Influence of Lifestyle Redesign ® on Health,Social Participation,Leisure,and Mobility of Older French-Canadians

Study Design and Participants

This pilot study used a mixed-methods concurrent triangulation design (Creswell & Plano Clark, 2017) including a preexperimental component with pretest (T₁), posttest (T₂), and 3-mo (T₃) and 6-mo (T₄) follow-ups and an exploratory descriptive qualitative clinical study (Miller & Crabtree, 2003). The sample included 16 community-dwelling older adults with and without disability. A sample size of 16 participants allowed detection of a standardized difference of .75 or greater between two means according to paired bilateral t tests based on a significance level of .05 and power of .80 (Machin et al., 2009). This difference was considered sufficient in studies that explored the influence of another intervention on leisure (Levasseur et al., 2016) and life-space mobility (Pigeon et al., in press). This sample size also allowed in-depth exploration and data saturation.

Eligibility criteria were (1) age 65 yr and older, (2) no or mild (group without disability) or moderate or severe (group with disabilities) loss of autonomy, (3) normal cognitive functions, (4) residence in a conventional or residential home for semi-independent older adults, and (5) French speaking. Participants were recruited from a previous study of people attending a day hospital and day center in a Health and Social Services Centre (HSSC) in Quebec (Canada) and from people living in a senior residence. The research ethics committee of the Eastern Townships HSSC approved the study (2015–488).

Data Collection Procedures

Participants were recruited until the predetermined sample size (n = 16 + 3, anticipating possible attrition) was reached. All participants signed an informed consent form and were met individually at home by a research assistant or occupational therapy student specially trained to administer the questionnaires. An experienced research assistant (Maryke Beaudry) conducted the qualitative interviews. At T₁, one sociodemographic and eight outcome questionnaires, four of which are reported here and the others elsewhere (Trépanier et al., 2019), were administered in approximately 120 min. At the end of the 6-mo intervention period (T₂), participants answered the same outcome questionnaires and, about 1 mo later, had a face-to-face semidirected individual interview lasting about 90 min. All interviews were digitally audiotaped, transcribed, and verified with respect to the wording used by participants. After the first few interviews, two authors (Maryke Beaudry and Mélanie Levasseur) discussed and adjusted the questions for subsequent interviews. Finally, 3 mo (T₃) and 6 mo (T₄) after the end of the intervention, participants answered the same questionnaires again.

Intervention

The French-Canadian Lifestyle Redesign intervention was led by an occupational therapist who took the University of Southern California 6-hr online introductory training course. This therapist was supervised on a weekly basis by an academic occupational therapist (Mélanie Levasseur) specializing in health promotion and clinical research who was familiar with the intervention (and very involved in the translation). Through this supervision, the intervening occupational therapist received regular feedback on her role and the intervention. Weekly 2-hr group sessions were held over a 6-mo period between August 2015 and March 2016. These sessions were based on 12 modules (e.g., occupation, health, and aging; transportation and occupation) from the second edition of the Lifestyle Redesign manual (Clark et al., 2015) and involved didactic presentations, peer exchanges, reflective exercises, direct experience, and personal exploration (Carlson et al., 1998). Every month, one group outing and individual meetings with the occupational therapist were planned. The meetings aimed to help participants integrate the group session content and engage in personalized meaningful activities. In the group sessions for participants with moderate or severe loss of autonomy, the occupational therapist was assisted by one or two volunteers.

Outcome Variables and Measures

Data on health, social participation, leisure, and life-space mobility were collected with four questionnaires. The 36-item Short Form Health Survey (SF–36) consists of 36 items covering eight domains related to physical and mental health: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional, and mental health (Ware et al., 2000). Higher scores indicate greater frequency of social participation. Widely used in research, including previous Lifestyle Redesign studies, the SF–36 has good psychometric properties, including good internal consistency (Cronbach’s αs of .83–.93 for the eight domains, .94 for the physical component summary, and .89 for the mental component summary; Gandek et al., 2004) and good test–retest reliability, even for testing after 6 mo (correlation coefficients of .60–.90 except for bodily pain, .43; Ware, 2000). The SF–36 is also sensitive to change (Gatchel et al., 1999), with a difference of 5 points in scale scores being clinically significant (Ware et al., 1993).

