Abstract
The importance of promoting active aging in people with intellectual disabilities is documented, but few studies have investigated adapted interventions. This pilot study documented the social validity and measured changes in physical activity (PA), self-efficacy, social support, and self-esteem following the completion of a physical activity intervention designed for the present study and co-constructed with the participants. Method: The authors used quantitative and qualitative designs. Participants were 11 adults (40–68 years old) with intellectual disabilities and 13 staff members. Results showed an improvement of the social support measures after the program and a reduction in PA at the follow-up phase. The intervention led to no significant changes in PA, self-esteem, or self-efficacy between the pretest and intervention phases. The program demonstrated a good social validity. This study showed great appreciation of the intervention by the participants but limited associations with the dependent variables. Our results help provide recommendations for potential further use and studies.
Introduction
Intellectual disability is defined by significant limitations in intellectual functioning (reasoning, planning, abstraction, judgment), confirmed by standardized assessment, as well as deficits in adaptive behaviors that affect autonomy in the conceptual, social, and practical domains (Schalock et al., 2021). People with intellectual disabilities present significant rates of physical and psychological health problems (Cooper et al., 2015; Totsika et al., 2022). In fact, certain types of problems are more common among them than within the general population (McMahon and Hatton, 2021; Morin et al., 2012; Perera et al., 2020). These health problems become more severe with age, as they are added to the normal effects of aging and can increase the risk of premature death (Bratek et al., 2017; Cunningham et al., 2020; Glover et al., 2017; Hermans et al., 2013). In parallel, adults with intellectual disabilities are generally less physically active than the general population and have a lifestyle that is considered sedentary (Borland et al., 2020; Dixon-Ibarra et al., 2013; Oviedo et al., 2017). Hsieh and colleagues (2017) reported that approximately 66% of people with intellectual disabilities aged 40 to 59 years and approximately 80% of those aged 60 years or older were considered sedentary or inactive based on data collected through a proxy responder. Maintaining a sedentary lifestyle not only worsens physical and psychological health problems, but can lead to greater use of health services and higher costs for the health system (Bratek et al., 2017; Wang et al., 2005). Certain factors were identified as barriers to physical activity (PA) participation among people with intellectual disabilities which can be both individual and environmental (Bossink et al., 2017; Jacinto et al., 2021). At the individual level, aging, limited concentration, and reduced motivation are frequently reported as obstacles to sustained participation (Jacinto et al., 2021; Salomon et al., 2019; van Schijndel-Speet et al., 2014). At the environmental level, authors highlight the lack of adapted spaces and insufficient opportunities to engage in PA within daily routines, as well as a shortage of material resources and trained staff to support participation (Bossink et al., 2017; Jacinto et al., 2021; Salomon et al., 2019; van Schijndel-Speet et al., 2014). Finally, structural barriers, such as limited transportation options, further reduce access to community programs and restrict autonomy in participation (Jacinto et al., 2021; van Schijndel-Speet et al., 2014). The issue of sedentary lifestyle and the vulnerability of older adults with intellectual disabilities highlight the importance of identifying strategies to promote PA among them (American Psychiatric Association [APA], 2013; Fisher and Kettl, 2005; Schalock, 2011).
Physical activity is a known strategy for improving certain health indicators and preventing the effects of aging, such as bone or joint problems, diabetes, and cognitive impairment (Lacunza Odriozola, 2021; World Health Organization [WHO], 2021). It can be defined as “any bodily movement produced by skeletal muscles that requires energy expenditure” and includes “leisure, travel (e.g. walking or cycling), work, household chores, play, sports, or planned exercise, in a daily, family, or community context” (WHO, 2010, p. 8). Recent studies showed that walking is a feasible, safe, and well-accepted activity for people with intellectual disabilities (Dunsky and Barak, 2025; Melville et al., 2015). Compared to more intense exercises like running and high intensity interval training (HIIT) or aerobic programs, walking requires minimal supervision, is affordable, and can be easily integrated into a routine, as it also considers the frequent limitations in strength, coordination and balance within this population (Wolan-Nieroda et al., 2023). For people with intellectual disabilities, PA is recognized as an effective practice for improving one’s well-being and quality of life (Carmeli et al., 2005, 2009; Tomaszewski et al., 2021). In particular, it promotes and predicts better self-esteem as well as a better sense of self-efficacy and social functioning (Bondár et al., 2020; Jo et al., 2018; Nemček, 2017; Peterson et al., 2008). Authors have also identified a decrease in symptoms of anxiety and depression in people with intellectual disabilities after 12 weeks of participation in a structured PA program (Borland et al., 2022; Carraro and Gobbi, 2012, 2014). In addition to generating psychological benefits, PA also improves certain physical components, including strength and muscular endurance (Jo et al., 2018; Lacunza Odriozola, 2021; van Schijndel-Speet et al., 2017).
