http://www.diva-portal.org
This is the published version of a paper published in Disability and Rehabilitation.
Citation for the original published paper (version of record):
Fritz, J., Wallin, L., Söderlund, A., Almqvist, L., Sandborgh, M. (2019)
Implementation of a behavioral medicine approach in physiotherapy: impact and
sustainability
Disability and Rehabilitation
https://doi.org/10.1080/09638288.2019.1596170
Access to the published version may require subscription.
N.B. When citing this work, cite the original published paper.
Permanent link to this version:
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=idre20
Disability and Rehabilitation
ISSN: 0963-8288 (Print) 1464-5165 (Online) Journal homepage: https://www.tandfonline.com/loi/idre20
Implementation of a behavioral medicine
approach in physiotherapy: impact and
sustainability
Johanna Fritz, Lars Wallin, Anne Söderlund, Lena Almqvist & Maria
Sandborgh
To cite this article:
Johanna Fritz, Lars Wallin, Anne Söderlund, Lena Almqvist & Maria Sandborgh
(2019): Implementation of a behavioral medicine approach in physiotherapy: impact and
sustainability, Disability and Rehabilitation, DOI: 10.1080/09638288.2019.1596170
To link to this article: https://doi.org/10.1080/09638288.2019.1596170
© 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
View supplementary material
Published online: 18 Apr 2019.
Submit your article to this journal
Article views: 327
RESEARCH PAPER
Implementation of a behavioral medicine approach in physiotherapy: impact and
sustainability
Johanna Fritz
a, Lars Wallin
b,c,d, Anne S€oderlund
a, Lena Almqvist
aand Maria Sandborgh
aa
School of Health, Care and Social Welfare, M€alardalen University, V€asterås, Sweden;bSchool of Education, Health and Social Studies, Dalarna University, Falun, Sweden;cDivision of Nursing, Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden;dDepartment of Health and Care Sciences, The Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden
ABSTRACT
Purpose: To explore the effects on and sustainability of physiotherapists’ clinical behavior when using facilitation to support the implementation of a behavioral medicine approach in primary health care for patients with persistent musculoskeletal pain.
Methods: A quasi-experimental pre-/post-test trial was conducted. Fifteen physiotherapists were included in the experimental group, and nine in the control group. Based on social cognitive theory and the Promoting Action on Research Implementation in Health Services framework, facilitation with multifaceted implementation methods was used during a six-month period. Clinical behaviors were investigated with a study-specific questionnaire, structured observations, self-reports and patient records. Descriptive and non-parametric statistical methods were used for analyzing differences over time and effect size.
Results: A sustained increase in self-efficacy for applying the behavioral medicine approach was found. Clinical actions and verbal expressions changed significantly, and the effect size was large; however, changes were not sustained at follow-ups. The behavioral changes were mainly related to the goal set-ting, self-monitoring and functional behavioral analysis components. No changes in clinical behavior were found in the control group.
Conclusion: Tailored multifaceted facilitation can support the implementation of a behavioral medicine approach in physiotherapy in primary health care, but more comprehensive actions targeting sustainabil-ity are needed.
äIMPLICATIONS FOR REHABILITATION
Tailored multifaceted facilitation can support the implementation of an evidence based behavioral medicine approach in physiotherapy.
Facilitation can be useful for increasing self-efficacy beliefs for using behavioral medicine approach in physiotherapist’s clinical practice.
Further research is required to establish strategies that are effective in sustaining behavioral changes.
ARTICLE HISTORY Received 1 November 2018 Revised 13 March 2019 Accepted 13 March 2019 KEYWORDS Physiotherapy; clinical competence; evidence-based practice; musculoskeletal pain; primary health care; self-efficacy
Introduction
A challenge in health care, including physiotherapy, is the imple-mentation of new evidence-based methods and guidelines [1, 2]. One in seven of all consultations in primary health care concerns musculoskeletal pain [3] and most of these patients seek or are referred to physiotherapy. Persistent pain markedly increases the individual’s disability in daily life activities, and one in four per-sons with persistent musculoskeletal pain believes that pain has had an impact on their employment status [4]. For patients with persistent musculoskeletal pain, a behavioral medicine approach in physiotherapy has shown positive effects on disability [5–7]. However, poor adoption of clinically relevant behaviors has been reported when implementing a behavioral medicine approach in physiotherapy [8–10]. Some changes in knowledge, beliefs and attitudes have occurred after participating in behavioral medicine courses, but changes in clinical actions have been less
forthcoming [4,11,12]. Insufficient skills training [8,10], complex-ity of the clinical intervention [12], physiotherapists’ perceptions of the intervention as ineffective [4], and inadequate duration of the educational input [4, 8] have been reported as reasons for poor adoption. Therefore, continuing to study how to support the implementation of a behavioral medicine approach in physiother-apy is necessary.