The Social Participation Scale measures the frequency of participation in 10 community activities and has good internal consistency (Cronbach’s αs of .85–.91; Richard et al., 2009). Higher scores indicate greater frequency of social participation; a difference of 1 point on each activity score is considered clinically significant. The Leisure Profile assesses involvement in leisure activities, attitudes toward leisure, and difficulties that might influence leisure activities; higher scores reflect greater leisure participation. It has acceptable interrater (κ = .21–.80) and test–retest (.41–.60) reliability (Dutil et al., 2007). The Life-Space Assessment (LSA) measures life-space mobility, specifically the range, independence, and frequency of movement over the preceding 4 wk; higher scores indicate better mobility (Baker et al., 2003). The LSA has shown excellent test–retest reliability (intraclass correlation coefficient = .87) and moderate to substantial concordance for 18 of 20 items (κ = .47–.73; Auger et al., 2009). It has good construct validity with observed physical performance and self-reported function (95% confidence interval [.82, .97]) and good sensitivity to change (Baker et al., 2003). Finally, a semistructured interview guide (see Supplemental Appendix A, available at the end of this document) validated by five qualitative research experts and pretested was used to explore the self-perceived effect of the adapted Lifestyle Redesign on participants.

Data Analysis

To foster transferability (Laperrière, 1997), the participants’ sociodemographic characteristics and outcomes were analyzed using descriptive statistics. Scores were compared with the Friedman test followed by the Wilcoxon signed-rank test for all participants and, in an exploratory manner, each group separately. Because of the exploratory nature of this study and the influence of seasonal variations on Quebecers’ health, social participation, leisure, and mobility, changes at any of the postintervention measurement times with p < .05 were considered to be potentially attributable to the intervention.

Interview transcripts underwent thematic content analysis using mix extraction grids (Miles et al., 2014). The data analysis involved (1) verbal data collection, (2) reading of data, (3) division of data into units of sense, (4) organization and reformulation of original data in disciplinary terminology, and (5) synthesis of results. Themes that emerged from the interview content were organized and renamed according to the Human Development Model–Disability Creation Process (HDM–DCP; Figure 1), a model of human development and disability (Fougeyrollas, 2010). The HDM–DCP illustrates interactions among intrinsic personal factors, extrinsic environmental factors, and participation (operationalized as life habits). To increase credibility, reliability, and confirmability (Laperrière, 1997), the first author (Mélanie Levasseur) co-coded one-third of the data, which had first been exhaustively analyzed by a specially trained research assistant (Maryke Beaudry). Levasseur closely supervised the analysis, making adjustments until consensus was reached regarding the participants’ perceptions of the intervention. Additional memos describing the thoughts, questions, and discussions of the research team were used. Analyses were conducted using IBM SPSS Statistics (Version 18; IBM Corp., Armonk, NY) or NVivo (Version 10; QSR International, Melbourne, Victoria, Australia).

OPEN IN VIEWER

Figure 1.

Human Development Model and Disability Creation Process (HDM–DCP) illustrating interactions among intrinsic personal factors, extrinsic environmental factors, and life habits.

Participants and Intervention

Of the 19 participants assessed at T₁, 1 had serious health problems preventing participation in the intervention and 1 died, leaving 17 older adults who completed the program and were interviewed (Table 1). Because 1 participant (Participant [P] 17) had vision and hearing problems that impeded questionnaire completion, only 16 were reassessed. At baseline, the 3 nonparticipants were not different from those who participated except for older age (p < .01), greater loss of autonomy (p = .01), and smaller life-space mobility (p = .047).

Participants were aged 65–90 yr (mean [M] ± standard deviation [SD] = 76.4 ± 7.6; median [Mdn] ± semi-interquartile interval [Q] = 74.0 ± 5.8). All were Caucasian, the majority were women (n = 10; 62.5%) and owners (n = 5; 31.3%) or tenants (n = 7; 43.8%) of their dwelling, and nearly half lived alone (n = 7; 43.8%). Half had 12 or more years of schooling (n = 8; 50.0%), and most had family income under CAN$40,000 (n = 12; 75.0%) and rated their health as good or excellent (n = 12; 75.0%; Table 1). Two groups as homogeneous as possible were created, one with 7 participants with disabilities and one with 10 participants without disability.

