Introduction

Physical activity and disease-specific physical activity recommendations have been created to provide a guideline on the mode, frequency, and intensity of physical activity, and how those factors can be modified in the presence of a chronic condition. For example, the Too Fit to Fracture physical activity and exercise recommendations [1, 2] provide guidance on effective exercise for fall and fracture prevention, and safe performance of physical activities of leisure and daily living for people with osteoporosis. Publication of guidelines or recommendations does not always reach patients. Therefore, knowledge translation (KT) interventions may be required to disseminate knowledge and change behavior among knowledge users.

To develop KT interventions, the Knowledge-to-Action cycle [3] suggests adapting knowledge to the context and identifying barriers to implementation. The Knowledge-to-Action cycle is a framework composed of two components: knowledge creation and the action cycle. The current study focused on the action cycle which is an iterative, dynamic process that suggests translating knowledge to action by identifying the problem, adapting the knowledge, assessing barriers, tailoring intervention, monitoring knowledge, evaluating outcomes, and sustaining the knowledge [3]. Barriers to and facilitators of exercise [4, 5] and physical activity [6] in older adults have been identified, and chronic health conditions have been cited as a major barrier to initiating and maintaining exercise [1]. Among individuals with osteoporosis, lack of interest, time, and knowledge, fear of falling, and comorbid conditions were the major barriers to physical activity [7], and improving health, feeling in control, and past experience were identified motivators to engaging in physical activity [7]. While previous work provides insight on barriers to participating in physical activity in general, it did not assess barriers related to implementing specific exercise recommendations [1]. Further, there is a need to move beyond generating lists of barriers and understand how barriers can be addressed in theory-informed KT interventions.

The Behaviour Change Wheel (BCW) [8] is a framework (see Fig. 1) designed to inform interventions or policy development based on barriers to knowledge uptake. The BCW theoretical model posits that Capability, Opportunity, and Motivation govern Behaviour (COM-B) [9]. By identifying barriers to behavior change using the COM-B model, we can inform a “behavioral analysis” of what needs to change, and guide the selection of intervention functions and policy categories. At the center of the Behaviour Change Wheel is the COM-B model, around which are nine intervention functions to address the target behavior and seven policy categories to enable the interventions [9]. As part of a multifaceted KT initiative, the objectives of the current study were to (1) identify barriers to and facilitators of implementing exercise and physical activity recommendations among middle-aged and older adults, using Too Fit To Fracture as the example, (2) perform a behavioral analysis, and (3) identify intervention and policy options using the COM-B model [9].

Fig. 1
figure 1

Behaviour Change Wheel (BCW) framework for designing interventions or policies based on barriers to knowledge uptake. BCW was used to develop an informed behavioral analysis of patient’s uptake of exercise and physical activity recommendations

Full size image

Methods

Study setting and framework

The current study is one part of a KT initiative conducted in Ontario, Canada, in partnership with Osteoporosis Canada. Our study represents the Knowledge-to-Action Cycle step, “assess barriers to knowledge use,” and the initial phases of the “selection, tailoring, and implementation of interventions” step [3]. To collect patient perspectives, a qualitative study design was employed. Focus group and one-on-one interviews were the primary source of data collection. We used the BCW [8] as a theoretical model to guide our identification of future KT interventions. The BCW is a non-linear framework that links essential conditions necessary for behavior change (capability, opportunity, and motivation) to interventions aimed at addressing deficits in the behavioral conditions, and to policies appropriate for facilitating the interventions to occur.

BCW step 1—define the problem in behavioral terms

A multidisciplinary team that included a primary care physician, specialists in geriatrics and internal medicine, physical therapists, kinesiologist, and a patient advocacy organization representative was assembled to define the problem in behavioral terms and outline the system of influencing behaviors. The “problem” or desired behavior was defined by the team as uptake of the Too Fit To Fracture recommendations [1, 2] (from here referred to as the recommendations) by individuals with osteoporosis. Examples of recommendation-driven behavior changes are that individuals adopt spine sparing strategies and participate in a multicomponent exercise program that includes aerobic physical activity and exercises to improve muscle strength, balance, and posture [1, 2].

