kg/m2 Baseline 27.37 (5.19) 29.12 (6.80) 12 weeks 26.9 (5.34) 28.96 (6.72) −0.05  Balance eyes open, s Baseline 33.57 (13.61) 31.19 (15.96) 12 weeks 36.45 (12.85) 31.54 (15.02) 0.18  Balance eyes closed, s Baseline 9.59 (9.71) 7.32 (7.19) 12 weeks 9.60 (8.59) 8.78 (6.47) −0.19  Sit to Stand, no. of stands Baseline 15.97 (4.52) 16.72 (5.96) 12 weeks 17.87 (4.62) 17.97 (6.91) 0.11  Sit and Reach, cm Baseline 24.74 (10.42) 23.74 (11.68) 12 weeks 27.65 (10.1) 24.88 (11.93) 0.16  Grip strength, kg Baseline 73.21 (20.73) 75.69 (17.3) 12 weeks 71.41 (19.4) 76.28 (18.64) 0.13  6MWT, m Baseline 651.64 (74.36) 627.84 (101.70) 12 weeks 679.7 (69.73) 638.18 (99.85) 0.21 SD standard deviation; SF-36 PCS Short Form 36 v2 physical component score; SF-36 MCS Short Form 36 v2 mental component score; CES-D Centers for Epidemiological Studies Depression; STAI-S State Trait Anxiety Index—State; STAI-T State Trait Anxiety Index—Trait; MSPSS multidimensional scale of perceived social support; PSQI Pittsburgh sleep quality index; GLTEQ Godin Leisure Time Exercise questionnaire; 6MWT 6-min walk test. aCohen d formula used = difference between two mean changes between groups (T2–T1) divided by pooled SDs. View Large Secondary Outcome Results Results of the LMM analysis of secondary outcomes including Short Form 36 mental health component score, Pittsburgh sleep quality index, Centers for Epidemiological Studies Depression scale, State trait anxiety inventory, Godin leisure time exercise questionnaire, 6MWT, 30-s sit-to-stand test, and grip strength are shown in Table 4. Quality of life Short Form 36 mental health component score There was a statistically significant group by time interaction effect for the Short Form 36 mental health component score (F [1, 61.52] = 7.02; p = .01) after controlling for the effect of care hours per week (Fig. 2), which indicated that the group effect varied with time. Simple effects testing indicated that for the exercise group, there was a statistically significant increase in their mental component scores across the 12 weeks (F [1, 62.23] = 29.27; p < .001). The mental health component score increased in the exercise group by 8.34 points and in the control group by 2.66 points based on estimated marginal means analysis (Table 5). The effect size of the intervention on the mental health component score was medium (d = 0.74; Table 5). Fig. 2. View largeDownload slide Estimated marginal mean SF-36 MCS score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. SF-36 MCS Short Form 36 v2 mental component score. Fig. 2. View largeDownload slide Estimated marginal mean SF-36 MCS score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. SF-36 MCS Short Form 36 v2 mental component score. Psychological health Depression Results of the LMM analysis for depression showed no significant group by time interaction effects; however, main effects for time were detected (F [1, 60.02] = 12.12; p = .001). Regardless of group assignment, both groups showed improved depression scores (i.e., less depressed) over 12 weeks. The covariate care hours per week was not significantly related to depression scores. In addition, the intervention effect to decrease depression scores (i.e., less depressed) was small (d = −0.41; Table 5). Anxiety Main effects for time were found for both state anxiety (F [1, 64.71] = 6.46; p = .01) and trait anxiety (F [1, 60.50] = 12.77; p = .001). Regardless of group assignment, both groups had decreased levels of anxiety over 12 weeks (Table 5). No group by time interactions were detected. Effect sizes for the exercise intervention on anxiety scores were small (Table 5). Sleep quality Sleep quality was significantly related to the covariate care hours (F [1, 97.58] = 4.07; p = .05). No group by time interaction effects or main interactions were found for sleep quality. Small effect sizes for the intervention on sleep quality were detected (d = −0.35; Table 5). Social support No differences on social support were detected between the groups or over time. Small effect sizes for exercise on social support were demonstrated (d = 0.22) Physical activity levels and physical fitness There were significant group by time interaction effects on the Godin leisure time exercise questionnaire scores (F [1, 69.26] = 6.84; p = .011) and the 6MWT (F [1, 56.03] = 6.90; p = .01; Figs. 3 and 4), indicating that the group effect varied with time. For the 6MWT, simple effect testing revealed that the exercise group and the control group both had a statistically significant increase in the number of meters they walked comparing baseline with 12 weeks. At 12 weeks, the difference between the exercise group and the control group in the number of meters they could walk in 6 min approached statistical significance (F [1, 76.04] = 3.93; p = .051). This suggests that the exercise group