There is weak evidence supporting the effects of the most commonly used implementation methods in health care, such as printed educational materials, educational meetings, reminders, audits and feedback, educational outreach, the use of local opin-ion leaders and tailored interventopin-ions, on clinical behaviors [13,
14]. Similar effect sizes, between 4 and 12%, have been reported across these implementation methods [13]. In physiotherapy, multifaceted implementation interventions, i.e., multi-method interventions, have been tested but offer no clear evidence regarding the effects [15].
CONTACTJohanna Fritz johanna.fritz@mdh.se School of Health, Care and Social Welfare, M€alardalen University, Box 883, V€asterås, SE-721 23, Sweden. Supplemental data for this article can be accessedhere.
ß 2019 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives License (http://creativecommons.org/licenses/by-nc-nd/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited, and is not altered, transformed, or built upon in any way.
DISABILITY AND REHABILITATION
The Promoting Action on Research Implementation in Health Services (PARIHS) framework proposes facilitation as a useful method to support the implementation of evidence-based guide-lines [16] and has also been confirmed in the primary health care context [17–19]. Facilitation is defined as both the role of a per-son who facilitates and the process of practice with certain char-acteristics [20,21]. The facilitation role differs by being formal or informal, whether the facilitator is recruited internally or exter-nally, and whether the facilitator is trained for the role [20]. The facilitation process is described as a social process using an inter-active experiential learning approach, providing problem-solving strategies and support in the context in need of improvement [22–24]. The main features of facilitation involve making change easier and supporting the development of new knowledge and skills by using multiple methods [20]. Although facilitation has been suggested as a key component of successful implementa-tion, it has seldom been examined as a method for supporting implementation in physiotherapy [25].
Capability beliefs and intentions have been identified as strong predictors of healthcare professionals’ clinical actions [26]. These assumptions are based on social cognitive theory, one of the most commonly used social learning theories in implementation research [26–29]. However, studies evaluating the impact of facili-tation on capability beliefs and intentions are sparse [17].
Implementation methods have mainly been studied in health care contexts other than physiotherapy and not in relation to a behavioral medicine approach. The implementation methods that would be useful to change physiotherapists’ clinical behavior remain unclear. In the current study, clinical behaviors refer to beliefs, observable clinical actions and verbal expressions [30]. Thus, the aim of this study was to explore the effects on and sus-tainability of physiotherapists’ clinical behavior when using
facilitation to support the implementation of a behavioral medi-cine approach in primary health care for patients with persistent musculoskeletal pain.
Methods
The TREND statement [31] was used for reporting this study.
Design
A quasi-experimental pre-/post-test trial was conducted [32].
Participants and settings
All physiotherapists working in primary health care in three county councils were asked to participate in the study. The physiotherapists were allocated to either an experimental or a control group based on the physiotherapists’ wishes and such that participants from the same clinic belonged to the same group. Fifteen physiotherapists working in seven primary health care units were included in the experimental group, and nine physiotherapists working in seven primary health care units were included in the control group. Dropouts and missing data occurred in both groups during the implementation intervention and follow-up periods (Figure 1). There were no significant dif-ferences between the groups regarding baseline characteristics (Table 1). Most of the participants had some kind of previous education regarding behavioral medicine, such as motivational interviewing or university courses in behavioral medicine for physiotherapists. Participation was voluntary, and all participants gave written informed consent after receiving oral and written
Figure 1. The participating physiotherapists, dropouts and missing data for each measurement point. 2 J. FRITZ ET AL.
information. Furthermore, the participants could withdraw with-out any consequences.
The physiotherapists were affiliated with three county councils of similar sizes located in the middle of Sweden (seeTable 1). In the experimental group, all physiotherapists at three of the clinics participated, while at the other four clinics, 25–80% of the physio-therapists participated. In the control group, 25–75% of the physi-otherapists at each clinic participated. The physiphysi-otherapists reported that the majority of their patients presented for physio-therapy because of persistent musculoskeletal pain (>4 weeks). All the clinics received financial reimbursement for the time spent in the project, corresponding to the physiotherapists’ wage costs.
Implementation intervention for the experimental group
The implementation intervention consisted of facilitation as a multi-faceted intervention [23,27,28,33]. The first author, acting as the facilitator, has comprehensive knowledge and skills of the behav-ioral medicine approach and the use of behavior change techni-ques, as well as extensive experience teaching students and physiotherapists using various learning methods. The facilitation methods used consisted of outreach visits, peer coaching, educa-tional materials, video recordings, individual goal setting, self-moni-toring, manager support and access to written patient information. For further description of the intervention, see Supplementary Material. The facilitation methods were tailored to each physiother-apist through setting individual goals of importance for learning, optional discussion subjects, problem solving related to individual obstacles, and use of optional educational materials.