In recognizing the physical and psychological benefits of PA, certain authors have implemented interventions that aim to increase physical activity and reducing sedentary behavior in older individuals (Böhm et al., 2016; Muellmann et al., 2018). Other authors have implemented such interventions for adults with intellectual disabilities (e.g., Bock et al., 2019; Hanashiro-Parson and Miltenberger, 2021; Jo et al., 2018; Yan et al., 2020). Some specific intervention strategies, inspired by behavior change techniques or previous studies, have shown positive changes in PA, including the use of positive reinforcement, monitoring, feedback, goal setting, social network involvement, decision making, and planned exercise sessions (Bergström et al., 2013; Krentz et al., 2016, 2016, 2016; Martin et al., 2015; Michie et al., 2013; Muellmann et al., 2018; Savage et al., 2022; van Schijndel-Speet et al., 2017).
Among the studies consulted, the majority focused on the quantitative effects of programs, like physical performance or anthropometric measurements. However, the perspectives or appreciation of participants and staff involved in the interventions are poorly documented (Jacob et al., 2023). Authors suggest that individuals with intellectual disabilities should be involved in the process of planning, creating, implementing, and evaluating programs that concern them so that they are designed according to their needs (Godin-Tremblay and Lussier-Desrochers, 2018; Tourigny et al., 2013).
Among the few studies in the field of intellectual disability and PA that include older adults, several have limitations regarding their application within this population (Krentz et al., 2016; van Schijndel-Speet et al., 2017). Specifically, most studies are conducted outside the participants’ usual living environment, meaning that the intervention was implemented in a location different from the participants’ living environment (e.g., in a sports facility that the participants only used in the context of the study) (Krentz et al., 2016). Additionally, some of the interventions studied are difficult to maintain in the long term due to high costs or the number of human or material resources required for their implementation (van Schijndel-Speet et al., 2013). Finally, few studies have measured the effects of interventions on psychosocial variables (Bondár et al., 2020; Jacob et al., 2023).
Current research therefore paints a limited picture of interventions aimed at increasing PA in older people with intellectual disabilities. The objective of this pilot study was to measure changes following the Watch Me Walk intervention that aims to promote daily physical activity. The intervention was co-constructed by the researchers of this study, the participants as well as the staff members or the partnered organization. More precisely, the study measured older adults’ level of PA, feelings of perceived self-efficacy and social support in the context of PA as well as self-esteem following the intervention. A second objective was to document the social validity of this intervention as perceived by staff members and by older adults with intellectual disabilities.
Method
Recruitment
Recruitment took place in two group homes of an organization that provides full-time care for adults with intellectual disabilities. Both group homes were located in the same administrative region in the province of Quebec, Canada. As part of this study, two categories of participants were recruited: resident participants (i.e., adults with intellectual disabilities) and staff participants (i.e., individuals who work and live in group homes, do not have an intellectual disability, and are responsible for supporting people with intellectual disabilities in completing their daily routines and task including recreational activities, cleaning, and hygiene). Prior to starting the study, an agreement with the management of the organization was obtained to implement the intervention in two of their group homes. After presenting the goals of the study, recruitment was carried out through verbal solicitation. Written consent was obtained from all participants. The inclusion criteria for resident participants were to: a) be an older adult, b) have an intellectual disability (all levels of intellectual disability were included), c) be able to answer oral questions, d) understand simple instructions, and e) be able to move around by walking with a support (e.g. a cane or walker) or independently. In the context of this article, the term “older” refers to individuals at age 40 or more, given that there is no clear consensus in the scientific literature on this subject and that, given the population studied, the effects of aging may be observed earlier in some of these individuals (University Institute for Intellectual Disabilities and Autism Spectrum Disorders [Institut universitaire en déficience intellectuelle et en trouble du spectre de l’autisme], 2018; Canadian Down Syndrome Society, 2019). For staff participants, they had to be employed in one of the two group homes in which the study took place, regardless of whether they worked full-time or part-time. No exclusion criteria were stated for this group of participants.
Resident participants
Among the 16 residents of the two group homes, four were under 40 years old and one had a physical condition that prevented the individual’s ability to walk 1 . The remaining 11 adults with intellectual disabilities agreed to participate, representing a 100 % participation rate for participants who met the inclusion criteria. In the cases where individuals were unable to consent to their participation (n = 2), the participant’s assent was obtained, in addition to the consent of the legal representative.
Characteristics of resident participants.
Note. In order to preserve confidentiality, but to ensure consistency when the results are presented, we have nominalized the resident participants with fictitious names. F = female, M = male, BMI = body mass index.
Staff participants
All staff members (N = 13) that worked in both group homes agreed to participate in the study and supervise the intervention. Staff participants were mainly women (n = 9) and were between 18 and 67 years old (M = 34.15). Seven of the staff participants had been working for the organization for over 2 years (n = 7), three had been employed for 1–2 years, and the remaining three had been working for less than 1 year. One of the staff participants reported having more than 21 years of experience working with people presenting an intellectual disability. A few had less than two years of experience (n = 5), between 3 and 5 years (n = 5), and between 11 and 20 years (n = 2). Staff participants had diverse educational backgrounds, ranging from civil engineering to finance or health fields.
