Implementation of a Personalized, Cost-Effective Physical Therapy Approach (Coach2Move) for Older Adults: Barriers and Facilitators.

BACKGROUND AND
PURPOSE: This article reports on a recent randomized clinical trial that showed a personalized approach to physical therapy (Coach2Move) by a physical therapist specialized in geriatrics (PTG) to be more cost-effective than usual physical therapy care in people with mobility problems (n = 130, mean age = 78 years).
METHODS: We used an explanatory mixed-methods sequential design alongside the randomized clinical trial to gain insight into (a) the contrast between the 2 interventions, (b) the fidelity of the Coach2Move delivery; (c) PTGs' experiences of Coach2Move; and (d) possible barriers and facilitators for future implementation. The study included 13 PTGs educated in the strategy and 13 physical therapists with expertise in geriatrics delivering the usual care. In total, 106 medical records were available for assessment: 57 (85%) Coach2Move, 49 (75%) usual care. Quantitative process indicators were used to analyze electronic medical records to determine contrasts in the phases of clinical reasoning. The fidelity of the delivery was tested using indicator scores focusing on 4 key elements of Coach2Move. In-depth interviews with Coach2Move therapists were thematically analyzed to explore experiences and facilitators/barriers related to implementation. RESULTS AND DISCUSSION: Indicator scores showed significant and clinically relevant contrasts in all phases of clinical reasoning, with consistently higher scores among PTGs, except for the treatment plan. Moreover, the fidelity of Coach2Move delivery was more than 70% in all phases, except the evaluation phase (53%). Experiences of Coach2Move were positive. In particular, extended intake allowing motivational interviewing, physical examination and an in-depth problem analysis, and shared goal setting were considered valuable. Facilitators for implementation were the addition of a Coach2Move medical record, frequent coaching by the researcher, and readiness to change in the therapist. Barriers were (1) having to use 2 parallel electronic medical record systems, (2) having to clear the calendar to schedule an intake of 90 minutes, (3) fear of losing income, (4) the sense that patients do not want to change their lifestyle, and (5) not acknowledging that increasing physical activity is an important goal for older adults with mobility problems.
CONCLUSIONS: Physical therapy based on the Coach2Move strategy is substantially different from usual care. Future implementation should focus on increasing regular evaluation and feedback, taking into account individuals' contextual factors, and improving organizational facilities while mitigating income loss.

RCT Entities:   Population Interventions Outcomes

INTRODUCTION

Stimulating self-management and increasing physical activity are considered essential components in the therapeutic management of frail older adults; however, this is very challenging.1 A successful new strategy is the systematically developed Coach2Move approach, provided by physical therapists (PTs) specialized in geriatrics (PTGs).2 In a recent randomized clinical trial (RCT), we demonstrated that Coach2Move physical therapy was indeed more effective than usual care physical therapy in increasing physical activity, improving health status, and reducing frailty.3 Moreover, the health care costs were lower in the Coach2Move group. These results warrant further implementation.

In short, the Coach2Move approach is based on the Hypothesis Oriented Algorithm for Clinical Reasoning II (HOAC-II)4 and places a strong emphasis on individual tailoring and setting priorities (Table 1).4,5 Moreover, the primary goal of the Coach2Move strategy is to increase physical activity in daily life as much as possible, increasing the long-term effectiveness related to health status and reducing frailty levels. Because we expected that specific in-depth knowledge in geriatrics would be beneficial in clinical reasoning, Coach2Move was provided by PTGs.2

Table 1.

Description of the Coach2Move Strategy and Usual Care Physical Therapy, and the Implementation Strategy for Older Adults (≥70 Years) Visiting the Physical Therapist With Mobility Problems