BCW steps 2 and 3—select and specify the target behavior

A patient’s behavior can be shaped by a system of behaviors, such as health care provider or exercise professional behavior (e.g., counseling, referrals; Fig. 2). The focus of the current behavioral analysis is the patient; we have previously published our consultations with physicians [10], and our analysis of barriers experienced by exercise professionals will be presented separately, as we hypothesize that some or all of the barriers experienced by patients will be unique, leading to patient-direct knowledge translation interventions. To select and specify the candidate behaviors, we considered options such as participating in exercise classes consistent with the recommendations, consulting an exercise professional with knowledge of the recommendations, or uptake of education and resources to self-manage osteoporosis with exercise and adopting spine sparing in daily life. The following criteria were used to prioritize the behavior: the likelihood that the behavior could be changed, the impact of the behavior change, the potential that the change would spill over to other behaviors, and the potential to measure the change. The research team considered the contexts in which the behavior(s) could occur, what behaviors needed to be performed and by whom, when and where they needed to do it, and with whom (e.g., which patients, or with which supporting health care professionals or services) (Fig. 2) [8].

Fig. 2
figure 2

Translating knowledge through a system requires consideration of a group of potential knowledge users. In the center, the desired behavior change is of the patient, but organizations, health care providers, social networks, and community opportunities may influence the patient’s behavior change

Full size image

BCW step 4—identify what needs to change (barriers and facilitators)

To finalize the selection (BCW step 2) and specification (BCW step 3) of behaviors, and inform what needs to change (BCW step 4), we conducted concurrent descriptive, qualitative studies with people with osteoporosis. Participants in this study were community-dwelling individuals living in Ontario who agreed to participate in knowledge exchange related to osteoporosis and physical activity. Our recruitment strategies were designed to ensure representation of individuals with interest in osteoporosis management, for whom the physical activity recommendations were salient. Osteoporosis Canada regional integration leads, which are individuals working for Osteoporosis Canada that implement programs through collaborative partnerships in communities across Ontario, planned focus groups or large group events and advertised the study to the Canadian Osteoporosis Patient Network (COPN). COPN is an electronic network of individuals who have underlying osteoporosis. Eight different Local Health Integration Networks (LHINs) in Ontario were selected for from both urban (100,000 inhabitants [11]) and rural communities (1000–99,999 inhabitants [11]). A LHIN is a publically funded health care service that works to plan, integrate, and improve health care access and patient care [12]. It was evident that males and individuals in northern Ontario were under-represented in the sample; therefore, targeted advertisements were distributed via COPN [13]. We were able to recruit men, but no individuals from northern Ontario responded to our recruitment attempts.

We used a survey to collect demographic information (e.g., age, sex, and contact information) and to ask whether they had heard of the recommendations. Participants at later sessions were asked about their current physical activity levels. Each interview, focus group, or large group session included a presentation of the recommendations, presented by one or two researchers (CM and/or LG). The purpose of the presentation was to provide context for the questions about implementing the recommendations. Focus groups were held in groups of 6 to 12 participants per moderator. In some regions, a knowledge exchange was held with a larger number of attendees, because of interest in the topic. Therefore, a large group interactive discussion with question and answer periods was conducted. The lead moderator (CM or LG) gave attendees small group discussion questions, and attendees could then share their perspectives with the larger group. We included transcripts from large group discussions among the focus group and interview data to ensure input from a more diverse group of knowledge users, and to compare/contrast emerging themes. Discussions or interviews were audio recorded and transcribed verbatim.

A semi-structured interview guide was used to prompt discussion about barriers to and facilitators of implementing the recommendations. Throughout data collection, it was noted that asking about personal experiences and goals provided more insight on barriers and facilitators than the explicit questions. Therefore, the interview guide was modified. Instead of asking what made exercising easier and harder, we asked the participant to describe a situation they were successful and unsuccessful in meeting an exercise goal. We also asked about the value of a Bone Fit™-trained instructor. Bone Fit™ (bonefit.ca) is a workshop for health care and exercise professionals on assessment and exercise prescription for osteoporosis [14]. Additionally, perceived knowledge gaps emerged regarding the definition of strength, balance, and aerobic exercise, so the interview guide was modified to probe those gaps.