Physical activity program: Watch Me Walk
The program, named Watch Me walk, was designed in collaboration with all resident and staff participants, while also considering the facilitators and barriers to PA documented in the scientific literature on intellectual disability and aging, recognized behavior change techniques as well as results obtained by other authors who implemented PA interventions to increase PA levels (Bergström et al., 2013; Krentz et al., 2016; Martin et al., 2015; Michie et al., 2013; Muellmann et al., 2018; Savage et al., 2022; van Schijndel-Speet et al., 2017). Before starting the program, preliminary consultations were held individually with each resident participant, as well as in the form of a focus group with staff participants, thus allowing the program components to be adapted according to the participants’ preferences, the reality of the two group homes and the available resources (material, schedule, space, participant abilities, etc.). More specifically, the researchers first presented staff participants with promising techniques and components identified in the literature. The proposed elements were subsequently discussed to ensure the intervention was aligned with the organization’s needs and specificities. These preliminary consultations informed key decisions, such as the scheduling, location for physical activity sessions. Each resident participant was then met individually to discussed their opinion and preferences regarding propositions that were identified as reasonable and valued by the staff. These individual consultations led to several participant-driven adaptations: using social reinforcement and a completion certificate, setting individualized rather than group goals, using personal logbooks instead of a tracking board, and gradually increasing session frequency without extending session length. The Watch Me Walk program included different components: the inclusion of planned PA into the weekly schedule of the group homes, social network involvement, PA monitoring, choice options, positive reinforcement, objective setting, and feedback.
The program consisted of group walking sessions (i.e. social network involvement) lasting 15 minutes (i.e. inclusion of PA into the schedule). The number of weekly planned PA sessions gradually increased from two session per week to four sessions per week as the implementation of the study progressed. The total number of planned PA sessions included in the intervention program was 36 for each group home. The time of day when the sessions took place was decided by the staff participants and could differ depending on the day and the group home. Once the sessions were announced by a staff participant, the resident participants were invited to choose, as a group, how they wanted to move (i.e., choice options); they were given the choice of walking outside in the neighborhood, or walking actively indoors, following a video broadcast (15MIN ACTIVE WALKING//beginner exercise; Kaori Channel, 2023) on a smart TV. Staff participants offered verbal encouragement and praise to the resident participants (i.e., positive reinforcement). Throughout the project, resident participants wore smartwatches that recorded their daily PA data (i.e., PA monitoring). At the end of each week, all resident participants decided on a new individualized step count goal to reach for the upcoming week with a staff participant. The new objective had to include an increase in step count (i.e., goal setting). Feedback was also provided on this goal before choosing the next one (i.e., goal feedback).
To ensure implementation fidelity, the principal researcher (JP) verbally explained the procedures to the staff members during an in-person group meeting. Staff members were able to ask questions, which were answered directly. The staff members were informed of their role in the implementation of the intervention. The logbooks used for collecting the number of sessions and steps completed by participants were presented and explained to the resident and staff participants, along with instructions for playing the video on a smart TV. A calendar was also given to each home to display the periods dedicated to physical activity, as determined in collaboration with the participants. Logbooks were also used to document the course of the intervention. The procedures were also clearly detailed in a written document provided to each group home.
Mesures
Socio-demographic questionnaire
All participants completed a socio-demographic questionnaire. Data collected included variables such as age, gender, weight and height, health problems, and level of education.
Physical activity
Resident participants’ daily physical activity was objectively measuring using a wearable ® Fitness ® Tracker smartwatch that calculated their number of steps. Before starting the program, each watch was paired with a smart tablet or cell phone (belonging to the resident participant or a staff participant) through an application (® Runmefit). Prior to starting the study, the first author (JP) assessed the reliability of the smartwatch. When walking 150 steps at three different times, she compared this manually counted number of steps with the number of steps given by the watch. The accuracy of this watch was judged acceptable (94 %) and it was better than another model of smartwatch, when comparing the results.
Adaptive functioning questionnaire
The Functional Screening Tool for Adults with Intellectual Disabilities (FST-ID; Ben-David et al., 2022) was translated into French by the authors for the present study, according to the method of Tassé and Craig (1999). With the permission of Ben-David et al., initial translation was conducted independently by two bilingual (French-English) individuals (two authors of the present study: JP and ST), then reconciled into a preliminary version. This was reviewed by a second independent committee of bilingual individuals, including additional researcher (DM) and a research assistant, to ensure equivalence in content, structure, and meaning. The preliminary translation (“pretest version”) was then assessed by potential respondents from various backgrounds relevant to intellectual disability in adults, who rated the clarity of instructions, items, and presentation on a five-point Likert scale. Items scoring 3 or below from at least two respondents were revised for clarity, while comments on content or construct definition were not incorporated to preserve the original tool. As the purpose was not to validate the instrument but to inform intervention adaptation, psychometric testing (Step 7 of the translation process proposed by Tassé and Craig (1999)) was omitted. It was used to measure the level of functioning of the resident participants within three domains of adaptive behaviors (conceptual, practical and social/emotional). It is a 17-item tool offering a scale from 0 to 4, with a higher score suggesting higher adaptive functioning. The instrument has good test-retest reliability with a two-week delay between administrations (r = .95), acceptable psychometric qualities (α between 0.62 and 0.96 for the different scales) and strong positive associations with the ABAS-II (r = .78) (Ben-David et al., 2022).