The implementation intervention period lasted six months. In the experimental group, the time allocated for the implementa-tion activities consisted of ten two-hour outreach visits led by the facilitator spread over six months and in total, four days allocated for self-studies, i.e., use of the educational material and discussion with colleagues (see Supplementary Material). The intervention started with a joint introduction (3 h) and ended with a final meeting (3 h) at the university. The introduction consisted of lec-tures about evidence supporting the behavioral medicine approach, an introduction to the behavioral medicine approach and the learning methods used in the implementation interven-tion. The final meeting focused on sharing experiences related to problem-solving solutions and strategies for maintaining the behavioral medicine approach.
The implementation intervention for the control group
The implementation intervention for the control group was selected to simulate the way new guidelines are usually dissemi-nated, i.e., written information and lectures. In the control group, the intervention started with a joint introduction (2 h) at the uni-versity consisting of the same lectures the experimental group received about evidence supporting the behavioral medicine approach and an introduction to the approach. The physiothera-pists received the same book given to the experimental group and were encouraged to read it and use the information in their clinical work. No further support for implementation was given.
Data collection
As far as we know, there is no tool concerning beliefs for using a behavioral medicine approach. The physiotherapists’ beliefs were therefore investigated using a study-specific questionnaire com-prising 24 questions where importance of, self-efficacy for and readiness for using the core components of the behavioral medi-cine approach were rated by the responders on an 11-point Likert scale. The core components in the behavioral medicine approach consisted of the following: patient’s goal setting, assessment of target behavior, patient’s self-monitoring of actions and cognition related to the target behavior in everyday life situations, individ-ual functional behavioral analysis, basic and applied skills acquisi-tion, and support of maintenance [12, 34]. Data were collected pre- and post-implementation and at the 3-month follow-up. Test-retest reliability of the questionnaire was evaluated on 59 undergraduate students with a two-week interval and analyzed with intraclass correlation, two-way mixed model, and absolute agreement. The results demonstrated good-to-excellent stability of the questionnaire, ICC¼ 0.95 (95% CI 0.90–0.98).
The physiotherapists’ behavioral medicine related clinical actions and verbal expressions were investigated with observa-tion, self-reports and analyses of documentation in patient records. Data were collected before and after the implementation intervention period and at the 3-, 6- and 12-month follow-ups.
Observation
To identify as many different observable clinical actions and ver-bal expressions as possible and to facilitate the observations, two treatment sessions per physiotherapist were video recorded by two of the authors (JF, MS). The video recordings were reviewed using an observation protocol by three researchers who were blinded to group affiliation and time point of measurement and not actively involved in the implementation intervention. The completed observation protocols were reviewed by one of the authors (JF) to ensure consistency between the reviewers, and corrections were made if needed. The observation protocol was study specific and consisted of 58 observable clinical actions and verbal expressions, assessed as present/not present, in relation to the core components of the behavioral medicine approach [12,
34], see Supplementary Material for details. Intra-rater reliability for the observation protocol was evaluated on four observers and analyzed with Cohen’s Kappa. The results showed substantial agreement; the Kappa value for the seven components ranged between 0.6 and 0.9. The highest agreement between the first and second observation was found for Goal setting, Kappa¼ 0.9, and the lowest agreement was found for Assessment of target behavior, Kappa ¼ 0.6.
Table 1. Physiotherapists’ baseline characteristics. Characteristics
Experimental group (n¼ 15)
Control group (n¼ 9)
Sex, male (M)/female (F) M¼ 5, F ¼ 10 M¼ 3, F ¼ 6
Age (years), median (min-max) 37 (23–63) 39 (24–57) Years of work as a physiotherapist, median (min-max) 9 (1–31) 5 (0.5–30) Years of work in primary health care, median (min-max)
9 (1–30) 3 (0.5–16)
Number of physiotherapists with previous behavioral medicine education
13 7
County council affiliation, county council A, B and C
A¼ 5, B ¼ 10 A¼ 3, B ¼ 3, C ¼ 3
Self-reports
The physiotherapists self-reported the content of each of the two treatment sessions in a log-book consisting of the 58 behavioral medicine related clinical actions and verbal expressions, assessed as present/not present, identical to the observation protocol.
Documentation in patient records
The physiotherapists’ notes regarding the two treatment sessions were reviewed by one of the authors (JF), blinded to group affili-ation, using a checklist similar to the observation. The checklist consisted of 38 of the 58 behavioral medicine related clinical actions and verbal expressions used in the observation protocol and log-book, assessed as present/not present.