Five participants with disabilities lived in the same residence where the group meetings were held. The older adults participated in an average of about 25 group meetings with the occupational therapist (M ± SD = 24.3 ± 2.2; Mdn ± Q = 25.0 ± 1.5), which amounted to 90% or more of the sessions and went on an average of four or five outings (e.g., restaurant, market, museum). Reasons for missing group meetings were most often being ill, working, or having an appointment. The participants attended 5–11 individual meetings with the occupational therapist (M ± SD = 6.1 ± 0.6; Mdn ± Q = 6.0 ± 0.0).

Participants reported mostly positive effects from the program, sometimes no effect, but rarely negative effects on their personal and environmental factors and social participation (Figure 2). The program fostered participants’ knowledge about health, social participation, leisure, and mobility. This knowledge stimulated them and, depending on their personal factors and with a safer and mobilized environment, encouraged their efforts to take action. With regard to personal factors, participants reported that the French-Canadian Lifestyle Redesign program modified their vision of themselves and others and empowered them. This vision and empowerment bilaterally influenced their willingness to act, which in turn similarly modified their health and relationship skills. In terms of interaction between personal and environmental factors, by facing challenges and taking action, the participants reported that they improved their social participation, leisure, and mobility. Social participation improvement included increases in health habits, activities in the community or with others, social interactions both within and outside the Lifestyle Redesign program, and, during interactions with others, mobility.

OPEN IN VIEWER

Figure 2.

Participant-reported influence of the French-Canadian Lifestyle Redesign^® program.

Health

Pre- and postintervention comparisons showed that for both groups as a whole, participants’ health did not change, but the mental component improved between T₃ and T₄ (Table 2). Surprisingly, physical role decreased between T₂ and T₃, indicating that participants’ physical health further affected time spent, accomplishment, and difficulties in daily activities. Although not statistically significant, emotional role, absence of pain, social functioning, and mental health scores showed a tendency to increase. According to group results for participants with disabilities, a decrease was observed in the physical component (T₂, Mdn ± Q = 29.1 ± 3.0 vs. T₃, 25.7 ± 4.0, p = .046), physical functioning (T₂, 7.5 ± 8.1, vs. T₃, 2.5 ± 6.3, p = .03, and T₄, 0.0 ± 3.8, p = .04), and physical role (T₂, 75.0 ± 19.5, vs. T₃, 31.3 ± 26.6, p = .04) but an improvement in pain (T₃, 46.0 ± 12.4, vs. T₄, 56.0 ± 13.8, p = .03). In participants without disability, the mental component increased (T₃, 54.5 ± 5.0, vs. T₄, 57.8 ± 4.3, p = .04), as did general health (T₁, 84.5 ± 10.9, vs. T₄, 87.0 ± 15.4, p = .048).

The majority of participants reported improved mental health (Figure 2): “I feel better, less depressed” (P12) and “[The program] makes me want to enjoy life again” (P10). Reduction in symptoms (e.g., stiffness) was also reported; one man with disabilities explained that during the group, “My legs hurt, but it was OK. . . . If something interested me and I liked it enough, I didn’t feel the pain” (P11). Participants without disability attributed positive health effects mainly to better health habits, such as increased physical activity.

Social Participation

Social participation did not change significantly after the intervention for participants (Table 2) but increased for those with disabilities (T₁, Mdn ± Q = 2.5 ± 7.5, vs. T₂, 7.0 ± 11.9, p = .03). Although not significant, a tendency toward improvement was also observed in participants without disability (T₁, 21.0 ± 9.0, vs. T₂, 26.0 ± 7.4; T₃, 27 ± 7.4; and T₄, 28.5 ± 6.5; ps = .14–.51).