Self-esteem
The Rosenberg Self- Esteem Scale (Rosenberg, 1965) consists of 10 items (with a maximum score of 40) that measures positive or negative feelings about oneself and self-attributed personal value. All items are score using a 4-point Likert scale (4 = strongly agree, 1 = strongly disagree). A total score below 25 indicates very low self-esteem, a score of 25 to 31 indicates low self-esteem, and a score higher than 31 indicates normal or high self-esteem. This instrument has demonstrated good reliability and its level of difficulty was appropriate for individuals with intellectual disabilities of all levels (Park and Park, 2019). The value of the test-retest correlation is at an acceptable level (r = .84) and its internal consistency index is acceptable (Cronbach’s alpha of 0.709; Park and Park, 2019). The French version of this instrument (Self-Esteem Scale; ÉES-10) was used in this study (Vallieres and Vallerand, 1990).
Sense of self-efficacy and social support in the context of physical activity
The French version of the three scales of the Self- Efficacy/Social Support for Activity for Persons questionnaire with Intellectual Disability (SE/SS-AID; Peterson et al., 2009) was used by adapting some terms for the study (e.g. replacing “your educator” with “a [staff participant]”). These self-reported scales measure feelings of self-efficacy in the context of PA, social support from staff, and social support from peers. Items are scores using a three-point scale (2 = yes, 1 = sometimes, 0 = no), for a total possible score of 12 points on the first two scales. The peer social support scale has a maximum score of 10. Cronbach’s alphas range from .70 to .74, reflecting a good internal consistency for all scales (Peterson et al., 2009). Test-retest reliability is low for the self-efficacy scale (r = .49) and excellent for the social support scales used (r between .76 and .78), with a two-week delay between both administrations.
Social validity questionnaire
Staff participants completed the Treatment Acceptability Rating Form-Revised (TARF-R; Carter, 2007) which has been translated (TARF-R-VF) into French by Turgeon (2021). The tool was adapted for the needs of this study. Specifically, some terms were adapted to the clientele and open questions were added (e.g. replacing “treatment” with “intervention” and replacing “your children” with “resident participants”). The instrument includes four open questions and 19 closed questions (5-point scale, where a value of 1 represents low social validity). The items address the degree of acceptance of the implemented program, the intention to implement it, the current and future vision of its effectiveness, as well as its benefits, disadvantages, clarity, and acceptability. The TARF-R demonstrated good internal consistency with a Cronbach’s alpha value of .92 (Carter, 2007).
Interview grid
The interview grid for resident participants was constructed using elements from studies on social validity in clients with neurodevelopmental disorders (Carter, 2007; Reimers et al., 1987, 1992; Reimers and Wacker, 1988; Rivard et al., 2024; Sankey et al., 2019; Winett et al., 1991; Wolf, 1978). It addresses the degree of appreciation of the program and its various components, the perception of benefits and constraints, as well as the components appreciated and less appreciated. Doctoral students in psychology with an expertise in intellectual disability and the authors of this article revised the interview grid. In addition, the interview grid was tested with these same students, as suggested by Julien-Gauthier and his colleagues (2009).
Procedure
The project received ethical certification from the Comité d’éthique de la recherche pour les projets étudiants impliquant des êtres humains (CERPE FSH) of the Université du Québec à Montréal. The design combined quantitative and qualitative methods. The quantitative portion of the project included a single-group quasi-experimental design with pre-test (A) and post-test (B) for psychosocial measures and a design of the ABC type (pre-test, intervention phase, maintenance) for the PA measurement. The study was conducted over a period of 19 weeks. It was divided into three phases, namely an 18-day baseline period (pretest), 12 weeks of intervention, and 5 weeks of follow-up/maintenance (posttest). The qualitative portion included posttest interviews.
Pretest
Starting on Day 1 of the 18-day pretest, resident participants wore the ® Fitness ® Tracker smartwatch that calculated their number of steps all day (i.e., from waking up to going to bed). Each evening, before going to sleep, assistants, along with the resident participants, recorded the total number of steps taken during the day in an individual logbook. Self-esteem (ÉES-10) as well as the feeling of self-efficacy and social support in the context of PA (SE/SS-AID) were measured at pretest. The questionnaires were administered to the resident participants in the form of a structured interview by the principal researcher (JP) in person.
Intervention phase and posttest
During the 12 weeks of the program implementation, participants continued to wear their smartwatch to measure their daily step count. Self-esteem, self-efficacy, and social support in the context of resident participants’ PA were again measured by replicating the pretest procedure at the end of the program. Social validity was also measured with resident participants through individual interviews conducted by the first author, and with staff participants through the completion of the TARF-R-VF.