Data analyses
For each physiotherapist, the scores in the observation protocols of the two treatments were merged, and a score of present clin-ical actions and verbal expressions in total and per component was calculated. In those cases when only one treatment session was video recorded, all observed data from the physiotherapist were excluded from that measurement point and handled as missing data (seeFigure 1). Scores in log books and checklists for patient records were handled in the same manner. The results are presented using descriptive statistics. The effects of the imple-mentation intervention over time, i.e., the changes and sustain-ability in ratings of beliefs and the changes in number of exhibited clinical actions and verbal expressions in observations, self-reports and patient records pre- and post-implementation, and at follow-ups, were calculated using Friedman’s ANOVA [35].
Due to missing data and attrition, the 3- and 12-month values were not included in Friedman’s ANOVA. To identify when the changes occurred, post hoc calculations were performed using the Wilcoxon Matched-Pairs Signed Rank test for within-group analy-ses [35]. Pre-values were compared with post- and follow-up val-ues. To identify in which behavioral medicine components the changes occurred, the Wilcoxon Matched-Pairs Signed Rank test was used for within-group analyses comparing pre- and post-val-ues [35]. The effect sizes were based on the observed clinical action and verbal expression outcomes pre- and post-implemen-tation intervention. The effect size was calculated within the experimental group using Pearson’s correlation coefficient r, where r¼ 0.10 represents a small effect, r ¼ 0.30 represents a medium effect and r¼ 0.50 represents a large effect [36]. The sig-nificance level was set at p 0.05. IBM Statistical Package for the Social Sciences (SPSS), version 24, was used for all analyses.
Ethical approval
The study was approved by the Regional Ethical Review Board, Uppsala, Sweden, Dnr 2015/385.
Results
Effects on the physiotherapists’ beliefs
The self-reported self-efficacy in the experimental group increased significantly for most of the core components in the behavioral medicine approach after the implementation intervention and was sustained after three months (see Table 2). A significant increase in perceived readiness for using the behavioral medicine Table 2. Median values and interquartile ranges (IQRs) for perceived beliefs about importance of, self-efficacy for and
readiness for using the behavioral medicine components in the experimental group, pre-, post-implementation interven-tion and at the 3-month follow-up.
Post hoc Difference over time Pre (n¼ 15) Post (n¼ 15) 3 months (n¼ 11) (df¼ 2, N ¼ 11) median (IQR) median (IQR) median (IQR) Chi2 p Importance†
patients’ goal setting 6 (5–8) 8 (6–9) 7 (6–8) 4.9 .09
biopsychosocial assessment 9 (8–10) 9 (7–10) 7 (7–9) 2.6 .28
promoting patients’ self-monitoring 5 (3.75–6.5) 6 (5–7) 5 (4–6) .24 .89
follow up of patients’ self-monitoring 6 (4–8) 8 (6–10) 7 (5–10) 4.0 .14
functional behavior analysis 5 (4–6) 7 (4.75–8.25) 6 (4–7) 4.3 .12
basic skills 9 (7–10) 10 (8–10) 8 (7–9) 5.5 .06
applied skills 8 (7–9) 8 (6.75–9.25) 7 (7–8) 1.5 .48
maintenance 8 (6–10) 9 (7–9) 8 (7–10) 1.4 .50
Self-efficacy†
patients’ goal setting 5 (4–6) 7 (6–8) 6 (5–7) 17.6 <.01
biopsychosocial assessment 6 (5–8) 7 (6–8) 7 (6–8) 1.9 .39
promoting patients’ self-monitoring 4 (3–5) 7 (6–8) 6 (5–7) 16.7 <.01
follow up of patients’ self-monitoring 4 (0–8) 8 (5–9) 6 (5–8) 12.4 <.01
functional behavior analysis 4 (1–4) 5 (4–7) 5 (4–7) 10.9 <.01
basic skills 6 (4–8) 8 (7–10) 8 (7–9) 9.0 <.01
applied skills 5 (3–7) 8 (5–8) 7 (7–8) 4.7 .10
maintenance 5 (4–6) 7 (6–8) 7 (6–8) 14.6 <.01
Readiness†
patients’ goal setting 6 (5–7) 8 (5–9) 7 (5–8) 3.7 .15
biopsychosocial assessment 7 (5–10) 7 (5–8) 7 (6–8) .41 .81
promoting patients’ self-monitoring 5 (3–7) 7 (2–10) 6 (5–7) 3.1 .22
follow up of patients’ self-monitoring 5 (0–9) 7 (5–9) 6 (5–8) 5.6 .06
functional behavior analysis 4 (2–5) 5 (4–7) 5 (4–6) 5.9 .05
basic skills 8 (6–9) 8 (7–10) 8 (7–9) 3.7 .15
applied skills 6 (4–8) 8 (5–8) 7 (6–8) 2.4 .30
maintenance 5 (4–8) 7 (6–8) 7 (7–8) 6.0 .05
†Scale 0–10. 0 means not at all, and 10 means very important/capable/ready. p 0.05, p < 0.01. 4 J. FRITZ ET AL.
core component functional behavioral analysis and maintenance was also found. However, the post hoc analyses could not confirm the functional behavioral analysis component. The increase in readiness for using the behavioral medicine core component maintenance occurred after the implementation intervention and was sustained after three months. No changes in beliefs were seen in the experimental group concerning the importance of using the behavioral medicine approach or in the control group concerning any beliefs.
