Outcomes of older adult physical activity participation that matter to health care professionals

Abstract

Background and Objectives

There is extensive heterogeneity in outcome domain selection and reporting in randomized controlled trials (RCTs) of physical activity interventions for older adults. Physical activity researchers need a core outcome set that recommends a minimum set of outcome domains to measure consistently, guided by input from health care professionals to ensure clinical relevance. This study aimed to identify the outcome domains related to older adults’ physical activity participation that are most important to health care professionals.

Research Design and Methods

Health care professionals completed an online survey in which they rated the importance of measuring 24 candidate outcome domains in future RCTs of physical activity interventions for older adults and ranked their top four.

Results

225 health care professionals participated (72% female; mean [SD] age, 37.8 [11.2] years; 21 general practitioners, 99 physiotherapists, and 105 kinesiologists). The 5 outcome domains most frequently rated as “highly important” (6–7 of 7) were Falls (92.0%), Quality of Life (91.5%), Independence (87.9%), Balance (84.4%), and Mobility (82.7%). The 5 outcome domains that most frequently appeared in participants’ top 4 importance rankings were Quality of Life (67.1%), Independence (50.2%), Risk and Management of Chronic Disease (40.2%), Mobility (33.3%), and Falls (32.9%).

Discussion and Implications

Health care professionals prioritized Quality of Life, Independence, Falls, and Mobility as key outcome domains to measure in RCTs of physical activity interventions for older adults. These results will help align research with clinical priorities, support physical activity promotion, and guide the development of a core outcome set.

Innovation and Translational Significance:

Older adult physical activity research is negatively affected by extensive outcome variability between studies and often lacks sufficient input from health care professionals. We conducted an online survey of health care professionals who promote or prescribe physical activity to older adults (105 kinesiologists, 99 physiotherapists, 21 general practitioners). Health care professionals prioritized Quality of Life, Independence, Falls, and Mobility as key outcome domains to measure in future clinical trials of physical activity for older adults. These findings will support clinically relevant outcome selection in older adult physical activity clinical trials and development of a standardized “core outcome set” to reduce outcome variability.

Background and objectives

Physical activity interventions evaluated in randomized controlled trials (RCTs) have shown benefits for older adults across a range of health outcome domains, such as improved management of many chronic diseases (Chodzko-Zajko et al., 2009), prevention of mobility disability (Pahor et al., 2014), reduced fall risk (Nelson et al., 2007), and more. However, many older adults remain inactive. In the United States, approximately 85% of older adults self-reported that they do not meet recommended physical activity guidelines (Elgaddal & Kramarow, 2024), and in Canada, approximately 75% of adults aged 65 to 80 years did not meet recommended guideline levels of physical activity based on objective measurement with accelerometers (Colley et al., 2023). This disparity between evidence and practice demands action to bridge the gap through consistently reported, relevant research outcomes.

There remains extensive heterogeneity in outcome domain selection and reporting in RCTs of physical activity interventions for older adults (Mackey et al., 2023), consistent with earlier observations (Keysor & Jette, 2001). Such outcome heterogeneity creates incompatible data across related RCTs, limiting the ability to synthesize findings across studies. This impedes the combination of research into highly regarded and influential scientific outputs like systematic reviews and meta-analyses (Clarke & Williamson, 2016; Williamson et al., 2017, 2020). To address this outcome heterogeneity, physical activity researchers need a “core outcome set” (COS) that provides clear guidance on a minimum and standardized set of outcome domains to measure and report in all RCTs of physical activity interventions for older adults (Mackey et al., 2023). Involving knowledge users, including relevant health care professionals, in the development of such a COS is a priority emphasized by international frameworks for COS development (Kirkham & Williamson, 2022; Kirkham et al., 2017; Williamson et al., 2017). This approach captures knowledge user priorities within recommended outcome domain selection and therefore increases the likelihood that selected outcome domains will be applicable to clinical practice.

Health care professionals—specifically general practitioners, physiotherapists, and kinesiologists—are often the first point of contact in helping older adults become active and can be instrumental in promoting and prescribing physical activity to them (Fortier et al., 2024; Moreno et al., 2019; Orrow et al., 2012; Studer et al., 2025). Their ability to effectively promote physical activity is enhanced when they have evidence about the impacts of physical activity on outcomes that are meaningful, useful, and applicable to their practice (Green, 2008; Vuori et al., 2013). Despite this, health care professional perspectives may be underrepresented when outcome domains (i.e., “what” to measure) are selected for RCTs evaluating physical activity interventions. This is problematic, as RCTs are highly influential in shaping clinical practice guidelines and health policy (Chan et al., 2025). Integrating health care professional perspectives into outcome domain selection for RCTs will help ensure the relevance of future physical activity research to clinical practice and enhance the ability of health care professionals to support physical activity among older adults.