Maintenance phase
At the end of the 12-week intervention, the program components stopped being implemented by the research team. For example, the group homes were no longer asked to scheduled PA sessions and participants were no longer instructed to set goals. However, participants continued to wear the smartwatch daily to track their number of steps during the maintenance phase. As during the other phases of the study, step counts were recorded in the logbook. The maintenance phase lasted five weeks.
Data analysis
The quantitative data were analyzed with nonparametric Wilcoxon tests (Harris and Hardin, 2013) using the IBM® SPSS® software, version 27 (IBM Corp, 2020). This test was chosen because of the small sample size and non-normal distribution of the data. For analysis using step counts, resident participants’ mean of the steps for each of the three phases was used. To calculate the mean, each resident participant’s total number of steps for each phase was divided by the number of days for the respective phase. In the cases where data was missing, the day was not included in the calculation of the mean.
Regarding the data from the qualitative design, interviews with the resident participants were first transcribed by a research assistant. The principal researcher (JP) made an initial selection of relevant excerpts to exclude certain parts of the interviews that were not relevant to the project. A deductive thematic analysis of the content of the interviews was then carried out (Braun and Clarke, 2006; Paillé and Mucchielli, 2021). In order to determine the themes (nodes), the author (JP) relied on scientific literature in the field of social validity and on the main concepts found in the TARF questionnaire (Carter, 2007; Miltenberger, 1990; Reimers et al., 1987, 1992; Reimers and Wacker, 1988; Winett et al., 1991; Wolf, 1978). The analysis was done according to the following nodes: appreciation and general satisfaction, understanding of the program, positive effects and improvements, undesirable side effects as well as potential risks or concerns. A verification of the clarity of the categories was then carried out according to the method proposed by Lincoln and Guba (1985), in order to increase the reliability of the analysis. From the selected excerpts, the main author and a research assistant independently coded the interviews using the NVivo software (version 12). For each of the excerpts where coding diverged, they held discussions to establish a consensus.
Results
Physical activity sessions and number of steps
In the first group home, 24 PA sessions took place in total, which represents 66.67 % of the total sessions that were planned. The rate of complete participation (where the person is physically engaged throughout the session) or partial participation (where the person participates by moving their feet while sitting, or they participate only for a few minutes) was 83.33 %. The choice of the type of PA was the walking video 66.67 % of the time. In the second group home, 23 PA sessions took place, representing 63.89 % of the planned sessions. The participation rate in this group home was 93.47 % and the video was chosen 55 % of the time. The planned 15-minute duration for the sessions was always respected.
All participants completed the 19-week study period. Data from all three phases of the study were therefore obtained, with the exception of days when participants forgot to wear their watch or lost their data due to their watch not being charged (M = 21 days). Of the study’s total 137 days (18 pretest days, 84 intervention days, and 35 maintenance days), 15.33 % of the data was missing.
Variations in the number of steps during the different phases of the study, in comparison with the number of steps at pretest.
Note. The ↓ symbol indicates a decrease in the number of steps compared to the pretest, while the ↑ symbol indicates an increase in the number of steps compared to the pretest. Numbers in bold are larger than the 5 000 steps/day threshold for a sedentary lifestyle.
Conversely, the results show a statistically significant difference between the intervention period and the maintenance period (Z = -2.67, p = .008), indicating a decrease in the number of steps once the program was completed. The effect size is strong (r = .57; Cohen, 1988). The median during the maintenance phase (Md = 3,293.28, M = 3,287.90, SD = 2,185.53) was indeed lower than during the intervention phase. During the maintenance phase (the five weeks following the end of the intervention) the majority (n = 10/11) of resident participants decreased their step count compared to the weeks during the intervention phase. The median of the maintenance phase was also significantly lower than that of the pretest phase (Z = -2.05, p = .041). The effect size is moderate (r = .44; Cohen, 1988).) The majority (n = 7/11) of resident participants took fewer steps in the maintenance period than the pretest phase.
Psychosocial measures
The psychosocial measures were administered to all participants for ethical reasons. However, participants who did not have a sufficient level of understanding or verbal expression were excluded (n = 3), so that the data would not be biased. Therefore, the following results concern eight participants.
Self-esteem
Pretest–posttest comparison of resident participants’ scores for self-esteem.
Sense of self-efficacy in the context of physical activity
Pretest–posttest comparison of resident participants’ scores for sense of self-efficacy in the context of physical activity.
Feeling of perceived social support in the context of physical activity
Pretest–posttest comparison of resident participants’ scores for feeling of perceived social support in the context of physical activity.