Theory-guided thematic analyses were performed by two members of the study team (CZ and CM). A third party resolved disagreements if they occurred. Data was first coded for major categories of information, where a category was defined as “concepts that pertain to the same phenomenon” [15]. Relationships and similarities among categories were discussed leading to the formation of themes. Thematic analyses were guided by the central behavior change model of the BCW, which posits that Capability (e.g., physical, perceived), Opportunity, and Motivation govern Behaviour (COM-B model) [8]. Each component is further divided into two types: capability can refer to the physical capability (e.g., physical skills, stamina) or psychological capability (e.g., knowledge, cognitive skills); opportunity can refer to the physical opportunity (e.g., resources, physical barriers, time) or social opportunity (e.g., norms, social cues); and motivation can refer to reflective motivation (e.g., goals, intentions, planning, beliefs) or automatic motivation (e.g., emotional or impulsive responses, habits, wants, and needs) [8]. Major themes were categorized into the most appropriate BCW component.

BCW steps 5 and 6: identify intervention options—intervention functions and policy categories

The BCW provides a guide for using the identified COM-B components to identify which strategies, among nine potential intervention functions and seven policy categories, would be expected to be effective. Candidate intervention functions and policy categories, and whether they meet the APEASE criteria (affordability, practicable, effectiveness/cost-effectiveness, acceptability, side-effects/safety, equity), are described [8]. BCW steps 7 and 8 will integrate consultations from multiple target groups (Fig. 2) and will be described elsewhere.

Results

BCW steps 2 and 3—select and specify the target behavior

Thematic analyses of patient consultations and team meetings reinforced the need to choose behaviors that consider patient preference, willingness to pay, and comorbid conditions. Among the potential modifiable patient behaviors, the following were selected as targets (Fig. 2):

  • Patient seeks advice on how to self-manage osteoporosis with a multicomponent exercise program consistent with the recommendations, and how to adopt spine sparing strategies in daily life

  • Patient considers seeking advice from a physical therapist or exercise professional that has expertise in applying the recommendations

BCW step 4—identify what needs to change (barriers and facilitators)

A total of 240 individuals, mean age 72 ± 8.3 years, participated in the study, from 12 cities in eight LHINs (Table 1). Emergent themes fell under each of the categories of the COM-B model, and were mapped to intervention functions (Table 2) and policy categories (Table 3).

Table 1 Demographic information of respondents and pre-survey response
Full size table
Table 2 Mapping themes to intervention functions using the BCW framework
Full size table
Table 3 Using the BCW framework to identify policy categories to deliver interventions
Full size table

Physical capability—theme: disease-related symptoms hinder exercise and physical activity participation

The presence of comorbidities was a reported barrier to participation in physical activity or exercise. For example, cancer survivors found the side effects of medications made physical activity very difficult. The lasting effects of chemotherapy treatment gave “symptoms, like hot flashes. Sometimes they [could] be so bad that when [the participant] was in cardiac rehab, [he] passed out.” Other chronic conditions such as fibromyalgia or chronic obstructive pulmonary disease were identified as limiting physical activity and making aerobic physical activity particularly challenging. Pain was reported as a barrier, such that it hindered even highly motivated participants from engaging in regular physical activity. One participant described that he “[had] been exercising five times a week with a regular program since [he] was 35 years… until [he] fractured his vertebrae and then couldn’t do any of those exercises anymore…[because] if you have pain as [he does] you can’t do [the exercises].” Alternative modes of exercise were facilitators to maintaining physical activity for participants that experienced pain due to arthritis, such as water aerobics because “the water was warm and [the participant felt] it was very good for [her].”

Psychological capability—theme: physical activity knowledge gaps

Participants reported specific gaps in knowledge pertaining to strength training, such that the participants “have never been told what exercises to do, and what not to do”, and asked “what’s the maximum for lifting, will I push myself too hard?” and “how [do I] modify the exercises?” When asked to describe a back extensor exercise, participants incorrectly provided examples such as bringing their knees to their chest while lying on their back. Good alignment and posture correction cues were poorly understood. Participants requested images to understand proper alignment. A subtheme was an emphasis on knowledge pertaining to the safety of physical activity. Participants not only wanted to know what exercises were effective and how to do them, but “what’s safe to do at home”, and emphasizing the importance to learn safe performance of daily activities:

I think it’s really important to learn how to do daily chores properly. For example, putting dishes in the dishwasher, or brushing your teeth, carrying your garbage out, lifting or picking up anything [because] those are the things you’re doing already.