Effects on the physiotherapists’ clinical actions and verbal expressions
There was a significant increase in the observed Chi2 (2, N ¼ 12)¼12.3, p < 0.01 and documented Chi2
(2, N¼ 12)¼9.5, p ¼ 0.01 clinical actions and verbal expressions when comparing pre-, post- and 6-month follow-up values. However, no difference was found regarding self-reported clinical actions and verbal
expressions Chi2 (2, N¼ 11)¼2.4, p ¼ 0.30. According to the post hoc tests on observed clinical actions and verbal expressions, the changes occurred after the implementation intervention (z¼ 2.70, N-Ties ¼ 13, p ¼ 0.01) but were not sustained at the three-, six- or twelve-month follow-up. According to the post hoc tests on docu-mented clinical actions and verbal expressions, the changes occurred after the implementation intervention (z¼ 2.88, N-Ties ¼ 14, p< 0.01) and were sustained at the three-month follow-up (z¼ 2.33, N-Ties ¼ 8, p ¼ 0.02) but not beyond. The median values of identified clinical actions and verbal expressions in the experi-mental and control groups are illustrated inFigure 2.
The changes in the experimental group were mainly related to the behavioral medicine components patient’s goal setting (obser-vation: z¼ 2.33, N-Ties ¼ 9, p ¼ 0.02; self-reports: z ¼ 2.36, N-Ties ¼ 9, p ¼ 0.02; documentation: z ¼ 2.45, N-Ties ¼ 6, p ¼ 0.01), assessment (documentation: z¼ 2.33, N-Ties ¼ 9, p ¼ 0.02), patient’s self-monitoring (observation: z ¼ 2.26, N-Ties ¼ 6, p ¼ 0.02; self-reports: z ¼ 2.64, N-Ties ¼ 8, p ¼ 0.01, documenta-tion: z¼ 2.27, N-Ties ¼ 6, p ¼ 0.02) and functional behavioral ana-lysis (observation: z¼ 1.98, N-Ties ¼ 8, p ¼ 0.05). The range of the numbers of changed observed clinical actions and verbal expres-sions was large within the experimental group (Figure 3); how-ever, no changes were observed in the control group. The effect size for observed behavioral medicine related changes in clinical actions and verbal expressions after the implementation interven-tion was large (r¼ 0.72) in the experimental group.
Discussion
The multifaceted facilitation methods seem to have been most successful in increasing the physiotherapists’ clinical behaviors regarding self-efficacy for using the behavioral medicine approach. The facilitation methods used in this study included several important components for increasing self-efficacy [27,37], such as enabling the physiotherapists to master increasingly chal-lenging tasks by setting individual goals, reinforcing the physio-therapists’ success and providing social support from peers and the facilitator. The results indicate that the facilitation methods used had the intended impact on self-efficacy.
The facilitation support also seems to have impacted the phys-iotherapist’s use of behavioral medicine related clinical actions and verbal expressions but not their ability to sustain an increased use of these clinical behaviors. Difficulties in sustaining clinical behaviors regarding actions and verbal expressions have been confirmed in many studies included in a review by Wiltsey Stirman et al. [38]. These authors found that partial continuation was more common than for the entire intervention; however, this result was not seen in our current study. They also found that resources and workforce stability particularly influenced sustain-ability as well as the context (policies, culture, structure), innov-ation itself (adaptability, effectiveness) and process (fidelity, monitoring, evaluation) [38].
Immediately after the implementation intervention, the physio-therapists’ clinical actions and verbal expressions increased regarding four of the seven behavioral medicine components, all concerning assessment and analyses of the patients’ problems. Rusk et al. [39] propose that an intervention needs to address multiple domains, helping the system to “tip over” and change, and that self-efficacy can be such a domain. Although beliefs about one’s capabilities have been shown to be important predic-tors for changing behavior [26], the increased self-efficacy in this study did not seem to have sufficient impact to change the physi-otherapists’ clinical actions and verbal expressions for all the Figure 2. Median values and interquartile ranges (IQRs) for observed,
self-reported and documented behavioral medicine-related clinical actions and verbal expressions for the experimental and control groups pre-implementation and over time.
behavioral medicine components. Intentions have also been shown to be important predictors of changing behavior [26]. According to the results, the physiotherapists did not value the behavioral medicine components as important in their clinical practice, which probably affected their intentions to use them. The physiotherapists’ readiness for using the behavioral medicine components varied. However, they did not doubt their own confi-dence in using the components, indicating that they might have been hindered by something else. Their perceived importance of the behavioral medicine approach could be one barrier, and con-textual factors affecting the implementation could be another.