Subject	Coach2Move Strategy	Usual Care Physical Therapy
Provider	Physical therapist with (a) additional education in physical therapy in geriatrics at master's level, (b) experienced in working with older adults, and (c) trained in the Coach2Move strategy and measurement instruments and standardized reporting in medical records.	Physical therapist with (a) experience in working with older adults, and (b) trained in using the measurement instruments and standardized reporting of medical records.
Primary focus	(Re)gaining roles and physical activity focused on empowerment of abilities, self-management, and using the context in the exercises focused on barriers related to impairment, activity, and participation levels.	Exercises focused on impairment, activity, and participation levels.
Therapy elements	Key elements added to the steps in clinical reasoning (90 min): History taking concerning the start and course of the complaints, exploring the question of help, with added motivational interviewing techniques addressing barriers and facilitators (physical, social, and environmental) in relation to physical activity, and exploring needs and beliefs; questions focus on both disabilities and abilities. Using the HOAC-II for problem analysis and diagnostics, eg, testing muscle force, joint mobility, endurance, pain, fatigue, skill performance, coordination, etc, guided by the disabilities found, with a strong focus on task manipulation to test abilities and to set priorities. Shared decision making on meaningful treatment goals to abrogate barriers and increase physical activity. Coaching on self-management and self-efficacy to perform exercise in daily activities at home and goal setting on being physically active at home. Repeated measurements to monitor the outcome and increase adherence and motivation. Focus on meaningful activities at home with help from family, friends, and/or professionals. Three tailored intervention profiles defined by the number (4/9/18 sessions) of intervention sessions needed, based on expected recovery potential, baseline level, coping style, and availability of environmental support. Increasing physical and social activity levels and exercise interventions based on the physical therapy diagnosis, including training in strength, endurance, balance, flexibility, functional training, etc.	Steps in usual care physical therapy clinical reasoning (30 min): History taking concerning the start and course of the complaints, exploring the issue of help, with a strong emphasis on disabilities. Using the HOAC-II for problem analysis and diagnostics, eg, testing muscle force, joint mobility, endurance, pain, fatigue, skill performance, coordination, etc, guided by the disabilities found. Defining treatment goals. Supervised training combined with exercise advice at home. Monitoring through observation (less objective measurements). Less focus on environmental social support. Open-ended: number of sessions to be determined during ongoing treatment. Exercise interventions depending on the physical therapy diagnosis, including training of strength, endurance, balance, flexibility, functional training, etc, and increasing physical activity level.
Implementation	Two-day training in Coach2Move strategy (motivational interviewing, HOAC-II with video cases, strong emphasis on task manipulation [what to change for enablement], shared decision making and SMARTI goal setting, and how to organize self-management). Explanation of and training in the measurement instruments used. Explanation of the use of Coach2Move supportive medical records with decision aids. Feedback and coaching during the RCT by the researcher (NdV). Three follow-up meetings to discuss problems encountered. Payment for the extra time to perform the intake.	One day training on the HOAC-II (same patient video cases). Explanation of and training in the measurement instruments used. Explanation of medical recording guidelines.

Abbreviations: HOAC-II, Hypothesis Oriented Algorithm for Clinicians II; RCT, randomized clinical trial.

One of the most persistent prejudices concerning the Coach2Move strategy is that PTs argue that they already apply a personalized approach and take personal and environmental factors into account, suggesting that Coach2Move and usual care physical therapy are, more or less, similar. For future implementation, we need to specify and show the difference between the 2 approaches in clinical practice and gain insight into the main barriers and facilitators for large-scale implementation. Therefore, we performed a preplanned process evaluation alongside the RCT aimed at answering the following questions:

Contrast: Is Coach2Move physical therapy different from usual care physical therapy?

Fidelity: Were PTGs loyal to the Coach2Move approach?

Experiences: What were PTGs' experiences of the Coach2Move approach?

Implementation: What were the facilitators/barriers related to the implementation of Coach2Move?

METHODS

To answer the questions outlined previously, we used an explanatory mixed-methods sequential design alongside the RCT.3 The quantitative findings were further explored using qualitative analyses. Quantitative data were collected from electronic medical records (EMRs) and qualitative data were obtained through individual interviews with the PTGs delivering Coach2Move.

Quantitative Data

The regular PTs who participated in the trial had experience with working with older adults and were trained in using the measurement instruments and reporting according to the standardized EMRs but did not undertake additional education. They used their own software system and were asked to send an anonymized copy of each episode to the researcher. The Coach2Move therapists used an online EMR including the same steps as the regular software systems (history taking, diagnostic tests, treatment plan, treatment delivered, and evaluation) but extended with some items to facilitate the Coach2Move strategy (eg, questions about abilities, environmental factors, and room for explicit hypotheses and goal-setting). Electronic medical records that were incomplete (<10% filled) or concerning cases in which more than 1 therapist was involved in treatment were excluded from the analyses.