The purpose of this study was to support the development of a COS for RCTs of physical activity interventions for older adults, this study aimed to identify which outcome domains associated with older adults’ physical activity participation are deemed most important by health care professionals. Our secondary objectives were to (a) explore differences in outcome domain importance by healthcare profession, gender, and age and (b) identify any adverse events or other concerns of health care professionals regarding older adults’ physical activity participation.

Research design and methods

Study design

This was a cross-sectional, anonymous, online survey of health care professionals. We targeted general practitioners (GPs), physiotherapists (PTs), and kinesiologists (Kins) based on consultation with patient partners and a review of relevant literature, highlighting where older adults might seek advice regarding their participation in physical activity (Fortier et al., 2024; Moreno et al., 2019; Orrow et al., 2012; Studer et al., 2025). The study was approved by the research ethics board at Simon Fraser University (SFU REB #30001863). Prospective participants who clicked the online survey link were directed to a webpage with study information and the consent form. All participants checked a box on the online consent form indicating consent to participate.

Participants

Individuals were eligible to participate if they self-reported on the online survey that they were currently practicing at least 1 day/week in the province of British Columbia (BC), Canada, as a GP, PT, or Kin, and that at least 20% of their patients were aged 65 years and older.

We aimed to recruit a representative sample of 300 health care professionals, yielding estimates generalizable to the greater population of BC health care professionals at a 95% confidence level and a ± 6% margin of error. Specifically, we aimed to recruit 100 GPs, PTs, and Kins each to allow generalization within each profession at a 95% confidence and a ± 10% margin of error. The target sample size of n = 300 also provided at least 90% power to detect differences of 30% or greater using chi-square goodness-of-fit tests at a significance level of α = 0.05. As this study was novel and exploratory, we opted to power it for detecting larger effect sizes (Althubaiti, 2022).

Recruitment

We recruited participants between February 2024 and February 2025 using various methods tailored to each of the three health care professional groups. To recruit GPs, we partnered with Doctors of BC’s Divisions of Family Practice. To recruit PTs, we partnered with the Physiotherapy Association of British Columbia and the University of British Columbia’s Department of Physical Therapy. To recruit Kins, we partnered with the BC Association of Kinesiologists. Each partner used different approaches to contact their members, such as publishing study advertisements in e-newsletters and sending email invitations to member lists. In addition, we directly contacted health care professionals from all three professions using publicly available email addresses, and we collaborated with Simon Fraser University’s Alumni Association to send email invitations to alumni of the Department of Biomedical Physiology and Kinesiology. Finally, we advertised the study on LinkedIn and X (formerly Twitter).

We incentivized health care professionals to participate in the survey by offering entry into eight $50 gift card draws and providing an educational PDF upon survey completion that included (a) key insights on what older adults want to gain from physical activity and (b) information on a health promotion program—including physical activity—for older adults in BC to which they could refer clients. To mitigate invalid survey responses, we offered a draw rather than guaranteed compensation and did not disclose specific eligibility criteria (e.g., requiring participants to self-report that ≥20% of their clientele were aged 65+ years). Survey cookies also blocked repeat access for screened-out participants.

Measures

Participants completed a brief (∼15-minute) online survey hosted on the platform SurveyMonkey. The survey was pilot tested by three Kins and two PTs to assess technical operations and pragmatic acceptability, including clarity, flow, and completion time (Kelley et al., 2003).

To assess opinions about outcome domain importance for evaluating the effects of physical activity interventions, participants were asked—from their perspective as health care professionals—to rate the importance of measuring each of 24 unique domains (categories) of outcomes in future RCTs of physical activity interventions for older adults on a 7-point Likert scale (1 = not at all important, 4 = somewhat important, 7 = extremely important). Before rating the outcome domains, participants were provided with brief definitions of “randomized controlled trials” and “outcomes” to ensure clarity of key concepts. The order of outcome domain presentation was randomized across participants; a response option of “I prefer not to answer” was provided for all domains; and participants could enter and rate “other” outcome domains that were not included among the list of 24. The 24 prepopulated outcome domains were derived from the standard outcome classification taxonomy endorsed by the COMET initiative (Dodd et al., 2018), our previous rapid review of RCTs of physical activity for older adults (Mackey et al., 2023), our previous interview study on outcome domains relevant to older adults (Young et al., 2024), and additional input from the research team.

Since we anticipated many outcome domains would be highly rated, participants were next presented with the list of outcome domains that they had rated as highly important (i.e., as a 6 or 7 of 7 on the Likert scale) and asked to rank the four outcome domains that they thought were most important to measure in future RCTs of physical activity interventions for older adults (1 = most important to 4 = fourth most important).

A previous set of interviews our research team conducted revealed that older adults ultimately desire their physical activity participation to help them maintain independence and quality of life (Young et al., 2024). Thus, we also asked participants in the current study—via open-ended questions—to indicate which specific outcome domains from physical activity participation they thought would most impact the ability of older adults to maintain their (a) independence and (b) quality of life.