had greater improvements in walk test scores over 12 weeks compared with the control group. For the Godin leisure time exercise questionnaire scores, simple effects testing indicated that for the exercise group, there was a statistically significant increase in their weekly exercise levels across the 12 weeks (F [1, 70.96] = 10.78; p = .002). In addition, at 12 weeks, there was a statistically significant difference between the exercise and the control group with respect to weekly exercise levels (F [1, 129.15] = 7.37; p = .008). Analysis using estimated marginal means showed the exercise group’s Godin leisure time exercise questionnaire score increased by 13.48, while the control group’s score decreased by 1.27 (Table 5). Effect sizes for the exercise intervention on the Godin leisure time exercise questionnaire scores were large (d = 0.80) and for the 6MWT were small (d = 0.21; Table 5). Fig. 3. View largeDownload slide Estimated marginal mean Godin Leisure Time Exercise Questionnaire (GLTEQ) score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Fig. 3. View largeDownload slide Estimated marginal mean Godin Leisure Time Exercise Questionnaire (GLTEQ) score across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Fig. 4. View largeDownload slide Estimated marginal mean 6-min walk test (6MWT) distance across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Fig. 4. View largeDownload slide Estimated marginal mean 6-min walk test (6MWT) distance across 12 weeks for exercise and control group. Error bars show 95% confidence interval. Other measures of physical fitness showed no group by time interaction effects. There was a main effect of time effect on the sit-to-stand test (F [1, 58.69] = 17.37; p = .001), with both groups showing improvements over time. Adherence to Intervention Physical activity levels (as determined from physical activity logs and class attendance) throughout the intervention showed the majority of participants achieved recommended levels of weekly aerobic activity. In addition, the majority were participating in resistance training exercises at least twice per week. The mean attendance (calculated as total number of classes attended out of a possible 24) across 12 weeks was 69%. Protocol adherence was monitored by the student co-investigator regularly attending exercise classes, through biweekly study meetings and through notification to the study team of potential protocol deviations. Harms There were no serious adverse events recorded for this trial. One participant had an exacerbation of a hiatus hernia. However, this participant was able to continue in the exercise class after clearance from his surgeon and modification to his strength training routine. Another participant had an exacerbation of a previous hip strain. After consultation with a physiotherapist, this participant decided to withdraw from the study. Discussion We conducted a randomized controlled trial intervention with cancer caregivers to determine whether exercise would improve self-reported quality of life, psychological health, measures of physical fitness, and physical activity levels. As there is limited research evaluating this type of intervention (i.e., a group-based structured aerobic and resistance training program) in cancer caregivers to date, our findings add some important new information to the current literature. For our primary outcome, the physical health component of quality of life, we found no differences between the exercise and the control group after 12 weeks. Several factors could account for this null finding. The physical health component of quality of life score is a composite of measures of physical functioning, impediments to one’s role based on poor health, bodily pain, and overall general health. The participants in our study had baseline physical health component scores that were higher than the expected population norms [39], had few medical morbidities, and had good overall health, meaning that further improvements to their physical component score would be difficult to detect (i.e., a ceiling effect). Despite the lack of statistically significant difference between the groups across 12 weeks, a small effect size of the intervention on the physical health component score was demonstrated. Northouse et al. [4] conducted a meta-analysis in 2010 and determined that caregiver interventions ranged between small and medium effect sizes. They argued that small effect sizes are in keeping with many interventions (physical activity and other types of support) tested in cancer patients and other chronic illness populations, and therefore, this effect should not be discounted as unimportant. Northouse et al. [4] also found that caregiver interventions delivered over longer periods of time generally showed greater effects. As we used perceived