The effect size of the facilitation intervention was large accord-ing to Field’s [36] effect size levels but also in relation to other studies in which different implementation methods were shown to have moderate effect sizes [13]. The physiotherapists’ experien-ces of these methods and their perception of which methods were most valuable for supporting their clinical behavior change contribute useful information for future studies.
Despite the lack of sustainability, the results could yet be seen as promising, compared to those of other studies where no changes in clinical actions have been found when implementing a behavioral medicine approach using local opinion leaders [4], a university course [11] or a training program in combination with supervision [12] as implementation methods. In a review by Alagoz et al. [17], all studies including facilitation of practice were reported to have significant effects. Similar to our study, frequent individualized follow-ups were important components of facilita-tion in these studies. Our findings are comparable to results from
previous studies in which facilitation has been found to be a use-ful implementation method, particularly when the focus is on changing clinical behavior [18,23,40].
No changes in clinical actions and verbal expressions were found for the behavioral medicine components concerning treat-ment of patients and support of patients’ maintenance of behav-ior change. One reason could be the complexity of the behavbehav-ioral medicine approach. According to the observation protocol (see
Supplementary Material), the behavioral medicine approach in the current study involved 58 different observable clinical actions and verbal expressions of the physiotherapist. The multiplicity of clinical behaviors for physiotherapists to adopt contributes to its complexity [9] and emphasizes the challenge for successful imple-mentation [41].
The interquartile ranges displayed in Figure 2 indicate a vari-ability in changed clinical actions and verbal expressions among the physiotherapists in the experimental group, implying that some individuals had implemented the behavioral medicine approach to a considerably larger extent than others. Further investigation of these individual differences is important for deep-ening the understanding of the mechanisms of impact for behav-ior change. A process evaluation of the current intervention will be reported elsewhere.
Strengths and limitations
When the sample is small, only non-parametric statistical analyses can be used, and they are not as powerful as parametric analyses Figure 3. The ranges of the numbers of changed observed behavioral medicine related clinical actions and verbal expressions in the experimental and control groups after the implementation intervention.
for detecting existing differences [35]. We are aware that small differences might not have been noted, and the results should therefore be interpreted with caution.
The physiotherapists were allocated to groups based on their own wishes. This could potentially lead to bias if physiotherapists already interested in, and with knowledge about the behavioral medicine approach chose to participate in the experimental group. However, we perceived that the interest in the behavioral medicine approach varied within both groups, and there were no significant differences of performed clinical behavior between the groups before the implementation intervention.
The observation protocol, log-book for self-reporting and checklist for analysis of patient records assessed the behavioral medicine related clinical actions and verbal expressions as pre-sent/not present instead of assessing the frequency, implying that there may be quality differences in the physiotherapists’ perform-ance depending on frequency as well as variations in performperform-ance [42]. However, we had no intention to identify the frequency or quality of performance of the behavioral medicine approach in this study. Another weakness related to the instruments used is that their sensitivity to detect changes has not been investigated. It is therefore possible that small changes occurred but were not registered in the observation protocol. To strengthen the results and decrease the risk for type I error, the observed, self-reported and documented data were triangulated [35]. The findings from observations and patient records in the experimental group pointed in the same direction, thereby strengthening the internal validity of the results.
A strength of the study was the design aimed to decrease threats to the internal validity of the study results. By including a control group whose implementation was supported in the way new guidelines are usually implemented, we were able to control for historical effects. Additionally, to minimize the effect of history, the implementation intervention period started at the same time for all participants, and all physiotherapists were assessed during the same time period. To avoid diffusion of the intervention [32], all participants at the same clinic had to belong to either the experimental or control group. The physiotherapists were also advised to not discuss the behavioral medicine approach or the implementation intervention with colleagues outside the clinic.
Two treatment sessions were video recorded for each physio-therapist at pre- and post-implementation intervention to strengthen the measurement validity. This strategy led to compre-hensive data that might have decreased a random effect. However, some behavioral medicine components, such as per-forming and communicating an functional behavioral analysis and supporting patients’ maintenance of behavior change and relapse prevention, probably occurred less frequently than other behav-ioral medicine components. Despite the comprehensive amount of data, these components may have been missing in the observed treatment sessions. This weakness may explain the scarce number of observed behaviors related to functional behav-ioral analysis, maintenance and relapse prevention. To avoid biased ratings, the observers were blinded to which group the physiotherapists in the video recordings belonged and which measurement point was observed.