In response to growing calls for researchers to systematically report adverse events in physical activity trials (Junqueira et al., 2023; Zorzela et al., 2016), we also explored health care professionals’ concerns about recommending physical activity to older adults. Using unaided, open-ended questions, we asked participants what concerns they would have about their older adult patients engaging in physical activity.

At the end of the survey, we ascertained participant demographic information, including age, sex assigned at birth (female, male, intersex, prefer not to answer), gender identity (woman, man, non-binary, do not identify with any of the above options, prefer not to answer), experience living as a trans person where gender identity does not align with sex assigned at birth (yes, no, prefer not to answer), identifying as a member of the LGBTQSIA+ communities (yes, no, prefer not to answer), and cultural background. Related to their professional status, we assessed hours per week of work within the profession, years of experience working in the profession, place of practice (private practice, long-term care/residential care/assisted living, public health facility (e.g., hospital), public recreation center, other), and the regional health authority within BC within which they practiced. Finally, we asked how many days in the past week (0–7) participants did a total of 30 minutes or more of physical activity, which was enough to raise their breathing rate.

Data analysis

Before analysis, we cleaned survey data to detect and remove fraudulent or suspicious responses. Specifically, we screened survey opening and completion times for identical or overlapping entries, examined inconsistencies in reported demographic information for suspicious responses (e.g., a 20-year-old GP) and meticulously examined open-ended questions for similarities, signs of computer-generated text, and other irregularities (Wang et al., 2023). Entries deemed suspicious (n = 19) were excluded before analysis.

All quantitative data analysis was conducted in RStudio (version 2022.12.0, build 353) using R (version 4.5.0, April 2025). We used ChatGPT (GPT-4o, OpenAI, July 2025) to assist in producing and refining R scripts. Specifically, ChatGPT supported translating our predefined analysis plan into code (e.g., “can you help build a looping chi-square test across these stratification groups?”) and aided in catching and fixing coding errors (e.g., “can you explain this error message?”). All AI-assisted code was directed by, executed, and reviewed by the authors to ensure its validity in accordance with AI ethical guidance (Flanagin et al., 2023).

Rating

We visually inspected the distribution of Likert scale ratings (1–7) for each outcome domain and observed a strong skew toward higher values for all domains. To facilitate clear comparisons of outcome domains, we dichotomized ratings into “highly important” (6 or 7 of 7) versus all other ratings. We then calculated the proportion of participants who rated each outcome domain as highly important. Next, we conducted a chi-square goodness-of-fit test across all outcome domains to identify which outcome domains were significantly (α = 0.05) more or less likely to be rated more highly important than expected under a uniform distribution.

After determining the overall importance of outcome domains to health care professionals, we conducted stratified chi-square tests of independence to explore differences in the proportion of highly important ratings across professional groups (PTs and Kins), genders (men and women), and age groups (20–29, 30–39, and 40+ years). There were insufficient numbers of GPs and non-binary individuals to include in comparisons across professional and gender groups, respectively. Age groupings were chosen to create roughly equal numbers of participants in each subgroup. We applied a Bonferroni correction to adjust for multiple comparisons. We did not report p-values for stratified comparisons with sample size <10.

Rankings

Our ranking analysis mirrored the approach used for the rating data. Each outcome domain was dichotomized as either “ranked in the top 4” or “not ranked.” Next, we calculated the proportion of participants who ranked each outcome domain in their top 4 and conducted the same series of analytical tests, including a chi-square goodness-of-fit test to assess differences in proportions, followed by stratified chi-square tests of independence across professional group, gender, and age categories. As with the rating analysis, statistical significance was set at α = 0.05, and a Bonferroni correction was applied to adjust for multiple comparisons. We did not report p-values for stratified comparisons with a sample size <10. To consolidate rating and ranking data, we sought to identify outcome domains that were both highly rated and frequently ranked and looked for natural breakpoints in the proportion of participants who rated each domain highly and ranked it in their top 4.

Open-ended data

This survey included two sets of open-ended questions, which required qualitative processing, as described below. The first set asked participants to identify which physical activity outcome domains they believed would be most impactful in promoting older adults’ independence and quality of life. For these questions, the first author conducted the initial categorization of responses, with input and confirmation from the last author for complex or ambiguous cases. When possible, responses were deductively coded to the predefined list of outcome domains provided to participants (e.g., “leg strength” and “core strength” were grouped under Strength). In other cases, responses were created inductively to reflect participant input (e.g., “maintaining household chores” and “being able to manage daily obligations” were categorized as Activities of Daily Living) (Braun & Clarke, 2021).

The second set asked participants to describe any concerns they associated with recommending physical activity to older adults. This question followed a similar processing approach, but codes were developed inductively as new responses emerged. After coding all the responses, we grouped them thematically for conceptual clarity in reporting.