exertion versus objective measures (i.e., heart rate monitor) to measure exercise intensity, it could be that participants were not achieving high enough intensity levels of exercise to improve the physical health component. Finally, a larger sample size might have provided greater power to detect effects of exercise on the physical component score of quality of life. Contrary to the null findings on the physical health component score of quality of life, the exercise intervention improved caregiver’s mental health component of quality of life score. This is a noteworthy finding, given that the caregivers in our study had low baseline mental health component scores compared with population norms. Our findings are similar to previous research on physical activity interventions in various caregiver populations, where changes to mental health component scores have ranged between 2 and 8 points [13, 17–19]. Based on estimated marginal means, the difference in mental health component scores between the exercise and intervention group was approximately 6 points which is beyond the suggested threshold for minimally important difference of one half a standard deviation [67] (4.1 points) for the MCS score [39]. The medium effect size of the intervention on the mental health component scores is also an important indicator that the caregivers who participated in regular physical activity noticed meaningful improvements in the mental health component of their quality of life. Further substantiating this effect, our follow-up interviews with caregivers [22] highlighted that caregivers perceived these improvements to their mental well-being as an important benefit of physical activity. The caregivers in our study had poor psychological health at baseline as evidenced by high levels of depression and anxiety, and poor sleep quality. This adds further evidence to the growing body of literature attesting to the negative impacts of caregiving on psychological health. Irrespective of group assignment, we found significant time effects on measure of depression, state anxiety, and trait anxiety. In other words, time alone seemed to improve caregiver’s anxiety and depression. Badger et al. [16] reported similar “time-only” effects on depression in their exercise intervention with caregivers and cancer patients. It could be that simply being enrolled in an intervention for caregivers provided a sense of being supported and taking control, which in turn improved aspects related to mood. From the interviews we conducted with caregivers after they completed the exercise intervention [22], the recognition of caregivers needing help and their ability to take control were important components of improving their overall feelings of well-being. The intervention was effective at improving physical activity levels and physical fitness measures. This was demonstrated through self-reported physical activity levels and measures of aerobic physical fitness. Similar improvements in physical activity levels and fitness measures have been reported in other caregiver exercise trials [9, 11, 13]. Although the improvements to aerobic fitness likely do not represent a clinically meaning difference, we believe this is an important outcome given that there was a small effect on aerobic fitness after just 12 weeks of exercise. In addition to improvements in physical activity levels and physical fitness, the caregivers in our study also achieved recommended levels of physical activity on a weekly basis and attended classes regularly. These findings demonstrate that caregivers are able to participate in regular physical activity with the support of a group-based exercise class, and the benefits of this regular activity include improvements to overall level of weekly physical activity and aerobic fitness. This is an important finding considering the reported barriers caregivers face in participating in regular physical activity [6, 7] and their increased risk for morbidities that could be mitigated through regular exercise. Improvements to physical activity levels could also stem from the social support gained from the group-based format of the class. It is well documented that social support is an important component of physical activity behavior [68]. In our qualitative interviews with caregivers, the group environment was consistently noted as an important component of the intervention [22]. Despite improvements to physical activity levels and aerobic fitness, there were no improvements to strength after 12 weeks of regular resistance training. This null finding could be explained by the following factors. As the participants had average strength at baseline, gains to their strength would be slower than if their strength at baseline would have been low [66]. In addition, many of the participants had no history of