To measure the quality of the physiotherapeutic process, a set of process indicators was developed on the basis of the guidelines for medical record reporting of the Royal Dutch Society for Physical Therapy6 and the HOAC-II.4,5,7 The first set of items was developed by the research team and was discussed in 3 Delphi rounds (April-June 2013) with an expert panel consisting of members of the research group (L.B., N.V., B.S., and R.N.), experts in the field of physical therapy (n = 3), geriatrics (n = 2), and indicator development (n = 2).7–9 The first Delphi round focused on the relevance and definitions of the items. In the second round, the items were linked to indicators for each step of the HOAC-II model.7–9 In the last Delphi round, the members of the panel reached final consensus, resulting in 31 items clustered into 6 overall indicators related to the therapeutic phases: “history taking,” “diagnostics,” “analysis,” “treatment plan,” “treatment delivery,” and “evaluation” (Appendix 1). To judge specific Coach2Move elements, we selected items that were combined into 4 specific indicator scores for the key elements of Coach2Move: “patient centeredness” (items 6.1-9.1), “focus on self-management, empowerment, and patient adherence to treatment goals” (items 20.1-20.2), “SMART and Inspiring (SMARTI) goal-setting” (items 18.1, 18.2, and 21), and “regular evaluation and feedback” (item 24).

Each EMR was independently scored by 2 researchers not blinded for the intervention (A.S., N.V., L.B.). Blinding was not feasible because of the difference in reporting between groups, as described previously. Differences in indicator scores between the 2 raters were discussed until consensus was reached. Two raters scored the EMRs pair by pair with adequate reliability (L.B. and N.V., intraclass correlation coefficient [ICC] = 0.85; A.S. and N.V., ICC = 0.89).

Qualitative Data

Between March and May 2014 (after completion of the RCT), semistructured interviews (45 minutes) were held with the Coach2Move therapists (PTGs) at their own physical therapy practice by 1 researcher (A.S.). Three PTGs did not really start the intervention and treated fewer than 3 individuals because of pregnancy, changing job, or illness; these PTGs were excluded to guarantee sufficient experience with and knowledge of Coach2Move in daily practice. Having supplied written information on the aim of the study, PTGs were invited by the researcher by telephone to participate. Informed consent was signed prior to the interview. An interview protocol was developed by the research group (Appendix 2). Each interview was audiotaped and transcribed verbatim (A.S.). Member checking was used to review the interview transcript for errors and misinterpretations.10 The results were analyzed (A.S.) and discussed (N.V.), and new topics brought up by the participants were included in the following interviews.11,12

Data Analysis

Descriptive statistics are used to describe the demographic characteristics of the therapists and to compute the indicator scores. The following steps were taken to answer the 4 questions:

Contrast: To determine whether Coach2Move physical therapy differs from usual care physical therapy, we compared the 6 overall indicator scores and the 4 Coach2Move-specific indicator scores of the 2 groups using a Mann-Whitney U test. A P value of less than .05 was considered statistically significant. Moreover, the perceived differences between Coach2Move physical therapy and usual care physical therapy were elicited in the semistructured interviews with the PTGs.

Fidelity: To determine the loyalty of the PTGs to the Coach2Move strategy, we made use of both quantitative and qualitative data. Because we trained the PTGs to perform their problem analysis of the clinical scenario conforming to the HOAC-II steps, we expected adequate scores on the 6 indicator scores related to the phases in clinical reasoning. We, therefore, deemed a score of greater than 70% on all indicator variables as adequate in terms of Coach2Move fidelity. Consequently, we compared the interview results for the PTGs who scored more than 70% for overall fidelity with those scoring lower to determine underlying thoughts and beliefs concerning the strategy.