Results

Participant characteristics

The survey was opened, and the consent form was completed a total of 364 times (Supplementary Figure 1). Of these, 79 individuals did not meet eligibility criteria: 26 did not report an eligible profession (i.e., not a GP, PT, or Kin), 49 did not meet the minimum requirement of having ≥20% of their clientele aged 65 years or older, 4 were not currently practicing, 19 were removed due to suspected fraudulent responses, and 9 responses were incomplete. Thus, 266 health care professionals started the survey, and 41 had their data removed due to insufficient survey completion (<10% completion). A final sample of 225 participants was included in the analysis, which included 21 GPs, 99 PTs, and 105 Kins.

Overall, the sample had a mean (SD) age of 37.8 (11.2) years, was predominantly women (72.2%) with a European background (64.9%), and a few reported lived experience as a trans person (1.4%) or identified as LGBTQ2SIA+ (6.0%) (Table 1). Participants worked an average of 34.4 (11.5) hours/week, had 12.0 (10.3) years of experience in their current healthcare profession, and primarily worked in private practice (70.0%). Participants were recruited from all of the seven regional health authorities in BC, and the majority of participants worked in the two largest health authorities: Fraser Health (36.7%) and Vancouver Coastal Health (33.0%). As a whole, the sample was physically active, with 59.5% reporting 30+ minutes of physical activity on 5–7 days each week. There were some differences observed across the three professional groups. Namely, Kins were younger, worked fewer hours per week, and had fewer years of work experience than PTs and GPs. In addition, PTs and Kins tended to report more physical activity than GPs.

Table 1.

Demographic characteristics of the total sample and stratified groups: general practitioners (GPs), physiotherapists (PTs), and kinesiologists (Kins).

Variable	Total	GP	PT	Kin	p
	N = 225	n = 21	n = 99	n = 105
Age (years), mean [SD]	37.8 [11.2]	40.5 [10.4]	43.2 [10.5]	32.3 [9.2]	<.001
Range	22–73	26–73	27–68	22–59
Hours worked (per week), mean [SD]	34.4 [11.5]	49.3 [11.9]	35.4 [9.0]	30.6 [10.9]	<.001
Range	5–70	30–70	15–60	5–60
Years in profession, Mean [SD]	12.0 [10.3]	11.3 [10.9]	16.7 [11.2]	7.1 [7.1]	<.001
Range	0–48	0–48	0–44	0–30
Current gender, n (%)					.693
Woman	160 (72.2)	14 (66.7)	70 (75.3)	76 (74.5)
Man	55 (27.4)	7 (33.3)	23 (24.7)	25 (24.5)
Non-binary	1 (0.4)	0	0	1 (1)
Missing	9	0	6	3
Sex, n (%)					.638
Female	160 (71.4)	14 (66.7)	69 (73.4)	77 (75.5)
Male	56 (28.1)	7 (33.3)	25 (26.6)	24 (23.5)
Intersex	1 (0.4)	0	0	1 (1.0)
Missing	8	0	5	3
Lived experience as a trans person, n (%)					N/A
Yes	2 (1.4)	0	1 (1.1)	1 (1.0)
No	213 (98.6)	21 (100.0)	92 (98.9)	100 (99.0)
Missing	10	0	7	4
Identified as LGBTQ2SIA+, n (%)					N/A
Yes	13 (6.0)	2 (9.5)	5 (5.4)	6 (5.9)
No	202 (94.0)	19 (90.5)	88 (94.6)	95 (94.1)
Missing	10	0	6	4
Cultural background, n (%)					.172
European	133 (64.9)	7 (41.2)	60 (67.4)	66 (66.7)
Asian	53 (25.9)	8 (47.1)	21 (23.6)	24 (24.2)
Other	26 (12.7)	2 (11.8)	10 (11.2)	14 (14.1)
Missing	20	4	10	6
Health authority, n (%)
Fraser health	79 (36.7)	11 (52.4)	24 (26.1)	44 (43.1)	.107
Vancouver coastal health	71 (33.0)	9 (42.9)	33 (35.9)	29 (28.4)
Vancouver island health	35 (16.3)	2 (9.5)	13 (14.1)	20 (19.6)
Interior health	29 (13.5)	1 (4.8)	15 (16.3)	13 (12.7)
Northern health	8 (3.7)	1 (4.8)	5 (5.4)	2 (2.0)
Provincial health services	7 (3.3)	0	7 (7.6)	0
First nations health	1 (0.5)	0	1 (1.0)	0
Missing	10	0	7	3
Practice location, n (%)					<.001
Private practice	152 (70.0)	17 (81.0)	58 (58.0)	77 (74.8)
Public health facility (e.g., hospital)	47 (21.7)	5 (23.8)	36 (38.7)	6 (5.8)
Long-term care/assisted living	14 (6.5)	0	2 (2.2)	12 (11.7)
Public recreation center	12 (5.5)	0	0	12 (11.7)
Missing	8	0	6	2
Days per week with 30+ minutes of physical activity, n (%)					.046
1–4	87 (40.5)	13 (65.0)	33 (35.1)	41 (40.6)
5–7	128 (59.5)	7 (35.0)	61 (64.9)	60 (59.4)
Missing	10	1	5	4