performing the strength training exercises, which meant that during the first 3–4 weeks of the program very little resistance was used. Gains in muscle strength are dependent on the overload principle whereby muscles need to be exercised at a workload that is greater than normal [66]. The portion of the time that participants spent overloading their muscles (i.e., 7–8 weeks) was likely not long enough to see strength gains. Limitations To our knowledge, this is the first randomized controlled trial in cancer caregivers to test the effects of a group-based aerobic and resistance training exercise program on physical and psychological outcomes. Although our findings add new information to the literature, some limitations must be considered. Common to most behavioral interventions, participants were not blinded to their treatment allocation that may lead to performance bias. To minimize other sources of bias, including detection bias, we blinded the fitness assessors. Our sample of caregivers had high socioeconomic status, and the participants were generally physically healthy. This may not be representative of the population of caregivers and therefore limits the generalizability of our findings. We used self-report physical activity logs to collect weekly physical activity in addition to class attendance. Future research on exercise interventions in caregivers could include accelerometer data to provide more accurate collection of physical activity levels. Only 56% of the participants returned 12 weeks of completed physical activity logs, while the remaining participants only partially completed their logs. Thus, our estimation of weekly physical activity levels is conceivably an overestimation, as those participants who completed all the logs were also likely to be highly adherent to the intervention. The participants who dropped out of the study did so predominantly because of being overwhelmed with their caregiving responsibilities. Finally, the broad age range of participants might have made it more difficult to detect an effect on some of the outcomes including the primary outcomes and social support. A limitation of this study was our inability to adequately capture the extent of the caregiving experience with respect to the nature and intensity of caregiving. This stems from lack of validated instruments to comprehensively measure this construct, our efforts to try to balance participant burden in the number of questionnaires they were completing, and the recognition that the caregiving experience is largely dependent on many personal, social, and situational factors [69]. We measured caregiving hours per week as a crude estimate of total caregiving responsibilities. We then controlled for caregiving hours per week as a crude estimate of changes that may have occurred to the caregiving status throughout the study. Other factors related to the caregiving situation, such as stage of illness and health of the care recipient or functional status (e.g., Eastern Cooperative Oncology Group performance status), an inventory of caregiving responsibilities, and other dependents, might also be important factors to measure and control for in future caregiver studies. In addition, although we controlled for caregiving hours per week, it should be noted that the caregivers in our study could have been less overwhelmed than the general population of caregivers. It should be noted that the results from this trial are specific to the design of the intervention, which included a group-based exercise and education program. It is difficult to attribute the results from the trial to exercise alone. We included a self-reported social support outcome measure in an attempt to try to examine whether there was a social component of the group-based program. We did not find any changes to social support, however, other measures of the group effect (e.g., social cohesion, role modelling, and sense of belonging) may also have been contributory. Implications for Clinical Practice and Future Research Directions Findings from this study suggest that a group-based exercise program for cancer caregivers may be effective at improving aspects of quality of life, physical fitness, and physical activity levels. The addition of physical activity programs to the range of supportive services offered to caregivers should be considered as an important additional mechanism in supporting health and well-being in this invaluable and vulnerable group. Our qualitative analysis of the impact of this intervention [22] also highlighted the perceived value of exercise as an intervention for caregivers and the need for more choice in the type of supports that are currently offered to caregivers. The time-only effects that were observed in many of the outcomes suggest that recognition of and attention to caregivers was