Experiences and implementation: To gain insight into the participants' experiences of Coach2Move and the facilitators and/or barriers related to its implementation, the interview data were analyzed using theoretical thematic analysis in 5 steps: (1) familiarization with the data; (2) generating initial codes; (3) searching for themes; (4) defining and naming themes; and (5) producing a report.13 The first 2 transcripts were analyzed by both researchers (A.S., N.V.) separately; thereafter, the codes that emerged from the data were compared and discussed. The subsequent interviews were analyzed by the first author (A.S.) and checked by the second author (N.V.).

All statistical analyzes were performed using SPSS 20 for Windows. The interviews were analyzed using QRS NVivo 10 for Windows.

RESULTS

Thirteen PTGs, male 7; mean age 46 (SD = 10) years, and 13 PTs, male 4; mean age 42 (SD = 13) years (Table 2), treated, respectively, 64 and 65 individuals. The average (SD) length of professional experience was 21 (11) and 17 (11) years for the PTGs and PTs, respectively. The PTGs treated on average 18 (SD = 10) individuals every week compared with 10 (SD = 5) in the usual care group. (See Table 2 for characteristics of the PTGs and PTs in the RCT.) The reasons for consulting physical therapy were diverse: low back pain, aerobic capacity problems based on (chronic) diseases, hip problems, knee problems, and several individuals with other problems.3

Table 2.

Characteristics of Physical Therapists in the Randomized Clinical Trial

Characteristics	PTG n = 13	PT n = 13
Age, mean (SD), y	46 (10)	42 (13)
Sex, female: n (%)	6 (46)	9 (69)
Experience as physiotherapist: mean (SD), y	21 (11)	17 (11)
Experience as geriatric physiotherapist: mean (SD), y	4 (4)	N/A
Average number of geriatric patients per week: mean (SD)	18 (10)	10 (5)

Abbreviations: N/A, not applicable; PT, physical therapists with experience in geriatrics; PTG, physical therapists specialized in geriatrics.

For the quantitative part of the study, we were able to include 57 (89%) EMRs in the Coach2Move group and 49 (75%) EMRs in the usual care group. Seven EMRs in the Coach2Move group were excluded because of treatment by multiple therapists (n = 3) or more than 90% missing data (n = 4). In the usual care group, 16 files were unavailable. These 16 files were not released by the participating PT after sending an e-mail 3 times and telephoning twice (the reason for withholding the files was not provided), so we decided to accept the missing values. The demographic and clinical characteristics of the participating individuals were comparable for both groups.3 For the qualitative part of the study, in the Coach2Move group, 7 of the 10 available PTGs agreed to participate in the interviews. The reasons for not participating were personal problems, lack of availability, and lack of time. After member checking, 1 PTG added a clarification to an interview transcript.

Contrast: Is Coach2Move Physical Therapy Different From Usual Care Physical Therapy?

The mean indicator scores (Table 3) were consistently higher in the Coach2Move group, except for the indicator “treatment plans.” The mean indicator scores for “Analysis” (70.9 vs 16.1) and “Treatment delivery” (85.3 vs 39.3), and “Evaluation and measurements” (52.6 vs 17.6) showed the largest between group contrast. The scores for “Evaluation” were low in both groups. In the usual care group, the “Analysis” showed the lowest mean score. Furthermore, there was a statistically significant difference between the 2 groups on the Coach2Move-specific subset of indicators (Table 3).

Table 3.

Scores on the Process Indicators From the Coach2Move Group and Usual Care Group

Subject	Coach2Move PTG Group	Usual Care PT Group	P
Number of therapists	13	9
Number of medical records included	57	47
Adherence to the 6 physiotherapeutic phases: mean, (SD), %
History taking	85.2 (12.8)	47.4 (12.2)	<.01
Diagnostics	83.3 (16.8)	56.1 (19.8)	<.01
Analysis	70.9 (23.8)	16.1 (20.4)	<.01
Treatment plan	70.5 (26.7)	71.0 (30.0)	.91
Treatment given	85.3 (23.1)	39.3 (24.1)	<.01
Evaluation measurements	52.6 (36.7)	17.6 (18.9)	<.01
Adherence to the 4 Coach2Move-specific aspects: mean, (SD), %
Patient centeredness	79.3 (18.8)	17.0 (13.0)	<.01
Focus on self-management, empowerment, and patient adherence to treatment goals	84.2(29.4)	21.9 (27.8)	<.01
SMARTI goal setting	73.1 (22.7)	56.1 (25.8)	<.01
Regular evaluation and feedback	40.4 (49.5)	2.0 (14.3)	<.01

Abbreviations: PT, physical therapist; PTG, physical therapist specialized in geriatrics; SMARTI, specific, measurable, acceptable, realistic, timely, inspiring.