Obtaining information on how to perform exercise and daily activities was a challenge reported by participants. The demonstrations that accompanied the knowledge exchange resulted in affirmations or statements suggesting that modeling is a facilitator, e.g., “what was good was when she demonstrated the hip hinge.”

Beyond a lack of knowledge on exercise concepts and how to perform activities of daily living, participants also expressed a lack of knowledge on seeking resources related to exercise. Participants wanted to know where to find qualified instructors, but also wanted to know where to get tools to aid in learning safe exercises. A summary of knowledge gaps is presented in Table 4.

Table 4 Physical activity knowledge gaps
Full size table

Physical opportunity—theme: access to exercise programs that meet needs and preferences

Available, accessible community exercise programs emerged as a facilitator of participation in exercise consistent with the recommendations. Many participants enjoyed group exercise classes for the social connections and considered them “a good way to teach people a safe way to do the exercise.” Group classes were described to “increase [participant’s] commitment… [and] would actually force [them] to exercise.” However, group classes did not address all participants’ needs or health conditions, provide sufficient progression, or individualized exercises. Some participants reported that they were dependent on classes; one participant explained that “left to my own devices, I don’t know.” However, participants felt frustrated when talking about available publicly funded exercise programs, explaining that the programs were difficult to get into, and were too short: “in another two weeks [I get] to see if I can get in the program again, but then they give me six weeks and dump me again.” Enrolling in a fall prevention program was described as follows: “All you have to do is fall… once you fall, then you get to find out [about the program].”

Physical opportunity—theme: limited resources and time

Financial constraints posed a barrier to participation in therapeutic exercise. Participants wanted to consult an exercise professional with expertise in fall and fracture prevention, but that “physiotherapists and kinesiologists are expensive.” Participants wanted self-management tools without costs. Limited group exercise class times and transportation constraints (e.g., the bus schedule does not coincide with the class time) made attending group classes difficult for some participants. Transportation was a greater concern for participants in rural communities, where public transit was not always an option. Many participants reported having constraints on their time, such as doctor’s appointments, or maintaining their home.

Social opportunity—theme: gender-related norms and physical activity preferences

The importance of social opportunity also emerged, such that interventions need to consider how social norms and patient preference influence behavior. Gender identity or roles may influence behavior. Male participants reported a general disinterest in group classes, or “sissy exercises.” Men reported that activities such as sports or chores around the house were how they preferred to be physically active. One man explained that stubbornness and denial led him to continue what he enjoys doing. He said: “I like to sail, I like to golf. I can’t imagine two stupider things to do with osteoporosis. And yet I love to do both.” Another participant agreed that golf was a hobby and would prefer to be taught how to modify the sport rather than quitting: “for golf I was told not to bend, but today she said that we should bend from the hip but no one told me this before.”

Opportunity-related barriers might also be represented by gender roles, e.g., maintaining their home, caring for family members: “my mom and my husband aren’t well. So you don’t get time in your day for yourself.”

Many participants described seeking exercise advice from their family physicians as a social norm, saying that they “don’t see anyone else apart from [their] family doctor.” However, participants also recognized that “family doctors are really busy and they’re not up to date on the latest data.” Therefore, this may be a missed social opportunity. Indeed some participants recognized that family doctors might not be the best health professional to provide exercise advice, knowing that many family doctors do not have training in exercise counseling, nor do they have sufficient time. Instead, participants that seek advice from an allied health professional “feel a lot more confident taking a class that was taught by someone Bone Fit™ trained and they knew what they were talking about.”

Reflective motivation—theme: personal and social incentives to exercise

Participants reported needing personal incentives to exercise, such as observing health benefits, maintaining independence, or feeling good after exercise. External motivators, such as a pet, the accountability of a class, or an exercise partner facilitated exercise participation. Group exercise classes created incentives by establishing accountability and social connections. Group classes were a motivating strategy because “you have the responsibility to show up.” Creating a sense of responsibility to exercise with at-home strategies such as “dancing,” “writing it down for reinforcement,” “[having] someone that has the same problems to talk to,” or establishing plans or triggers, e.g., “doing balance while doing the dishes,” were described as facilitators.