an important component of the study (i.e., Hawthorne effect). Indeed, the caregivers we interviewed all thanked the student co-investigator for designing and conducting a study to support caregivers and suggested that recognition of caregiving by health care providers and researchers is an important supportive mechanism. Building on this work and previous research evaluating exercise in caregivers, the following future research directions could be considered. The baseline physical health status of our participants was better than we expected. Health status and functional level at baseline may be an important indicator of the type of participant who would self-select to be in an exercise study. Those caregivers with poor overall health, multiple morbidities, or poor functional level may not volunteer to be in an exercise study, which paradoxically is the type of participant that such an intervention should be targeting. Although we had the capacity to accommodate those caregivers who were older and in poorer health, we did not generally attract these types of caregivers to our study. Recruitment efforts to target caregivers in poorer health, those with lower socioeconomic status, or minority populations, and exercise interventions designed to be flexible to accommodate lower baseline functional levels would be important considerations for future research. We found the mental health component of quality of life improved; however, other measures of mental health such as depression and anxiety did not improve significantly. Future research could evaluate the impact of psychological support and exercise combined (e.g., walk and talk, or combined exercise and support group). Given the high psychological morbidity in our sample, exercise in addition to psychological support may be more effective at improving overall mental well-being. In the meta-analysis by Northouse et al. [4], caregivers interventions that had a combination of the types of support offered showed greater effect sizes than single types of support. In addition, our qualitative findings [22] suggested that other indicators of caregiver well-being, such as positivity, life satisfaction, and feelings of control, may be important outcomes to measure to determine the impact of exercise on caregivers’ psychological well-being. Determination of sustainable exercise programming, such as at community centers or cancer-specific agencies, would also be an important next step. Because this work was part of a doctoral research program that had limited funding and time constraints, the length of our intervention was likely too short. To provide more robust estimation of the effects of exercise and to address questions about maintenance of exercise behaviors, long-term programs should be the goal. Researchers partnering with community agencies where exercise programs for cancer patients are already being offered, or partnering with caregivers using participatory action research, would be a logical next step. Finally, we intentionally designed a group-based exercise and education program for caregivers based on theories of behavior change and the well-documented effect of social support in exercise adherence. Given caregivers overwhelming time demands and potential inability to leave their care recipient, it would be important in future work to examine home-based versus group-based exercise program to determine whether a balance between feasibility and outcomes could be achieved. Acknowledgements This study was funded by The Canadian Institutes of Health Research (CIHR) Doctoral research award, The Izaak Walton Killam scholarship, and the Psychosocial Oncology Research Training program. Compliance with Ethical Standards Authors’ Statement of Conflict of Interest and Adherence to Ethical Standards Authors Colleen A. Cuthbert, Kathryn King-Shier,Dean Ruether, Dianne Tapp, Kathryn Wytsma-Fisher, Tak S. Fung, and S. Nicole Culos-Reed declare that they have no conflict of interest. All procedures, including the informed consent process, were conducted in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Ethical Approval This research involved human participants. Ethics approval from the local governing ethics review board was obtained prior to any contact with participants. Informed Consent All participants signed an informed consent form. References 1. Sinha M . Spotlight on Canadians, Results from the General Social Survey: Portrait of Caregivers . Statistics Canada; 2012 . Available at http://www.statcan.gc.ca/pub/89-652-x/89-652-x2013001-eng.pdf. Accessibility verified July 2016 . 2. National Alliance for Caregiving/AARP . 2015 Report: Caregiving in the US . Available at http://www.caregiving.org/caregiving2015/. Accessibility verified July 25, 2016. 3. Northouse L , Williams AL , Given B , McCorkle R . Psychosocial care for family caregivers of patients with cancer . J Clin Oncol . 