In our qualitative analysis, we focused on the difference experienced between the Coach2Move strategy and usual care. All PTGs mentioned that the exercises in the Coach2Move strategy did not differ much from their usual physical therapy practice. However, 5 PTGs stated that the primary focus on patient involvement in goal setting and self-management in the Coach2Move strategy did really differ from their usual care strategy, leading to greater commitment on the part of the patients, greater adherence to the intervention goals, and sustained activity. Another main difference was a shift from a disability-focused approach to an ability-focused approach and involving the environment in both history taking and intervention.

We tend to take the patients by the hand and tell them, here we are now and this is where we want to go. Whereas now we ask where do you want to go? And how do you think you can reach this? What do you need? What can I do for you and what can you do yourself to get where you want to go? (PTG3)

What's new is the focus on disablement and enablement. To look at the opportunities and to appoint this, I use my own EMR, the International Classification of Functioning (ICF) model, as well, so I state the personal and environmental factors. But not as comprehensively as in the Coach2Move strategy. (PTG 4)

Fidelity: Were PTGs Loyal to the Coach2Move Approach?

Quantitative analysis demonstrated high fidelity scores of the PTGs in all 6 physiotherapeutic phases, except for the evaluation phase. Regarding the overall fidelity score, 10 of the 13 PTGs scored more than 70% on the total indicator score.

We subsequently compared the interview results of the PTGs with an adequate fidelity score with those with an inadequate score. The PTGs with an adequate fidelity score agreed with the fundamental perspective of the Coach2Move approach that the primary focus of a physical therapeutic intervention for older adults should be on increasing an individual's level of physical functioning and the ability to self-manage; also, they accepted the premise that a focus on increasing physical activity will consequently positively affect functional ability as well as social participation. They did not change their primary focus but did acquire new and additional tools to realize the primary goals.

...nevertheless, we always have an active policy. I think that in this practice there is not a huge difference. The only difference is that we focus still more on motivating patients to be active in their leisure time, and to be active outside therapy time and to continue this after therapy. I think this is an essential difference. Previously, I thought that if patients stopped therapy, their functioning slowly decreased. Nowadays, functioning is better maintained, they stay active. (PTG 10)

The PTGs with an inadequate fidelity score reasoned the other way around. They stated that PTs first need to improve an individual's functional ability preparatory to stimulating that person to become more physically active, and they rejected the notion that physical activity is a necessity to stay healthy. They argued that most individuals need help for pain or impairments. Their primary focus is that (hands-on) treatment of body functioning and structures is the core of their profession.

What I also think of Coach2Move is that nowadays everybody thinks that people are not active enough. Then I think, “yeah right,” there are more problems besides being inactive. It annoys me a bit. I think there are enough people who participated in the RCT who were usually active and then this strategy does not have any added value because the main problem wasn't being inactive. (PTG9)

Experiences: What Were PTGs' Experiences of the Coach2Move Approach?

In the interviews (n = 7), 3 main themes were brought up by the PTGs: (1) overall positive experience; (2) extended intake; and (3) goal setting. All PTGs believed that they were able to apply the Coach2Move strategy in their daily physical therapy practice. One PTG (PTG9) made a critical note that inactivity is not always the main problem and, therefore, Coach2Move is not always the most appropriate strategy. The other PTGs stated that they supported the Coach2Move strategy and recognized the importance of improving physical activity and participation by means of a coaching intervention.

I think it is an excellent strategy, it suits my working method. In my opinion, it is a good tool to let patients take control. (PTG10)

Concerning the advice in Coach2Move strategy to use an extended intake of 1.5 hours, 5 PTGs reported that they had already extended their first consultation to 1 hour prior to the RCT. Two other PTGs (PTG1 and PTG7) mentioned that they used multiple diagnostic sessions because patients were not able to focus for 1.5 hours. The PTGs noticed that history taking, in particular, differed from their regular history taking, with a more extended focus on the contrast between disabilities and abilities related to the roles of the individual.