At the post-implementation assessment of clinical actions and verbal expressions, there were two dropouts, one in each group. No differences between these individual physiotherapists and the rest of the two groups of physiotherapists were found at the pre-implementation assessment. Therefore, these dropouts probably did not affect the overall findings for clinical actions and verbal expressions post implementation. During the follow-up period,
there were several dropouts and missing data. When data from one measurement point were missing for a participant, the par-ticipant was excluded from the analysis, which led to a small sam-ple. To achieve as large a sample as possible, Friedman’s test was calculated using pre- post- and 6-month follow-up values.
In conclusion, facilitation, as used in this study, contributed to changes in physiotherapists’ clinical behavior, but the changes were not sustained at follow-up. Although self-efficacy beliefs for relevant clinical behaviors increased, this increase was not enough to affect sustained changes in physiotherapists’ clinical behavior. The findings indicate that future studies should focus on targeted actions to achieve sustained behavior change.
Disclosure statement
No potential conflict of interest was reported by the authors.
Funding
Funding was provided by AFA Insurance, Sweden. Assistance with data collection was provided by Ann-Christin Johansson, and Thomas Overmeer is greatly appreciated.
ORCID
Johanna Fritz http://orcid.org/0000-0002-4616-521X
References
[1] Grol R. Successes and failures in the implementation of evi-dence-based guidelines for clinical practice. Med Care. 2001;39:46–54.
[2] Stevenson K, Lewis M, Hay E. Does physiotherapy manage-ment of low back pain change as a result of an evidence-based educational programme? J Eval Clin Pract. 2006;12: 365–375.
[3] Jordan KP, Kadam UT, Hayward R, et al. Annual consult-ation prevalence of regional musculoskeletal problems in primary care: an observational study. BMC Musculoskelet Disord. 2010;11:144.
[4] Breivik H, Collett B, Ventafridda V, et al. Survey of chronic pain in Europe: prevalence, impact on daily life, and treat-ment. Eur J Pain. 2006;10:287–333.
[5] Hill JC, Whitehurst DGT, Lewis M, et al. Comparison of stratified primary care management for low back pain with current best practice (STarT Back): a randomised controlled trial. Lancet. 2011;378:1560–1571.
[6] Lamb SE, Lall R, Hansen Z, et al. A multicentred rando-mised controlled trial of a primary care-based cognitive behavioural programme for low back pain. The Back Skills Training (BeST) trial. Health Technol Assess. 2010;14:1–253. [7] SBU. Rehabilitering vid långvarig sm€arta: en systematisk
lit-teratur€oversikt [Rehabilitation of patients with chronic pain conditions: a systematic review]. Stockholm (Sweden): The Swedish Council on Technology Assessment in Health Care; 2010. (SBU-rapport nr 177/1þ 2). Swedish.
[8] Gray H, Howe T. Physiotherapists’ assessment and manage-ment of psychosocial factors (Yellow and Blue Flags) in indi-viduals with back pain. Phys. Ther Rev. 2013;18:379–394. [9] Sanders T, Foster N, Bishop A, et al. Biopsychosocial care
and the physiotherapy encounter: physiotherapists’
accounts of back pain consultations. BMC Musculoskelet Disord. 2013;14:65.
[10] Synnott A, O’keeffe M, Bunzli S, et al. Physiotherapists may stigmatise or feel unprepared to treat people with low back pain and psychosocial factors that influence recovery: a systematic review. J Physiother. 2015;61:68–76.
[11] Overmeer T, Boersma K, Main CJ, et al. Do physical thera-pists change their beliefs, attitudes, knowledge, skills and behaviour after a biopsychosocially orientated university course?. J Eval Clin Pract. 2009;15:724–732.
[12] Sandborgh M, Åsenl€of P, Lindberg P, et al. Implementing behavioural medicine in physiotherapy treatment. Part II: adherence to treatment protocol. Adv Physiother. 2010;12: 13–23.
[13] Grimshaw J, Eccles M, Lavis J, et al. Knowledge translation of research findings. Implement Sci. 2012;7:50
[14] Grimshaw J, Eccles M, Thomas R, et al. Toward evidence-based quality improvement: evidence (and its limitations) of the effectiveness of guideline dissemination and implementation strategies 1966–1998. J Gen Intern Med. 2006;21:14–20.