2012 ; 30 ( 11 ): 1227 – 1234 . Google Scholar CrossRef Search ADS PubMed 4. Northouse LL , Katapodi MC , Song L , Zhang L , Mood DW . Interventions with family caregivers of cancer patients: Meta-analysis of randomized trials . CA Cancer J Clin . 2010 ; 60 ( 5 ): 317 – 339 . Google Scholar PubMed 5. Stenberg U , Ruland CM , Miaskowski C . Review of the literature on the effects of caring for a patient with cancer . Psychooncology . 2010 ; 19 ( 10 ): 1013 – 1025 . Google Scholar CrossRef Search ADS PubMed 6. Beesley VL , Price MA , Webb PM ; Australian Ovarian Cancer Study Group; Australian Ovarian Cancer Study—Quality of Life Study Investigators . Loss of lifestyle: health behaviour and weight changes after becoming a caregiver of a family member diagnosed with ovarian cancer . Support Care Cancer . 2011 ; 19 ( 12 ): 1949 – 1956 . Google Scholar CrossRef Search ADS PubMed 7. Rha SY , Park , Y , Song SK , Lee CE , Lee J . Caregiver burden and health-promoting behaviors among the family caregivers of cancer patients . Eur J Oncol Nurs . 2015 ; 19 : 174 – 181 . Google Scholar CrossRef Search ADS PubMed 8. Kent EE , Rowland JH , Northouse L et al. Caring for caregivers and patients: Research and clinical priorities for informal cancer caregiving . Cancer . 2016 ; 122 ( 13 ): 1987 – 1995 . Google Scholar CrossRef Search ADS PubMed 9. Connell CM , Janevic MR . Effects of a telephone-based exercise intervention for dementia caregiving wives: A randomized controlled trial . J Appl Gerontol . 2009 ; 28 ( 2 ): 171 – 194 . Google Scholar CrossRef Search ADS PubMed 10. Hirano A , Suzuki Y , Kuzuya M , Onishi J , Ban N , Umegaki H . Influence of regular exercise on subjective sense of burden and physical symptoms in community-dwelling caregivers of dementia patients: A randomized controlled trial . Arch Gerontol Geriatr . 2011 ; 53 ( 2 ): e158e163 . Google Scholar CrossRef Search ADS 11. King AC , Baumann K , O’Sullivan P , Wilcox S , Castro C . Effects of moderate-intensity exercise on physiological, behavioral, and emotional responses to family caregiving: A randomized controlled trial . J Gerontol A Biol Sci Med Sci . 2002 ; 57 ( 1 ): M26 – M36 . Google Scholar CrossRef Search ADS PubMed 12. Lowery D , Cerga-Pashoja A , Iliffe S et al. The effect of exercise on behavioural and psychological symptoms of dementia: The EVIDEM-E randomised controlled clinical trial . Int J Geriatr Psychiatry . 2014 ; 29 ( 8 ): 819 – 827 . Google Scholar CrossRef Search ADS PubMed 13. Hill K , Smith R , Fearn M , Rydberg M , Oliphant R . Physical and psychological outcomes of a supported physical activity program for older carers . J Aging Phys Act . 2007 ; 15 ( 3 ): 257 – 271 . Google Scholar CrossRef Search ADS PubMed 14. Danuculov MAD , Kozasa EH , Ribas KT , Galduroz JCF , Garcia MC , Verreschi ITN . A yoga and compassion meditation program reduce stress in familial caregivers of Alzheimer’s disease patients . eCAM . 2013 ; doi: 10.1155/2013/513149 . 15. Waelde LC , Thompson L , Gallagher-Thompson D . A pilot study of a yoga and meditation intervention for dementia caregiver stress . J Clin Psychol . 2004 ; 60 ( 6 ): 677 – 687 . Google Scholar CrossRef Search ADS PubMed 16. Badger T , Segrin C , Dorros SM , Meek P , Lopez AM . Depression and anxiety in women with breast cancer and their partners . Nurs Res . 2007 ; 56 ( 1 ): 44 – 53 . Google Scholar CrossRef Search ADS PubMed 17. Martin AC , Keats MR . The impact of yoga on quality of life and psychological distress in caregivers for patients with cancer . Oncol Nurs Forum . 2014 ; 41 ( 3 ): 257 – 264 . Google Scholar CrossRef Search ADS PubMed 18. Milbury K , Mallaiah S , Lopez G et al. Vivekananda yoga program for patients with advanced lung cancer and their family caregivers . Integr Cancer Ther . 2015 ; 14 ( 5 ): 446 – 451 . Google Scholar CrossRef Search ADS PubMed 19. Winters-Stone KM , Lyons KS , Dobek J et al. Benefits of partnered strength training for prostate cancer survivors and spouses: Results from a randomized controlled trial of the Exercising Together project . J Cancer Surviv . 2016 ; 10 ( 4 ): 633 – 644 . Google Scholar CrossRef Search ADS PubMed 20. Cuthbert CA , King-Shier K , Ruether D , Tapp DM , Culos-Reed SN . What is the state of the science on physical activity interventions for family caregivers? A systematic review and RE-AIM evaluation . J Phys Act Health . 2017 ; 14 ( 7 ): 578 – 595 . Google Scholar CrossRef Search ADS PubMed 21. Cuthbert CA , King-Shier K , Tapp DM , Ruether D , Jackson C , Culos-Reed SN . Renewing caregiver health and wellbeing through exercise (RECHARGE): A randomized controlled trial . Contemp Clin Trials . 