Shared goal setting was recognized as a key element of the Coach2Move strategy. Four PTGs noted that patients are not used to stating explicitly what they really want to achieve. This requires specific skills on the part of the PTG, using probing questions and listening carefully to determine the underlying goals. Two PTGs (PTG4 and PTG7) indicated that the main difference in goal setting in Coach2Move was that goals should be “inspiring,” thus motivating patients to adhere to the treatment plan and to invest time and energy in reaching these inspiring goals.

I am keen to coach inspiring goals. So SMART becomes SMARTI and these are time-limited inspiring goals, this is what attracts me. I find this very important and this is what I have learned from the Coach2Move strategy. (PTG7)

Implementation: What Were the Facilitators/Barriers Related to the Implementation of Coach2Move?

Qualitative analysis showed that barriers and facilitators were apparent not only at the therapist level but also at the organizational and patient levels. At the organizational level, all PTGs considered the use of 2 EMRs a major burden and drawback. As the Coach2Move strategy was not implemented using the regular EMRs, the PTGs used 2 electronic records for every individual: 1 for their regular registration and 1 for the trial registration. However, PTGs were positive about the items added in the EMR as guidance for working with the HOAC-II algorithm. Also, extended intake was sometimes difficult to schedule. In addition, a point raised by PTG1 concerned the consequences of fewer treatment sessions in the Coach2Move intervention, resulting in fewer sessions and, therefore, less payment. This might lead to less income for Coach2Move therapists.

But there is a bigger problem within physical therapy. If you treat every patient using Coach2Move, then you treat patients in fewer treatment sessions. This will be a financial disadvantage and this makes it hard to implement in the future. (PTG1)

During the RCT, PTGs were coached in how to use the Coach2Move strategy by the primary researcher of the RCT (N.V.). This coaching was perceived as an additional organizational facilitator. Five of the 7 PTGs appreciated this kind of feedback and found it useful for improving their skills.

At the patient level, PTGs mentioned that the individual's readiness to change was an important success factor for the Coach2Move strategy. All PTGs reported that they tried to activate and motivate patients to become more physically active by stimulating them to perform relevant home activities and exercise. All PTGs mentioned that this was easier if individuals were interested in increasing their activity levels prior to the treatment.

If someone is not yet willing to change? Will it work then? Well ... this does not work all the time, but then I usually let them make a matrix. What does this behavior bring you? And what would other behavior bring you? ... What do you want to achieve? ... If you could walk for 30 minutes and you were able to do the groceries in the shopping center ... or if you could go out with your daughter again—what is the added value for you? And then you see that people start thinking about it. And this starting to think about it is important for me, because then they are already changing. (PTG3)

In addition, the individual's ability to apply self-management was an important factor for the success of the Coach2Move intervention. Both the cognition of the individual and the attitude of the informal caregiver were considered to influence self-management ability. If the opinion of the informal caregiver was that physical activity is not important, this negatively influenced the treatment results. If individuals did not show any ability to self-manage their problems, the PTGs tended to focus more on problems concerning function (eg, balance, muscle strength), rather than on activities and participation.

Barriers and facilitators at the therapist level are related to the skills of the PTGs. The skill of motivational interviewing was considered very helpful by 3 PTGs. The main difference from the other PTGs is that these 3 PTGs let the individual explain why he or she should be more active. These PTGs also mentioned that they had already had training in motivational interviewing before they started the Coach2Move strategy. Furthermore, the skills of listening carefully and trying to determine the real underlying problems and goals were suggested as essential for a therapist working according to the Coach2Move strategy.