Group exercise may also create anti-incentives. For example, group classes were not unanimously preferred, and some participants found “it easier to do it on [their] own, [as] it can be embarrassing.” Many participants felt uncomfortable in group exercise classes because of body image concerns, or wanted an exercise demonstrator that they could relate to. For example, they felt that a young, fit exercise instructor might not be relatable or appreciate their physical limitations.

Automatic motivation—theme: fear of injury

The participants expressed a lack of confidence and fear of exercising with osteoporosis. One participant explained that “[she] can’t do the exercises at home … because [she’s] afraid of doing them wrong, to fall….” A history of exercise participation may be a facilitator to beginning an exercise program designed for people with osteoporosis, as a respondent explained that exercise is “something you have to start early.”

The fear subtheme may also influence psychological capability, such that their lack of confidence and exercise self-efficacy is limiting their ability to initiate or adapt exercise, simply because they do not know what exercises are safe to do. Therefore, motivation and psychological capability may be inter-related for people with osteoporosis.

Fear of injury related to environmental conditions reoccurred as a subtheme. During winter months, participants described more time spent inside and less time going to their exercise classes for fear of slipping on ice or snow: “I haven’t been able to do too much because of the concern of ice… so I haven’t ventured out much.”

Age may influence risk perceptions and associated decision-making. Some of the older women with osteoporosis expressed a fear of fracturing and falling and were more resistant to uptake an exercise routine, suggesting fear is a demotivating factor to exercise. However, younger women that have known someone with osteoporosis may use the fear of frailty as a motivator to learn to exercise safely with osteoporosis. A younger women explained that her motivation to learn how to safely exercise was her “mother had osteoporosis and she lost a lot of height and wasn’t given any treatment… I was diagnosed in my 50’s and have been treated for a long time now.” She wanted to learn more about what she can do to prevent the physical changes she had observed in her mother.

A common barrier was not having a trustworthy exercise instructor or physical therapist, which is parallel with the theme pertaining to fear. Participants felt that exercise instructors “do not have the appropriate training in the management and prevention of falls and fractures” and that “there is a lack of trained professionals.” One participant explained that “[he] was injured by a physiotherapist and hurt [his] back and it set [him] back so not all of them know what they are doing and know what to do with a compression fracture in the spine.” After describing the Bone Fit™ program to the participants, some expressed interest in finding a Bone Fit™-trained instructor in their area.

BCW steps 5 and 6: identify intervention options—intervention functions and policy categories

We used BCW to map the barriers to potential intervention functions. However, not all intervention functions meet the APEASE criteria. Based on the identified barriers, whether there is a need to change the barriers, and the APEASE criteria, the following intervention functions were selected: training, education, and modeling (Table 2). The intervention functions selected mapped onto potential policy categories and using the APEASE criteria, the following policy categories were selected: communication/marketing, guidelines, and service provision (Table 3).

Discussion

We identified barriers and facilitators to implementation of physical activity and exercise evidence, including several that may be unique to individuals with osteoporosis, such as fear of injury, trust in health care professionals, and physical activity knowledge gaps. Other barriers that emerged included the following: comorbid chronic disease-related symptoms, physical activity knowledge gaps, access to programs and trained professionals, limited time and resources, physical activity norms and preferences, personal and social incentives or planning, and patients’ fear of injury. We identified training, education, and modeling as intervention functions, and communication/marketing, guidelines, and service provision as policy categories. Our next steps are to integrate consultations with patients, physicians, and exercise professionals to develop knowledge translation tools [16].

Although some of the barriers to physical activity participation that we identified have been previously reported [17], barriers that may be unique to individuals with osteoporosis emerged. Fear of fracturing and time constraints were previously cited as major barriers to physical activity for individuals with osteoporosis [17,18,19]. A unique finding was that patients’ fear of fracturing led to a lack of trust in health care providers or exercise professionals. Interestingly, our interviews with health care professionals also revealed a fear of referring patients who were frail or at risk of fracture to exercise professionals [10], suggesting that any KT strategies need to instill trust in both patients and health care providers. Other examples of physical activity barriers unique to our study include certain physical activity knowledge gaps (e.g., how to translate good alignment to activities of daily living) and how to modify exercise for comorbid conditions. Participants also reported not wanting to lift heavy weights out of fear of fracturing. However, moderate intensity progressive resistance training, with an emphasis on form and alignment, is encouraged for individuals with osteoporosis [2]. Unlike previous studies, we asked about barriers in the context of specific types, frequencies, and intensities of exercise revealing gaps in knowledge about exercise terminology or practices. Many of our participants lacked knowledge on engaging back extensor muscles, correct posture, and volume and load for resistance training exercises, which are key components of the physical activity recommendations for osteoporosis [1, 2]. Therefore, translation of research needs to address patient knowledge about exercise terminology, and fear among both patients and providers, such that we provide exercise professionals and providers with the basic knowledge to tailor advice on safe movement and physical activity to the patient.