Social validity
TARF-R-VF
Seven of the 13 staff participants completed the TARF-V-VF. A score above three suggests a positive assessment of the item measured, a score of three represents a neutral assessment, and a score below three suggests a negative assessment of the item (Carter, 2007). Staff participants gave the program a median social validity score of 4.0 out of a maximum of 5.0. The items with the highest scores were those concerning the willingness to recommend the program to other settings, resources, or organizations (Md = 5.00, M = 4.71, SD = 0.49), appreciation of the procedures used in the program (Md = 5.00, M = 4.57, SD = 0.53) and understanding of the intervention (Md = 5.00, M = 4.56, SD = 0.79). Staff participants gave the lowest scores to the following items: “To what extent do you think there might be disadvantages to implementing this program?” (Md = 3.00, M = 2.71, SD = 0.95), “How disruptive will the application of this program be to the living environment?” (MD = 3.00, M = 3.00, SD = 1.29), and “To what extent will the implementation of this intervention be costly?” (MD = 3.00, M = 3.71, SD = 0.95). These results suggest that staff participants concerns are generally neutral or very mild.
Interviews
The results of the qualitative analysis are divided into three sections, according to the nodes: assessment of the program and general experience, positive effects and adverse effects (Table 6). The quotes presented were translated by the first author (JP).
Assessment of the intervention and general experience
The PA intervention Watch Me Walk was generally well-received by all resident participants, with many stating that it was enjoyable and fun. Louise said of the intervention components, “It was really interesting and fun. I liked everything about it”. When asked about each component of the program (i.e., integration of PA into the schedule, involvement of the social network, monitoring of PA, choice options, reinforcement, goal setting, and feedback), all resident participants reported a positive appreciation of them. For example, about the choosing the type of PA, Georgia said “I liked it because we weren’t always doing the same thing… and, well, it was motivating”. They also mentioned that these components had an important and positive impact on their motivation to be physically active. Madison also reported the important role that goal setting played in her motivation: “It kind of boosted me. […] I’m not sure that, uh, on my own [without a goal], I would have done it”. When asked about the elements they liked most, some resident participants mentioned wearing the watch and the movements shown in the video. Three resident participants also mentioned that they enjoyed the experience because they were able to meet a new person (the author of the project) and chat with her.
Resident participants also reported that they were highly motivated to move and had a broad understanding of the importance of PA for their health. For example, Georgia mentioned this in relation to the program: “It made me want to continue [to walk and move] for my health.” […] and to become in good shape”.
Perceived positive effects
Resident participants also reported the benefits of the program, including experiencing positive emotions or feeling physical and mental well-being. For example, several reported feeling good after completing a PA session or after reaching their steps goals, and six resident participants added feeling “happy” or “proud”. Louise mentioned that “it felt good. […] I was excited to do it, and I couldn’t wait to do it”. Resident participants mostly reported doing more PA, walking more and being more motivated to move than before the program.
Perceived adverse effects
Summary of qualitative results from resident participants.
Discussion
The objective of this pilot study was to measure the effects of the Watch me Walk program, which was co-constructed with people with intellectual disabilities and staff members living in two group homes in Quebec. More specifically, the dependent variables of interest were PA, feelings of self-efficacy and perceived social support in the context of PA, and self-esteem. As concluded by other studies (e.g., Ammann et al., 2013; Melville et al., 2011; Turgeon et al., 2024), the results of this pilot study support the social validity of a co-constructed PA program implemented in a natural living environment with older adults with intellectual disabilities. The different components were implemented throughout the study and the participation rate in the walking sessions and goal setting activities was high. As reported by other authors, it is possible to conclude that older adults with intellectual disabilities in their 40s, 50s, and 60s are able to follow a PA program that requires little supervision, as well as to participate in its conceptualization (Bondár et al., 2020; Lynnes et al., 2009; Stanish et al., 2001; Valbuena et al., 2019). This intervention was supervised by staff members, which are identified as suitable providers in successful interventions (Nutsch et al., 2021). However, the levels of PA achieved by participants with intellectual disabilities before, during or after the intervention do not reach health recommendations, as also shown by other studies (Dixon-Ibarra et al., 2013; Oviedo et al., 2017; Stancliffe and Anderson, 2017; Temple, 2010). In fact, most of the participants are considered to have a sedentary lifestyle throughout the study, with daily step count under 5 000 (Cheng et al., 2023). The number of participants with a median step count above this threshold was two during the pretest phase, two participants during the intervention phase, and one participant during the maintenance phase.
Contrary to expectations, the intervention did not lead to a significant increase in participants’ median step counts during the 12-week implementation period. However, when the components of “planned group PA”, “positive reinforcement”, “choice options”, “goal setting”, and “feedback” were removed during the maintenance phase, participants’ step counts decreased by more than 15% compared to the pretest and intervention phases. It is possible that the lack of a significant difference between pretest and posttest can partly be explained by the implementation of a specific component of the program during the pretest phase: step monitoring with the smartwatch. Indeed, it can be hypothesized that the addition of this component increased the number of steps by motivating resident participants to move and by indirectly encouraging staff participants to mobilize participants to be more active during this period (Chaudhry et al., 2020; de Vries et al., 2016; Gal et al., 2018; Larsen et al., 2019; Michie et al., 2013). In other words, since they were aware of their participation in a study related to PA, resident participants and staff participants potentially showed more awareness, enthusiasm, and valued for PA, even before the remaining components of the program were implemented during the intervention phase (Bernal et al., 2021; Bondár et al., 2020; McCambridge et al., 2014). This potential threat to validity can be interpreted as a form of “study effect” or “Hawthorne effect” whereby merely being observed or involved in a structured program may increases participants’ motivation and engagement (Landsberger, 1958). For example, for resident participants, wearing the watch and knowing that their steps were monitored may have acted as a constant reminder of the importance of PA, fostering a sense of accountability and pride in their achievements. For staff participants, the awareness of supporting an intervention with a clear research purpose may have created a stronger commitment to encourage residents, integrate movement opportunities into daily routines, and value PA as part of their professional role. For these reasons, the baseline level of step counts obtained in the study is possibly higher than the true level of PA.