2016 ; 50 : 273 – 283 . Google Scholar CrossRef Search ADS PubMed 22. Cuthbert CA , Culos-Reed SN , King-Shier K , Ruether D , Bischoff MB , Tapp DM . Creating an upward spiral: A qualitative study of caregivers’ experience of participating in a structured physical activity program . Eur J Cancer Care . 2017 ;26(6): e12684 . 23. Ware JE Jr , Sherbourne CD . The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection . Med Care . 1992 ; 30 ( 6 ): 473 – 483 . Google Scholar CrossRef Search ADS PubMed 24. National Cancer Institute . Coping with cancer for family and friends . Available at www.cancer.gov/about-cancer/coping/family-friends. Accessibility verified July 2016 . 25. Canadian Society for Exercise Physiology . Canadian physical activity guidelines 2012 . Available at www.csep.ca. Accessibility verified July 28, 2015 . http://dx.doi.org/10.1249/MSS.0b013e318213fefb. 26. Cuthbert CA , King-Shier K , Tapp D , Ruether D , Culos-Reed SN . Exploring gender differences in self-reported physical activity and health among older caregivers . Oncol Nurs Forum . 2017 ; 44 ( 4 ): 435 – 445 . Google Scholar CrossRef Search ADS PubMed 27. Warburton DER , Jamnik VK , Bredin SSD , Gledhill N . On behalf of the PAR-Q+ Collaboration . The physical activity readiness questionnaire for everyone (PARQ+) and electronic physical activity readiness medical examination (ePARmed-X+) . Health Fitness J Canada . 2011 ; 4 ( 2 ): 3 – 23 . 28. Garber CE , Blissmer B , Deschenes R et al. American College of Sports Medicine Position Stand . Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise . Med Sci Sports Exerc . 2011; 43 ( 7 ): 1334 – 1359 . CrossRef Search ADS PubMed 29. Borg GA . Psychophysical bases of perceived exertion . Med Sci Sports Exerc . 1982 ; 14 ( 5 ): 377 – 381 . Google Scholar CrossRef Search ADS PubMed 30. Bandura A . The primacy of self-regulation in health promotion . Appl Psychol . 2005 ; 54 ( 2 ): 245 – 254 . Google Scholar CrossRef Search ADS 31. Schwarzer R . Social-cognitive factors in changing health-related behaviors . Curr Dir Psychol Sci . 2001 ; 10 ( 2 ): 47 – 51 . Google Scholar CrossRef Search ADS 32. Buysse DJ , Reynolds CF III , Monk TH , Berman SR , Kupfer DJ . The Pittsburgh Sleep Quality Index: A new instrument for psychiatric practice and research . Psychiatry Res . 1989 ; 28 ( 2 ): 193 – 213 . Google Scholar CrossRef Search ADS PubMed 33. King MB , Whipple RH , Gruman CA , Judge JO , Schmidt JA , Wolfson LI . The Performance Enhancement Project: Improving physical performance in older persons . Arch Phys Med Rehabil . 2002 ; 83 ( 8 ): 1060 – 1069 . Google Scholar CrossRef Search ADS PubMed 34. Carpenter JS , Andrykowski MA . Psychometric evaluation of the Pittsburgh Sleep Quality Index . J Psychosom Res . 1998 ; 45 ( 1 ): 5 – 13 . Google Scholar CrossRef Search ADS PubMed 35. Radloff LS . The CES-D scale: A self report depression scale for research in the general population . Appl Psychol Meas . 1977 ; 1 : 385 – 401 . Google Scholar CrossRef Search ADS 36. Pinquart M , Sorensen S . Differences between caregivers and non-caregivers in psychological health and physical health: A meta-analysis . Psychol Aging . 2003 ; 18 ( 2 ): 250 – 267 . Google Scholar CrossRef Search ADS PubMed 37. Pinquart M , Sörensen S . Helping caregivers of persons with dementia: Which interventions work and how large are their effects ? Int Psychogeriatr . 2006 ; 18 ( 4 ): 577 – 595 . Google Scholar CrossRef Search ADS PubMed 38. Turner-Bowker DM , Bartley PJ , Ware JE. SF-36 Health Survey & SF Bibliography: (1988–2000) . Lincoln, RI : QualityMetric Inc .; 2002 . 39. Hopman WM , Towheed T , Anastassiades T et al. Canadian normative data for the SF-36 health survey . Can Med Assoc J . 2000 ; 162 ( 3 ): 265 – 271 . 40. Spielberger CD. State-Trait Anxiety Inventory: Bibliography . 2nd ed . Palo Alto, CA : Consulting Psychologists Press ; 1989 . 41. Zimet GD , Powell SS , Farley GK , Werkman S , Berkoff KA . Psychometric characteristics of the Multidimensional Scale of Perceived Social Support . J Pers Assess . 1990 ; 55 ( 3–4 ): 610 – 617 . Google Scholar CrossRef Search ADS PubMed 42. Godin G , Shephard RJ . A simple method to assess exercise behavior in the community . Can J Appl Sport Sci . 1985 ; 10 ( 3 ): 141 – 146 . Google Scholar PubMed 43. Jacobs DR Jr , Ainsworth BE , Hartman TJ , Leon AS . A simultaneous evaluation of 10 commonly used physical activity questionnaires . Med Sci Sports Exerc . 1993 ; 25 ( 1 ): 81 – 91 . Google Scholar CrossRef Search ADS PubMed 44. Canadian Society for Exercise Physiology . Physical Activity Training for Health (PATH): CSEP − Health and Fitness Program’s Health-Related Appraisal and Counselling Strategy . Ottawa, Ontario, Canada: Canadian Society for Exercise Physiology ; 2013 . 45. Rikli R , Jones JC. Senior Fitness Test Manual . 2nd ed . Windsor, Ontario, Canada: Human Kinetics Inc.; 2002 . 