A coach should especially search for the key to intrinsic motivation. That is important, to let the patients themselves reach the conclusion. (PTG4)

Lots of people are able to learn this strategy, but they just do not have the discipline to do it by themselves. They are not able to motivate themselves because they always have excuses for why it is not possible. And those, I think ... in my opinion, those excuses are not relevant, but for them the excuses are relevant. But then you will not get through to them, to let them experience that if you are more active for a longer time it will actually lead to feeling well. (PTG2)

DISCUSSION

The Coach2Move strategy differed from usual care physical therapy in almost all physiotherapeutic phases, but the main contrast was defined by a more individual-centered problem analysis and the use of inspiring goals based on shared decision making, leading to greater adherence and self-management. The content of preplanned treatment plans did not seem to differ between groups; however, the delivery was more stringent in the Coach2Move group. Moreover, the treatment plans were based on a transparent analysis in the Coach2Move group, recognizable by the mean indicator score of 70%, while the mean indicator score for this phase was the lowest in the usual care group (16%). “Regular evaluation and feedback” was low in both groups but was still twice as high in the Coach2Move group. The fidelity of most PTGs to the Coach2Move strategy was adequate, although beliefs and attitudes influenced adherence to the protocol. The semistructured interviews showed that the PTGs had, overall, a positive experience with the Coach2Move strategy, although the experience was influenced by their views of physical therapy in general. Facilitators of the strategy were the intrinsic motivation of therapists to focus on enablement and on coaching to improve activity levels. At the individual (and informal caregiver) level, intrinsic motivation to become (or stay) active and experiences of success were identified as facilitators, while lack of motivation in combination with cognitive impairments and inadequate support in the social environment were mentioned as barriers. Reimbursement for the longer intake time and the content of the EMR were facilitators, while the lack of connection with their own medical record was a barrier. Moreover, the lower number of intervention sessions based on the intervention profiles was mentioned as a barrier because the income would be reduced.

The fidelity scores within the Coach2Move group were good and quite similar to the fidelity scores for other physical therapy guidelines.14–16 An intensive and multifaceted implementation strategy was applied in the RCT because high fidelity was considered a prerequisite to show the extra effect above the also effective usual care physical therapy.3,17 The PTGs stated that coaching, in particular, was valuable in implementing the new strategy in terms of diagnostics and intervention delivery. However, coaching is quite an intensive and expensive strategy and may not be feasible for large-scale implementation. As an alternative, peer assessment in addition to education and training may be successful as this has been shown to improve adherence to guidelines.18–21

All PTGs had a positive experience overall with the Coach2Move strategy, although there were different views concerning the role of the PT in general, which influenced fidelity scores. The therapist's personal view of health care is an influencing factor, and a favorable perspective is possibly a prerequisite for high adherence to individual-centered approaches. A recent international study also recognized that individual centeredness in guidelines should be increased, resulting in 14 recommendations, including intervention-centered goals, use of the ICF terminology in problem analysis, adopting a shared decision-making method, and incorporating individual-reported health outcome measures22; all these elements are combined in the Coach2Move strategy. The contrastive analysis between Coach2Move and usual care illustrates that the PT samples information on all different ICF factors but needs guidance to combine them in clinical practice. Also, the PT needs to shift from a biomedical to a biopsychosocial perspective on health, defining health as the ability to self-manage and adapt, with less focus on illness and interventions, and taking into account strengths rather than weaknesses.23 Although appreciated, this concept requires substantial personal input and competencies, as also demonstrated in a recent study: patients perceive health as multidimensional, but physicians (including PTs) assess health from a more narrow and biomedical standpoint.24 This study shows the added value of a more personalized strategy.25 Dynamic, multidimensional knowledge is the basis of PTGs' education and also the Coach2Move strategy. The systematic algorithm in HOAC-II guides the PTGs in focusing on a tailored intervention, with SMARTI defining goals that can be embedded in daily life and empowered by the social environment. In an earlier pilot study on Coach2Move, we interviewed patients to gather insights into their experiences of the strategy. Individuals treated according to the Coach2Move strategy especially valued the fact that they knew what to do to improve their physical performance, and that they received regular evaluation and feedback.26 Adequate shared decision making and explicit knowledge transfer regarding the treatment plan and therapy progress will most likely result in a higher adherence on the part of the individual with the preplanned treatment goals and exercises, and will increase self-esteem.