Several participants reported increased activities in the community or with others (Figure 2); as 1 participant explained, “We take more time. . . . We go to the restaurant, visit people, act like good neighbors” (P10). Others resumed previous activities or started new ones. After an individual session, 1 participant reported, “It induced me to find volunteer work that I like” (P9). Although the program encouraged participants to act—“I do more things now because I order myself to do something; before the program, I just sat here, in my armchair, waiting for time to go by” (P12)—being more active did not always transform into changes in activities. Some participants faced participation challenges, and others struggled with perseverance in experimenting or searching for meaningful activities. Nevertheless, meeting people, benefiting from exchanges with others, and contributing to the group were among the important contributions of the program: “I get a lot out of being with others and being able to chat” (P4). It also helped optimize interpersonal relationships: “[The program] made me want to be more open, to socialize” (P13).

Leisure

Pre- and postintervention comparisons showed that participants’ leisure did not change except for an increase in interest between T₁ and T₂ (Table 2). Frequency of activities decreased between T₂ and both T₃ and T₄, as did the desire to modify leisure practices between T₂ and T₄. Impairments increased between T₁ and T₂ but decreased between T₂ and T₄ (Table 2). Finally, there were fewer physical environment obstacles at both T₁ and T₂ than at T₄, but social environment obstacles decreased between T₁ and both T₃ and T₄. Results for participants without disability revealed an increase in interest in leisure activities (T₁, Mdn ± Q = 23.0 ± 1.8, vs. T₂, 25.0 ± 1.4, p = .02) but a decrease in frequency (T₂, 21.5 ± 2.1, vs. T₃, 19.0 ± 1.3, p = .03). The desire to modify leisure practice also decreased for this group (T₁, 17.1 ± 1.3, vs. T₄, 15.0 ± 1.1, p = .01; T₂, 17.5 ± 1.7, vs. T₃, 15.5 ± 1.4, p = .049, and T₄, 15.5 ± 1.4, p = .01). Impairments in participants without disability also changed over time (T₂, 6.0 ± 3.0, vs. T₁, 3.0 ± 2.6, p = .02; T₃, 2.5 ± 2.1, p = .02; and T₄, 2.0 ± 2.4, p =.03). In participants with disabilities, positive attitudes toward leisure increased after the intervention (T₁, 12.5 ± 1.4 vs. T₃, 15.5 ± 1.5, p = .04).

Although some participants reported no change in leisure, others planned to do more or actually increased the frequency of activities—for example, physical exercise such as walking regularly (Figure 2): “I started doing it again. I walk for an hour or hour and a half every morning and sometimes in the afternoon” (P5). Although not all participants maintained the activities, they resumed or modified previous leisure activities or tried new ones, including more intellectual stimulation or physical exercise: “[The occupational therapist] showed us proprioception, so I do balance exercises” (P3). Participants also reported trying meditation: “I started exploring it. It feels good, relaxing” (P6).

Mobility

Life-space mobility decreased between T₁ and T₂, during wintertime, and increased again between T₂ and T₄ (Table 2). Maximum space mobility with any type of assistance did not change after the intervention except for a decrease in mobility without human assistance between T₁ and T₂ and a decrease in mobility without any assistance between T₂ and T₄. A decrease followed by an increase in life-space mobility was also observed in participants without disability (T₂, Mdn ± Q = 73.0 ± 8.3, vs. T₁, 83.0 ± 7.3, p = .01, and T₄, 82.0 ± 6.3, p =.02).

Participants reported that the program resulted in increased mobility (Figure 2): “[The program] got me out of my room” (P17). Several participants visited new places: “There are places where I’ve never been and where I would never have gone, either. . . . I hadn’t been out to eat since my stroke. . . . I went back to the pub twice after [the program] with my children” (P11). Nevertheless, for several participants, travel did not differ after the program, especially if they drove their own car, which allowed them to have good mobility before the program.