Note. Continuous variables (age, average hours worked per week, and years in profession) were compared between the three professional groups via a one-way ANOVA. Categorical variables (gender, sex, cultural background, health authority, practice location, and days per week with 30+ minutes of physical activity) were compared between the three professional groups using chi-square tests. For gender, sex, cultural background, health authority, and practice location, comparisons were limited to binary groupings—men vs women; male vs female; European vs other; Fraser Health vs Vancouver Coastal Health vs other; and Private Practice vs other, respectively. Variables with insufficient cell counts (n < 10) (lived experience as a Trans person and identified as LGBTQ2SIA+) were excluded from stratified analysis. Multiple responses were permitted for cultural background, health authority, and practice location; thus, cell counts for these variables could sum to more than column totals.

Outcome domains of older adult physical activity participation

Ratings

Figure 1 shows the proportion of health care professionals who rated each of the 24 candidate outcome domains as highly important (6 or 7 of 7). The five outcome domains most frequently rated as highly important by health care professionals were Falls (92.0%), Quality of Life (91.5%), Independence (87.9%), Balance (84.4%), and Mobility (82.7%). Conversely, the five outcome domains least frequently rated as highly important by health care professionals were Endocrine (30.8%), Nerve (35.0%), Immune (37.8%), Flexibility (39.3%), and Stress Management (48.0%). Results from a chi-square goodness-of-fit test with a Bonferroni-adjusted standardized residual analysis indicated that these 10 outcome domains were rated as highly important at rates that differed significantly from the expected uniform distribution.

Figure 1.

Percentage of participants who rated each outcome domain as “High” (6–7), “Moderate” (3–5), or “Low” (1–2) in importance for measurement in future randomized controlled trials of physical activity interventions for older adults. Note. Four outcome domain labels were shortened for figure presentation: Disease = Risk and Management of Chronic Disease, Injury = Risk and Management of Injury, Stress = Stress Management, and Psychological = Psychological and Emotional. Percentages were calculated based on the total valid responses per outcome domain (n = 221–225). Variation in n reflects incomplete survey responses, as detailed in Supplementary Table 1. Asterisks (*) indicate outcomes for which the proportion of “High” ratings was significantly different (p < .05) than expected based on chi-square analysis comparing “High” (6–7) vs combined “Moderate” and “Low” (1–5) ratings.

No significant differences were observed in the likelihood of an outcome domain being rated as highly important across profession (PT vs Kin), gender (men vs women), or age group (20–29, 30–39, 40+ years) (Supplementary Table 1).

Rankings

Figure 2 shows the proportion of health care professionals who ranked each of the 24 candidate outcome domains—or added an additional outcome domain—as being among the top four most important. The five outcome domains that most frequently appeared in the participants’ top four rankings were Quality of Life (67.1%), Independence (50.2%), Risk and Management of Chronic Disease (40.2%), Mobility (33.3%), and Falls (32.9%). The five outcome domains that least frequently appeared in participants’ top four rankings were Flexibility (0.5%), Endocrine (0.9%), Nerve (0.9%), Immune (1.4%), and Stress Management (2.3%). Results from a chi-square goodness-of-fit test with Bonferroni-adjusted standardized residual analysis indicated that 17 outcome domains were ranked among the top four most important significantly more (6 outcome domains) or less often (11 outcome domains) than expected.

Figure 2.

Percentage of participants who ranked each outcome domain among their top four most important to measure in future randomized controlled trials of physical activity interventions for older adults. Note. Four outcome domain labels were shortened for figure presentation: Disease = Risk and Management of Chronic Disease, Injury = Risk and Management of Injury, Stress = Stress Management, and Psychological = Psychological and Emotional. Outcome domains are ordered by frequency of selection. Colors represent the proportion of participants who ranked each outcome as their first, second, third, or fourth most important. Asterisks (*) indicate outcomes for which the proportion of Top four rankings was significantly different (p < .05) than expected based on a chi-square analysis comparing observed vs expected frequencies.

Four additional outcome domains were identified by participants. For clarity, these were coded into the closest relevant prespecified outcome domains (i.e., “previous injury” was categorized under Risk and Management of Injury; “multimodal outcomes” under Quality of Life; “functional goal achievement” and “function” under Independence).