Many other factors that were highlighted as barriers to PA can also explain the lack of improvement in the number of steps during the intervention phase (Bossink et al., 2017; Jacinto et al., 2021). For example, as suggested by Bergström and colleagues (2013) and Melville and colleagues (2015), interventions that are too low in intensity, duration, or frequency may not lead to significant changes in participants’ PA. In the case of the Watch Me Walk program, two to four 15-minute walking sessions per week may not be sufficient to cause a significant increase in step counts. Other elements that were not assessed in this study that may have acted as barriers to participants PA levels include lack of concentration, preference for sedentary activities, physical discomfort, and the time required for the activity (Bossink et al., 2017; Jacinto et al., 2021). These elements could explain why some resident participants improved their daily average (n = 5) of steps during the intervention, while others did not (n = 6).
The results also show a decrease in the number of steps when the program was no longer being implemented (maintenance phase). This suggests that the level of PA achieved during the intervention is not maintained when certain components of the intervention program are removed. It is possible that the removal of certain components such as goal setting and team work (planned PA sessions with the adults living in the group home) decreased participants intrinsic and extrinsic motivation towards PA. Nevertheless, initial concerns expressed by some participants (e.g., fear of falling, not reaching goals, or damaging the watch) did not materialize and did not hinder participation, indicating good tolerance to potential constraints and high program feasibility. Finally, the interest generated by certain components, such as activity tracking with a smartwatch and the option to choose between outdoor walks and video-guided sessions, highlights the value of integrating personalized and flexible elements into future programs to maximize adherence and participant satisfaction (Nutsch et al., 2021; Salomon et al., 2019).
Although the analyses did not reveal any significant increase in the measure of PA during the intervention, as well as in the feeling of self-efficacy or self-esteem, the resident participants felt increased supported by their peers and by the staff participants after the intervention, and this element encouraged them to engage in PA. The involvement of the social network and of social reinforcement were both perceived positively by staff and resident participants. Social interactions, whether walking in a group, conversing with the researcher, or sharing structured time with staff, emerged as catalysts for participation. These findings are consistent with the observed quantitative improvements in perceived social support and align with previous research highlighting the value of group-based formats and social reinforcement in PA interventions for this population (Bossink et al., 2017; Nutsch et al., 2021; Salomon et al., 2019). Other authors have also concluded that social support should be integrated in interventions aimed at increasing PA in adults with intellectual disabilities (Böhm et al., 2016; van Schijndel-Speet et al., 2014). In addition, beyond quantitative indicators, participants’ experiences were marked by motivational and social dimensions that favored engagement. The enjoyment, fun, and sense of accomplishment reported by several participants suggest that the program fostered intrinsic motivation, a key factor in sustaining physical activity among people with intellectual disabilities (Deci and Ryan, 2000; Nutsch et al., 2021). Feelings of pride and well-being after achieving goals, as well as the anticipation of upcoming sessions, point toward a reinforcement of perceived competence, consistent with mechanisms described in Self-Determination Theory and with evidence linking positive affect to long-term adherence to PA (Deci and Ryan, 2000; Ryan and Deci, 2000; Teixeira et al., 2012).
Despite the acceptable-good psychometric qualities of the self-esteem and self-efficacy in the context of PA questionnaires, the small number of items combined with our small sample size may partly explain the lack of significant changes in these outcomes as a result of limited variance (Cohen, 1988). In addition, the questionnaires used may not have been sensitive enough to demonstrate certain changes over a short period of time. The data from these tools also appeared to be highly context sensitive, so self-esteem scores may have been negatively influenced by life events of the resident participants (e.g., several people contracted a virus that weakened their immune system when taking psychosocial measures at posttest). Despite the lack of a significant difference for the self-efficacy measure, the effect size approaches the conventional threshold, which represents a notable trend that could indicate a significant effect.