Health

Contrary to the lack of changes in health found in the current study, previous studies on the original Lifestyle Redesign program showed that it prevented or slowed a decline in health compared with control groups (Clark et al., 1997, 2001, 2012). Moreover, secondary analyses of the second RCT (Clark et al., 2012) showed that higher activity frequency was associated with fewer depressive symptoms through enhanced social connections (Juang et al., 2018). Such mediating mechanisms and the qualitative results from the current study point to the complexity of the effects of this intervention on health. For example, the decreased physical role—that is, the impact of physical health on time, accomplishment, and difficulties in regular daily activities—might be attributable to the participants’ greater awareness of their impairments. Notably, physical role and vitality were especially influenced by Lifestyle Redesign in the first intervention study (Clark et al., 1997).

Social Participation

In line with two previous studies (Johansson & Björklund, 2016; Matuska et al., 2003), the current study found that an adapted version of Lifestyle Redesign tended to increase social participation. According to participants in the current study and in the second original RCT (Blanchard, 2010), the intervention fostered not only social activities and interactions but also personal and environmental factors that are prerequisites to social participation, such as relationship skills and a social network. These benefits were multifaceted and diverse, especially in social support and healthy activity. Other interventions have been shown to foster social participation in older adults (Raymond et al., 2013), including those with disabilities (Levasseur et al., 2016). Nevertheless, maintaining, experimenting with, and searching for activities often require personalized assistance (Piché et al., in press), such as that provided in Lifestyle Redesign.

Leisure

Further assistance might also be needed to modify and maintain leisure activities. Older adults are not always physically and emotionally able to participate in social and leisure activities (Levasseur et al., 2016). Adapting leisure activities to older adults’ capacities often requires the expertise of an occupational therapist and a recreation specialist. Other leisure studies found increased frequency of leisure activities (Chang et al., 2015; Kao & Chang, 2017), including in older adults with disabilities (Desrosiers et al., 2007; Levasseur et al., 2016). Interventions for leisure activities are currently not sufficiently targeted in Quebec community occupational therapy practice (Turcotte et al., 2015). Education focused on the meaning of activities for the client (Dattilo, 2016; Kleiber, 2001; Lee & Payne, 2016) and awareness of the benefits and importance of these activities (Dattilo, 2015, 2016; Kleiber, 2001; Mitchell et al., 2014; Mundy, 1998) appears to be effective in increasing leisure activities (Carbonneau et al., 2011; Kao & Chang, 2017).

Mobility

Because mobility is strongly influenced by the weather, changes in these participants’ mobility over time may be attributed to the winter and, for some, to living in a residence. The participants nevertheless reported an increase in mobility, which sometimes involved changes in the perceptions of members of their network, such as a family member or health assistant who had concerns about them traveling. Such concerns and help from Lifestyle Redesign to overcome challenges in public transportation have previously been observed (Blanchard, 2010). Similarly, one study on personalized assistance improved older adults’ travel habits and increased the number of places they visited as well as their ability to travel alone (Pigeon et al., in press), which is often restricted during aging (Yen et al., 2009).

Participants With Disabilities

The program’s influence differed according to participant characteristics, including medical conditions. In those with disabilities, the decreases in the physical component, functioning, and role and the increase in impairments may be attributed to the need to cope with serious health problems and disabilities that interfere with increasing activities. Such challenges were reported in studies of the original Lifestyle Redesign program (Blanchard, 2010), Lifestyle Redesign adaptations (Horowitz & Chang, 2004; Lund et al., 2012), and other interventions (Levasseur et al., 2016). It is important to adapt the program to the group’s specific needs (Clark et al., 2015), especially for older adults with disabilities and, as discussed by Blanchard (2010), in accordance with participants’ beliefs, values, and predispositions.

End of the Intervention

Because the follow-up was only 6 mo after the intervention, and because the measurements were likely influenced by the weather, it is difficult to judge the sustainability of the changes. Nonetheless, when interviewed 1 mo after the intervention, participants reported that they missed the group, negatively affecting their morale. Those with disabilities needed the assistance of the intervention to maintain some benefits, such as getting out. Consequently, it is important to prepare participants who need social interactions and assistance for the end of the intervention and to allocate the necessary resources to maintain the benefits, as was found for another intervention with older adults with disabilities (Levasseur et al., 2019). Future studies need to document facilitators of and challenges to both the intervention and the sustainability of changes.