Across professions, PTs ranked Risk and Management of Injury as significantly less important than Kins (16.7% vs 37.9%, p = .036) (Supplementary Table 2). Outcome domain ranking of Risk and Management of Injury also differed significantly (p = .019) by age group: health care professionals aged 20–29 years ranked it in their top four most important 45.9% of the time, compared to 27.1% among those aged 30–39 years and 17.1% among those aged 40+ years. No statistically significant differences in outcome domain rankings were noted across the genders represented in this study.

Promoting quality of life and independence

Health care professionals identified multiple physical activity outcome domains as most likely to impact older adults’ ability to maintain independence (Table 2). The five most frequently selected were Strength (65.5%), Balance (56.5%), Mobility (52.5%), Brain and Cognitive (30.0%), and Falls (23.5%).

Table 2.

Outcome domains of physical activity that health care professionals believed to be most impactful for older adults maintaining their independence and quality of life.

Independence		Quality of life
Outcome domains	Count (%) (n = 200)	Outcome domains	Count (%) (n = 204)
Strength	131 (65.5)	Social	92 (45.1)
Mobility	113 (56.5)	Psychological	74 (36.3)
Balance	105 (52.5)	Mobility	73 (35.8)
Brain and cognitive	60 (30.0)	Brain and cognitive	63 (30.9)
Falls	47 (23.5)	Sleep	63 (30.9)
Endurance	43 (21.5)	Enjoyment	62 (30.4)
Disease	31 (15.5)	Independence	60 (29.4)
Injury	29 (14.5)	Strength	56 (27.5)
Energy	23 (11.5)	Disease	54 (26.5)
Flexibility	18 (9.0)	Loneliness	47 (23.0)
Psychological	17 (8.5)	Energy	45 (22.1)
Muscle	17 (8.5)	Balance	41 (20.1)
Quality of life	15 (7.5)	Stress	40 (19.6)
Cardiovascular	15 (7.5)	Injury	36 (17.6)
Social	14 (7.0)	Falls	26 (12.7)
Activities of daily living	13 (6.5)	Endurance	23 (11.3)
Bone	10 (5.0)	Cardiovascular	19 (9.3)
Nerve	9 (4.5)	Flexibility	16 (7.8)
Enjoyment	8 (4.0)	Muscle	15 (7.3)
Sleep	8 (4.0)	Bone	11 (5.4)
Immune	7 (3.5)	Immune	11 (5.4)
Loneliness	6 (3.0)	Endocrine	8 (3.9)
Endocrine	5 (2.5)	Nerve	8 (3.9)
Stress	5 (2.5)	Activities of daily living	4 (2.0)

Note. Four outcome domain labels were shortened for table presentation: Disease = Risk and Management of Chronic Disease, Injury = Risk and Management of Injury, Stress = Stress Management, and Psychological = Psychological and Emotional.

Health care professionals also identified several outcome domains as likely to impact older adults’ ability to maintain their quality of life (Table 2). The five most frequently identified outcome domains by participants were Social (45.1%), Psychological and Emotional (36.3%), Mobility (35.8%), Sleep (30.9%), and Brain and Cognitive (30.9%).

Concerns about older adult engagement in physical activity

Health care professionals were also asked what concerns they might have when recommending or prescribing physical activity to older adults. Reported concerns were grouped into four themes. Of these, two themes—clinical and safety concerns—reflected perceived risks of potential adverse events associated with physical activity. Clinical concerns (health conditions that may interact with or complicate physical activity) were most frequently reported, with subthemes including general comorbidities or contraindications (36.9%), cardiovascular or pulmonary concerns (19.4%), cognitive concerns (7.8%), pain or fatigue concerns (6.9%), muscle, bone, or joint concerns (6.5%), and medication interactions (3.2%). Safety concerns (immediate risks of engaging in physical activity) were also highly prevalent, including falls or balance concerns (34.1%) and general safety or injury risk (33.6%). Behavioral concerns reflected providers’ doubts about older adults’ ability to engage in physical activity consistently or to an appropriate standard and included support needs or self-regulation concerns (18.4%) and compliance or adherence concerns (11.1%). Finally, some health care professionals reported no concerns—either outright (1.8%) or with the caveat that physical activity was appropriately individualized (9.2%). Despite any concerns they indicated, 100% of participants reported that they do recommend physical activity to their older adult patients or clients.

Discussion and implications

To support the development of a COS for RCTs of physical activity interventions for older adults, the purpose of this study was to provide insight into which outcome domains associated with older adult physical activity participation are deemed most important by health care professionals who promote and prescribe physical activity to them, including GPs, PTs, and Kins. Notably, four outcome domains—Quality of Life, Independence, Falls, and Mobility—ranked among the top five most important across both rating and ranking questions. Each was rated as highly important by over 80% of participants and selected among the top four most important by at least 30% of participants. In addition, Risk and Management of Chronic Disease was less often rated as highly important (73.2%) but ranked among the top four outcomes by 40.2% of participants. Conversely, Balance was frequently rated as highly important (84.4%) but was less frequently ranked among the top four outcomes (22.8%).