Strengths and limitations
The co-development of the Watch Me Walk program with key stakeholders (i.e., staff members and the adults with intellectual disabilities) and researchers is a strength of this project. Co-developping such interventions is helpful in ensuring that the intervention is tailored to participants characteristics, environments, living reality, while ensuring that the program is empirically informed. Regarding limitations, the small sample and the heterogeneous profiles of the resident participants limit the possibility of generalizing the results of this study. The limited sample size also made it impossible to isolate the effect of each of the components of the intervention program (i.e., integration of PA into the schedule, involvement of the social network, monitoring of PA, choice options, reinforcement, objective setting and feedback) and to include control groups. Future research should seek to identify the components with the largest effect size for increasing older adults with intellectual disabilities’ PA levels. The implementation of a posttest phase nevertheless represents a strength of the study since it made it possible to note a decrease in PA once the intervention was over.
Measuring PA using a monitoring tool is more accurate than self-reported questionnaires, since the latter often overestimates the level of PA (Tucker et al., 2011). Some authors use accelerometers for their accuracy (Oviedo et al., 2019; Saint-Maurice et al., 2020). However, in the context of this project, the ® Fitness ® Tracker smartwatch was better suited because it was easy to use and enabled the collection of a simple, inexpensive, and accessible measure of physical activity (i.e., the number of steps) for people with intellectual disabilities and staff members. This choice of the ® Fitness ® Tracker watch as a step monitoring tool is also a strength of this study. Other authors have also highlighted the relevance of using technology in these types of programs (Irvine et al., 2013; Martinez-Millana, 2022; Muellmann et al., 2018; Torrado et al., 2022). Although the watch was appreciated and did not cause any major constraints, using walking as a measure of PA comprises certain limitations. First, the data obtained in this project did not allow us to quantify the overall PA of people with intellectual disabilities (e.g. swimming, yoga, weightlifting and cycling), but informed us on their number of steps. Finally, given that the study only measured number of steps, it is not possible to quantify participants PA in terms of intensity (light, moderate, vigorous). Given that higher intensity PA is often associated with larger effects on physical and psychological outcomes, future research should assess how PA intensity influences such outcomes in older adults with intellectual disabilities.
Future directions
Based on the qualitative data of the present study and others previous findings, PA sessions should be scheduled (Michie et al., 2013; Schroé et al., 2020; Turgeon et al., 2024; van Schijndel-Speet et al., 2017). The integration of PA into participants’ daily schedules seemed be a valuable component. These findings are consistent with those of previous research, where integrating PA into the daily routine of older individuals with intellectual disabilities was considered to successfully increase PA participation (Nutsch et al., 2021; Salomon et al., 2019). In addition, Vlot-van Anrooij and colleagues (2020) suggest that programs should be designed to require minimal preparation or setup and be delivered directly within the living environment. Although walking was the only form of PA including in the intervention, giving participants the opportunity to choose between outdoor walks and indoor video-guided walking was appreciated by participants.
When creating intervention withing natural settings such as group homes, researchers should adapt the frequency and duration of sessions should to ensure adherence and long term implementation (Bergström et al., 2013; Melville et al., 2015; Nutsch et al., 2021). It would also be appropriate, when developing such programs, to add a component that addresses education/awareness of the importance of being physically active and the different actions to increase one’s PA, not only during walking activities, but also in other contexts (Nutsch et al., 2021). Social support is also a component that should be included and highlighted in such programs (Böhm et al., 2016; Melville et al., 2015; van Schijndel-Speet et al., 2014). Burke and colleagues (2010) found a strong and positive association between the implementation of group goals and group performance in the context of PA, emphasizing the relevance of exploring this method of intervention for people with intellectual disabilities. Finally, considering participants’ appreciation of the smartwatches and the absence of major constraints regarding their use, this technology should be considered when including PA monitoring components into interventions. Future researchers should also have staff participants wear physical activity monitors. Collecting such data could provide valuable insights into how the program influences the social environment, enhances staff awareness of PA, and potentially creates a ripple effect beyond the resident participants.
Conclusion
The Watch Me Walk program did not result in any significant changes in the number of steps taken daily. However, four of the 11 participants did see important increases during the implementation of the intervention. More research is needed to understand effective programs and components that can lead to significant lifestyle changes in older adults with intellectual disabilities, while keeping in mind that Carty et al. (2021) point out that “Doing some PA is better than doing none.” (p. 90). In addition, no significant changes were observed in the resident participants’ feeling of self-efficacy or self-esteem, but the program was perceived very positively and all active and staff participants recognised its benefits. It was perceived as acceptable, enjoyable, motivating, generally easy to implement and relevant to encourage older people with intellectual disabilities to be more physically active. Despite efforts to adapt the intervention to the resources of the environment and a preliminary consultation with all stakeholders involved regarding the feasibility and modalities of the intervention, all the results suggest that certain barriers remain. Feelings of social support should be at the heart of future intervention models in order to facilitate PA in these individuals, and future studies should focus on feelings of self-efficacy as a central tenet of such programs.
Footnotes
Acknowledgements
The authors would like to thank the staff that work at the organization for collaborating on the project and the legal tutors and participants who agreed to take part in this study. We would also like to thank research assistants for their help in analysis and translation.
Funding
The Chaire Dr William-Barakett de déficience intellectuelle et troubles du comportement has provided financial support.
Declaration of conflicting interests
The authors have no conflict of interest to report.