Another unique finding emerging from our work is that knowledge translation interventions and programs may need to consider preferences or determinants of health inequities, or their intersectionality [20]. Examples identified in our consultations with patients include gender, cost, age, geography, multi-morbidity, and social norms. Health inequity and intersectionality may influence communication strategies, modeling, the choice of language used, the strength of the persuasion of the communication strategy, and the way promotional information is processed [21]. For example, marketing strategies segment sex and gender because subtle changes in language and emotional appeal will have different effects on men and women’s attitude to the advertisement and influence purchase intentions [22]. Communication strategies should consider intersectionality, with varying peer models and address fears, gender-related roles, identities, or preferences. For example, men in the current study reported that they preferred sport-related exercise rather than attending a group exercise class. A KT strategy might address risk perceptions in men, while also providing examples of implementation strategies (e.g., how to golf safely) that are relevant to men, using peer models.

Using the BCW, we identified guidelines, service provision, and communication and marketing as policy categories that could be used to translate physical activity or exercise recommendations. Guidelines have been established [1, 2], but selecting behavior techniques that enable patient-centered tailoring and implementation is the next step. Service provision via exercise classes or exercise professionals that have expertise related to exercise and osteoporosis may address both knowledge and fear. Supervised exercise strategies such as 1:1 telephone counseling and professional advice have been shown to improve short-term physical activity participation [23]. However, the high demand on support personnel may not be feasible or create long-term exercise behavior change [24, 25]. Traditional educational sessions using both written [26, 27] and visual [27] materials may increase patient knowledge. Physicians have identified a preference for strategies such as using a pamphlet, a one-page handout, or a list of qualified individuals for referral [10]. We may need to consider innovative and cost-effective ways to deliver exercise services. For example, an exercise program called Seniors Maintaining Active Roles Together (SMART), run by Victorian Order of Nurses for Canada (VON), and implemented by volunteers, teaches older adults a set of 15 exercises both in home and group exercise programs (www.von.ca). Participants described the program as a “valuable asset for people” and that they were more inclined to exercise on their own after learning the exercises [28], suggesting an increase in exercise knowledge. Communication strategies can also be used to enhance knowledge translation. Social media [29, 30] or multimedia interventions [31, 32] have been shown to be effective at translating knowledge and improving disease self-management.

Strengths and limitations

Strengths of the current study were the diversity of participants with respect to region, age, and gender. The depth of our inquiry as it pertains to sources of fear and physical activity knowledge gaps resulted in unique findings, and addressed exercise and physical activities of leisure and daily living rather than in general. We also applied a behavior change framework to link identified barriers to knowledge translation interventions. Participants were recruited through a convenience sample, which may introduce sampling bias. Multiple interview methods were used, which might be a limitation contributing to variability in depth and breadth of data collected. However, there appeared to be no differences in emerging themes across the different modes of data collection. Another limitation of the study is that the sample size was larger than is usually for qualitative research. It is possible that the depth within sub-cohorts (region, age, and gender) may have been missed. Further, this study did not strongly represent the opinions of individuals in rural communities in northern Ontario.

Conclusion

Our study identified unique barriers to implementation of physical activity and exercise recommendations for individuals with osteoporosis: fear of injury, access to exercise programs that meet needs and preferences, disease-related symptoms hindering exercise and physical activity preferences, gender-related norms and physical activity preferences, and social and personal incentives to exercise. Mapping these barriers onto the BCW provided suggested training, education, modeling, and persuasion as intervention functions. Interventions and programs may need to consider preferences or determinants of health inequities, or their intersectionality through the modeling intervention function, and develop a multi-media approach to address the training and education intervention functions. Future interventions can be delivered through communication/marketing, guidelines, and service provision.