Broad and holistic outcome domains such as Quality of Life and Independence were likely prioritized because they encompass multiple other outcome domains. For example, Quality of Life may be prioritized over more specific outcome domains such as Psychological and Emotional or Social outcome domains, as it encompasses and effectively supersedes them. This pattern of outcome domain nesting is consistent with previous findings from our laboratory, which showed that older adults also prioritize holistic outcome domains like Quality of Life and Independence from physical activity (Young et al., 2024). Broader concepts have also been theorized to be more salient and easily envisioned by health care professionals, thereby promoting their perceived importance (Festen et al., 2021).

Independence and Quality of Life were previously identified by older adults as the most important outcome domains because they encompass multiple others (Young et al., 2024). To build on this understanding of outcome domain nesting and summarize broader patterns, we examined which outcome domains health care professionals perceived as most influential in promoting Independence and Quality of Life in older adults. For maintaining Independence, seven of the ten most frequently cited outcome domains were physical in nature, based on our thematic outcome domain mapping (Young et al., 2024); these were Strength, Balance, Mobility, Brain and Cognitive, Endurance, Flexibility, and Energy. Three of the ten were clinical in nature; these were Falls, Risk and Management of Chronic Disease, and Risk and Management of Injury. In contrast, the outcome domains found to be most influential in promoting the Quality of Life of older adults were more diverse. The ten most frequently cited outcome domains for maintaining Quality of Life included three physical outcome domains (i.e., Mobility, Brain and Cognitive, and Strength), three psychological outcome domains (i.e., Psychological and Emotional, Sleep, and Enjoyment), two social outcome domains (i.e., Social and Loneliness), one clinical outcome domain (i.e., Risk and Management of Chronic Disease), and one overarching outcome domain (i.e., Independence). This distinction highlights how health care professionals view physical and clinical outcome domains as central to maintaining Independence, while maintaining Quality of Life is seen as requiring a more multidimensional set of outcome domains. Notably, health care professionals most frequently identified the social outcome domain as key to maintaining Quality of Life. This finding corroborates existing literature on the multidimensional nature of Quality of Life (Phillips et al., 2013) and aligns with our previous recommendation to develop higher-quality measures of social outcomes and use them more frequently in RCTs examining physical activity interventions for older adults (Young et al., 2024).

Falls and Mobility were also identified as two of the most highly prioritized outcome domains. Falls and mobility limitation are well recognized as significant sources of morbidity for older adults that interfere with independence and quality of life (Pahor et al., 2014; Sherrington et al., 2019). While a growing body of evidence suggests that certain physical activity interventions can reduce the incidence and severity of falls, there is still uncertainty about the efficacy of some types of exercise interventions, such as walking programs, due to limited evidence (de Souto Barreto et al., 2019; Sherrington et al., 2019). This highlights the need for consistent outcome reporting across trials of physical activity interventions. Similarly, there is strong evidence that mobility is a modifiable outcome in older adults, and mobility disability can be prevented through engagement in structured physical activity (Pahor et al., 2014). Continued reporting on mobility outcomes in physical activity trials is essential to contribute to a stronger evidence base about physical activity interventions that are relevant to health care professionals.

We also explored whether outcome domain priorities differed by profession, gender, or age group. Outcome domain ratings were consistent across all subgroups, with no significant differences observed between PTs and Kins, men and women, or across age categories. Outcome domain rankings were also largely consistent across all subgroups, with only a couple of exceptions; namely, Risk and Management of Injury was ranked among the top four most important outcome domains less frequently by PTs than by Kins. Younger professionals also rated Risk and Management of Injury as more important compared with more experienced practitioners. Their views on aging may be more related to socialization and less to their own developing aging journey (Kornadt et al., 2019). Such socially shaped perspectives may unintentionally reinforce deficit-based views of aging, including assumptions of frailty or injury risk (Levy & Banaji, 2002). It is also possible that early-career health care professionals have had limited opportunities for sustained positive intergenerational contact, which may shape perceptions around aging-related risks (Henry et al., 2024). This finding aligns with calls to promote positive intergenerational contact within health care professional training contexts as a way of fostering richer experiential understanding of aging (Henry et al., 2024).