Barriers and facilitators related to the successful application Coach2Move were similar to those identified in other studies: organization and the characteristics of patients and therapists.16,27,28 Although they were paid for extra time, a number of therapists were unable to schedule enough time. However, the most important organizational barrier was the use of 2 EMRs in the study environment. Thus, the positive effect of the revised EMR will be much greater when this barrier is resolved. Moreover, we call for the opportunity to employ open text codes. The PTGs used rich information, while in the usual care group information was reduced to quite well-formulated treatment plans in standardized text fragments already provided by the EMR they used. Thus, these plans seemed to be SMART, but patients were not involved in choosing personalized goals or in determining how and when the goals would be reached. The effect was that the indicator scores for the content of the treatment plans were comparable, but the plans were poorly followed up and goals were not reached, possibly due to a lack of adherence by the therapists themselves and the patients lacking feedback.

Another important barrier to further implementation was the fear of lower income on the part of the PTGs. We implemented 3 patient profiles with 4 or less sessions, 5 to 9 sessions, or 9 to 18 sessions to tailor the intervention in self-management and to increase the awareness among patients that the treatment period would end at a certain point. The trial results showed that the results were indeed higher in the Coach2Move group, while the number of sessions needed was lower. In future implementation, it needs to be stipulated that PTGs are at least partly to be reimbursed for the lower income generated.

Some PTGs mentioned a lack of patient motivation and low cognition and support of caregivers as barriers. Some PTGs found it difficult to adhere to Coach2Move when their patients lacked motivation to become more active, or when they were not supported in the process by their caregivers. Moreover, PTGs struggled if an individual did not show the ability to self-manage his or her problems. In such cases, the PTGs tended to shift the focus back to problems in physical functioning (ie, balance, muscle strength), rather than seeking motivation and shared goals in activities and participation. This behavior is possibly related to the competencies of the therapist: in another study, Jensen et al25 noticed that experts mentioned not meeting patients who were not motivated or had no goal. Thus, in the choice of implementation strategies, the negative interaction between the competence of the therapist and the motivational focus of patients should be taken into account.25 Readiness to change is an important condition for changing behavior, such as adhering to physical activity.29 The therapist needs to adapt the approach to the stage of change of the individual, as pointed out in the motivational interviewing training.30 Nonmotivated individuals need stimulation to consider the consequences of their choices, rather than starting by trying to convince them what to do.29 Advanced techniques in motivational interviewing can be used to provide the correct decision-making climate.30 Here, we found that the PTGs with prior training in motivational interviewing found it easier to motivate the “unmotivated” patients.

The major strength of this study is the mixed-methods approach, enabling exploration of the complete spectrum of the process; a limitation is the inability to blind the raters for the intervention. On the contrary, the reliability between the 2 raters was high (ICC = 0.87). The qualitative data were based on an interview guide and member checking was used after the interviews. Although there is always the possibility that responses in interviews will be socially desirable, the member check and the fact that the participating therapists were highly motivated do not make this likely. Assessing fidelity through scoring performance indicators in an EMR could be seen as a limitation. Theoretically, there can be a gap between reporting and real practice. The study by Richoz et al31 showed that a large majority of Swiss PTs were not accurate in their clinical recording. However, the outcome of the RCT confirms a contrast between the 2 approaches. The fact that not all PTGs were interviewed (for various reasons) might have led to bias and an overestimation of the motivation to use Coach2Move. Moreover, recall bias might have been present as the PTGs were first introduced to the Coach2Move strategy more than 2 years ago and the last treatment of individuals was several months ago. However, most PTGs (5 out of 7) indicated that they were still working according to the Coach2Move strategy on a daily basis.

In conclusion, we have demonstrated that physical therapy according to the Coach2Move strategy is substantially different from usual care physical therapy in nearly all physical therapeutic phases of the HOAC-II and in terms of individual centeredness, focus on self-management, individual empowerment, and adherence to treatment goals and SMARTI goal setting. The majority of PTGs showed high fidelity with the Coach2Move strategy, although there was room for improvement in terms of regular evaluation and feedback. In combination with the fact that PTGs appreciated the Coach2Move approach, the multifaceted implementation strategy thus appeared successful. To enable further optimization of the intervention results, future implementation needs an implementation strategy that takes into account professional competencies, organizational facilities (eg, related to intake time and EMR), the potential financial consequences in terms of income, and the individual patient contextual factors.

APPENDIX 1

Process Indicators (Translated From Dutch)