46. Burr JF , Bredin SS , Faktor MD , Warburton DE . The 6-minute walk test as a predictor of objectively measured aerobic fitness in healthy working-aged adults . Phys Sportsmed . 2011 ; 39 ( 2 ): 133 – 139 . Google Scholar CrossRef Search ADS PubMed 47. Rantanen T , Guralnik JM , Foley D et al. Midlife hand grip strength as a predictor of old age disability . JAMA . 1999 ; 281 ( 6 ): 558 – 560 . Google Scholar CrossRef Search ADS PubMed 48. Jackson A , Langford NJ . The criterion-related validity of the sit and reach test: Replication and extension of previous findings . Res Q Exerc Sport . 1989 ; 60 ( 4 ): 384 – 387 . Google Scholar CrossRef Search ADS PubMed 49. Lemmink KA , Kemper HC , de Greef MH , Rispens P , Stevens M . The validity of the sit-and-reach test and the modified sit-and-reach test in middle-aged to older men and women . Res Q Exerc Sport . 2003 ; 74 ( 3 ): 331 – 336 . Google Scholar CrossRef Search ADS PubMed 50. Eyigor S , Karapolat H , Durmaz B . Effects of a group-based exercise program on the physical performance, muscle strength and quality of life in older women . Arch Gerontol Geriatr . 2007 ; 45 ( 3 ): 259 – 271 . Google Scholar CrossRef Search ADS PubMed 51. Hand BD , Cavanaugh S , Forbes W , Govern J , Cress E . Changes in health related quality of life and functional fitness with exercise training in older adults who attend senior centers . Activit Adapt Aging . 2012 ; 36 : 29 – 54 . 52. Iversen MD , Fossel AH , Katz JN . Enhancing function in older adults with chronic low back pain: A pilot study of endurance training . Arch Phys Med Rehabil . 2003 ; 84 ( 9 ): 1324 – 1331 . Google Scholar CrossRef Search ADS PubMed 53. Li WC , Chen YC , Yang RS , Tsauo JY . Effects of exercise programmes on quality of life in osteoporotic and osteopenic postmenopausal women: A systematic review and meta-analysis . Clin Rehabil . 2009 ; 23 ( 10 ): 888 – 896 . Google Scholar CrossRef Search ADS PubMed 54. Whitehurst MA , Johnson BL , Parker CM , Brown LE , Ford AM . The benefits of a functional exercise circuit for older adults . J Strength Cond Res . 2005 ; 19 ( 3 ): 647 – 651 . Google Scholar PubMed 55. Brant R . Inference for means: comparing two independent samples . Available at http://stat.ubc.ca/~rollin/. Accessibility verified July 25, 2016 . 56. Research randomizer . Available at www.randomizer.org. Accessibility verified April 30, 2015. 57. Freidman LM , Furberg CD , DeMets DL. Fundamentals of Clinical Trials . 3rd ed . New York, NY : Springer ; 1998 . Google Scholar CrossRef Search ADS 58. Quene H , van der Bergh H . On multi-level modeling of data from repeated measures designs: A tutorial . Speech Commun . 2004 ; 43 : 103 – 121 . Google Scholar CrossRef Search ADS 59. West BT . Analyzing longitudinal data with the linear mixed models procedure in SPSS . Eval Health Prof . 2009 ; 32 ( 3 ): 207 – 228 . Google Scholar CrossRef Search ADS PubMed 60. Cohen J. Statistical Power Analysis for the Behavioral Sciences . 2nd ed . New York, NY: Lawrence Erlbaum Associates ; 1988 . 61. Maas CJM , Hox JJ . Sufficient sample sizes for multilevel modeling . Methodology . 2005 ; 1 : 85 – 92 . Google Scholar CrossRef Search ADS 62. Schulz KF , Altman DG , Moher D ; CONSORT Group . CONSORT 2010 statement: Updated guidelines for reporting parallel group randomized trials . Ann Intern Med . 2010 ; 152 ( 11 ): 726 – 732 . Google Scholar CrossRef Search ADS PubMed 63. Public Health Agency of Canada . Fact sheet: Government of Canada chronic disease initiatives . Ottawa, ON ; 2011 . Available at http://www.phac-aspc.gc.ca. Accessibility verified July 2016 . PubMed PubMed 64. Godin G . The Godin – Shepard leisure time physical activity questionnaire . Health Fitness J Canada . 2011 ; 4 ( 1 ): 18 – 22 . 65. Camarri B , Eastwood PR , Cecins NM , Thompson PJ , Jenkins S . Six minute walk distance in healthy subjects aged 55 – 75 years . Respir Med . 2006 ; 100 ( 4 ): 658 – 665 . Google Scholar CrossRef Search ADS PubMed 66. Heyward VH. Advanced Fitness Assessment and Exercise Prescription . 6th ed . Champaign, IL : Human Kinetics ; 2010 . 67. Norman GR , Sloan JA , Wyrwich KW . Interpretation of changes in health-related quality of life: The remarkable universality of half a standard deviation . Med Care . 2003 ; 41 ( 5 ): 582 – 592 . Google Scholar PubMed 68. Bandura A . Health promotion by social cognitive means . Health Educ Behav . 2004 ; 31 ( 2 ): 143 – 164 . Google Scholar CrossRef Search ADS PubMed 69. Fletcher BS , Miaskowski C , Given B , Schumacher K . The cancer family caregiving experience: An updated and expanded conceptual model . Eur J Oncol Nurs . 2012 ; 16 ( 4 ): 387 – 398 . Google Scholar CrossRef Search ADS PubMed © Society of Behavioral Medicine 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Annals of Behavioral MedicineOxford University Press

Published: Jan 25, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off