The results of this study will directly inform the selection of outcome domains for a COS (Mackey et al., 2025), with the aim of improving consistency in outcome reporting and enhancing the relevance of evidence from RCTs of physical activity interventions for older adults (Clarke & Williamson, 2016; Williamson et al., 2017). Our work responds to calls from COS creation frameworks to explicitly involve health care professionals in the generation and refinement of outcome domains (Kirkham & Williamson, 2022; Kirkham et al., 2017; Williamson et al., 2017). Specifically, the results of this study align with established evidence that overarching outcome domains such as Quality of Life and Independence, as well as more specific outcome domains like Mobility and Falls, are relevant to health care professionals in the context of outcome selection and are important to consider for COS creation. These outcome domains are in line with previous research based on the outcome domain preferences of older adults (Young et al., 2024). Creating more timely, consistent, and relevant research on older adult physical activity may help to attenuate some of the current barriers to physical activity promotion among health care professionals—namely insufficient knowledge and limited practical resources—by enabling more efficient evidence comparison and synthesis and increasing the clinical relevance of RCT findings (Barton et al., 2021; Green, 2008; Petrella et al., 2007; Vuori et al., 2013). Before a COS is available, the results of this study will serve as a resource for outcome selection in RCTs of physical activity with older adults to help ensure that trial results are clinically relevant.

Our work also responds to increasing calls within the literature, including guidance from PRISMA and CONSORT, for more consistent and transparent reporting of adverse event outcomes or harms of an intervention (Junqueira et al., 2023; Zorzela et al., 2016). When asked to identify any concerns they might have in recommending physical activity to their older adult clientele/patients, health care professionals frequently cited concerns related to a perceived risk of experiencing an adverse event. The most frequent concerns were related to management of clinical conditions and safety. Interestingly, the potential for injury is not mentioned in the literature as being a barrier for physical activity promotion (Barton et al., 2021; Vuori et al., 2013). This suggests that while health care professionals acknowledge some risks of physical activity for older adults, they may view the benefits of physical activity as outweighing them (O’Brien et al., 2017; Reid et al., 2022). Improving the transparent and consistent reporting of adverse events that are relevant to health care professionals will strengthen their confidence in recommending physical activity interventions for older adults.

Health care professionals also had several behavioral concerns around older adults’ ability to sustain quality independent engagement in physical activity. These concerns may reflect limited evidence related to adherence and sustainment following physical activity interventions, particularly after intervention completion (Dzewaltowski et al., 2004; Estabrooks & Glasgow, 2006). Our findings of behavioral concerns align with calls to consider additional maintenance-related outcomes (e.g., adherence) in RCTs (Hodgson et al., 2023).

The results of this study are novel in that they integrate and prioritize a comprehensive set of physical activity outcome domains for older adults, as they are perceived by health care professionals who are involved in physical activity promotion and prescription. Previous studies have made valuable contributions by asking health care professionals to rank small subsets of outcomes (i.e., Independence, Staying alive, Pain Relief, and Symptom relief) in specific clinical settings (Festen et al., 2021; Seghers et al., 2023). Likewise, other research has shown health care professionals’ understanding of some physical activity benefits for older adults, such as mobility, strength, and well-being (Giri et al., 2025). To the best of our knowledge, the current study is the first to comprehensively combine rating, ranking, and subgroup analysis of outcome domain priorities among health care professionals in the context of older adult physical activity. Future research should aim to bring together the perspectives of researchers, health care professionals, and older adults to identify shared priorities for outcome domains for inclusion in a COS.

This study has certain limitations. First, the use of a voluntary response survey design may have introduced self-selection bias, as participants may have had a preexisting interest in or higher perceived value of physical activity. Second, we recruited only a small number of GPs despite an extensive recruitment period and multiple recruitment strategies; thus, it is uncertain whether our sample represents the broader GP population in BC, and we lacked statistical power to test differences in outcome importance between GPs and other included health care professional groups. Third, restricting recruitment to BC, one province in Canada, may limit the generalizability of the results to other settings. It seems likely that health care professionals’ preferences around priority outcome domains are largely influenced by both educational background and current clientele. In BC, roughly one-third of PTs are educated internationally, which may contribute to diversity in educational perspectives (Canadian Institute for Health Information, 2024). Conversely, BC’s older adults share unique traits among Canadians, such as the lowest prevalence of chronic diseases in Canada, which may shape how health care professionals perceive certain outcomes (Basham, 2020). Given the diversity of educational training and professional background among health care professionals who promote and prescribe physical activity to older adults, a COS will be especially important to support a more consistent evidence base about older adult physical activity that can inform the training and practice of such health care professionals.

In conclusion, health care professionals prioritized several outcome domains to measure in RCTs of physical activity interventions for older adults. The most consistently prioritized across both ratings and rankings of importance for GPs, PTs, and Kins were Quality of Life, Independence, Falls, and Mobility. These findings can inform outcome selection for future RCTs of physical activity interventions for older adults and provide an important foundation for developing a grounded, relevant COS in this area.

Supplementary material

Supplementary material is available at Innovation in Aging online.

Funding

This work was supported by the Drummond Foundation (research grant to D.C.M.); Michael Smith Health Research British Columbia (scholar award to D.C.M.); and the Canadian Institutes of Health Research (Canada Graduate Scholarship—Master’s to P.J.Y.).

Conflicts of interest

None declared.

Data Availability

Data will be available upon reasonable request after consultation with the authors. This study was not preregistered.