[15] van der Wees PJ, Jamtvedt G, Rebbeck T, et al. Multifaceted strategies may increase implementation of physiotherapy clinical guidelines: a systematic review. Aust J Physiother. 2008;54:233–241.
[16] Rycroft-Malone J, Seers K, Chandler J, et al. The role of evi-dence, context, and facilitation in an implementation trial: implications for the development of the PARIHS framework. Implement Sci. 2013;8:28.
[17] Alagoz E, Chih M-Y, Hitchcock M, et al. The use of external change agents to promote quality improvement and organizational change in healthcare organizations: a sys-tematic review. BMC Health Serv Res. 2018;18:42
[18] Baskerville NB, Liddy C, Hogg W. Systematic review and meta-analysis of practice facilitation within primary care settings. Ann Fam Med. 2012;10:63–74.
[19] Nagykaldi Z, Mold JW, Aspy C. Practice facilitators: a review of the literature. Fam Med. 2005;37:581–588.
[20] Cranley LA, Cummings GG, Profetto-Mcgrath J, et al. Facilitation roles and characteristics associated with research use by healthcare professionals: a scoping review. BMJ Open 2017;7:e014384.
[21] Dogherty EJ, Harrison MB, Graham ID. Facilitation as a role and process in achieving evidence-based practice in nurs-ing: a focused review of concept and meaning. Worldviews Evid Based Nurs. 2010;7:76–89.
[22] Berta W, Cranley L, Dearing JW, et al. Why (we think) facili-tation works: insights from organizational learning theory. Implement Sci. 2015;10:141.
[23] Harvey G, Kitson A. Implementing evidence-based practice in healthcare: a facilitation guide. London (UK): Routledge/ Taylor & Francis Group; 2015.
[24] Stetler CB, Legro MW, Rycroft-Malone J, et al. Role of "external facilitation" in implementation of research findings: a qualitative evaluation of facilitation experien-ces in the Veterans Health Administration. Implement Sci. 2006;1:23.
[25] Glegg SMN, Livingstone R, Montgomery I. Facilitating interprofessional evidence-based practice in paediatric
rehabilitation: development, implementation and evalu-ation of an online toolkit for health professionals. Disabil Rehabil. 2016;38:391–399.
[26] Godin G, Belanger-Gravel A, Eccles M, et al. Healthcare pro-fessionals’ intentions and behaviours: a systematic review of studies based on social cognitive theories. Implement Sci. 2008;3:36.
[27] Bandura A. Self-efficacy: the exercise of control. Basingstoke (UK): W. H. Freeman; 1997.
[28] Bandura A. Social cognitive theory: an agentic perspective. Asian J Soc Psychol. 1999;2:21–41.
[29] Glanz K, Bishop DB. The role of behavioral science theory in development and implementation of public health inter-ventions. Annu Rev Public Health. 2010;31:399–418. [30] Sundel M, Sundel S. Behavior change in the human
serv-ices: behavioral and cognitive principles and applications. 5. ed. London (UK): Sage; 2005.
[31] Des Jarlais DC, Lyles C, Crepaz N. Improving the reporting quality of nonrandomized evaluations of behavioral and public health interventions: the TREND statement. Am J Public Health. 2004;94:361–366.
[32] Kazdin AE. Research design in clinical psychology. Boston (MA): Pearson; 2017.
[33] Fritz J, S€oderb€ack M, S€oderlund A, et al. The complexity of integrating a behavioral medicine approach into physio-therapy clinical practice. Physiother Theory Pract. 2018. DOI:10.1080/09593985.2018.1476996.
[34] Åsenl€of P, Denison E, Lindberg P. Individually tailored treat-ment targeting motor behavior, cognition, and disability: 2 Experimental single-case studies of patients with recurrent and persistent musculoskeletal pain in primary health care. Phys Ther. 2005;85:1061–1077.
[35] Plichta SB, Kelvin EA, Munro BH. Munro’s statistical meth-ods for health care research. Philadelphia (PA): Wolters Kluwer Health/Lippincott Williams & Wilkins; 2012.
[36] Field A. Discovering statistics using IBM SPSS statistics: and sex and drugs and rock ’n’ roll. 4. ed. London (UK): Sage; 2013.
[37] McAlister AL, Perry CL, Parcel GS. How individuals, environ-ments, and health behaviors interact: social cognitive the-ory. In: Glanz K, Rimer BK, Viswanath KV, editors. Health behavior and health education: theory, research and prac-tice. 4 ed. San Francisco (CA): Jossey-Bass; 2008.
[38] Wiltsey Stirman S, Kimberly J, Cook N, et al. The sustain-ability of new programs and innovations: a review of the empirical literature and recommendations for future research. Implement Sci. 2012;7:1–19.
[39] Rusk RD, Vella-Brodrick DA, Waters L. A complex dynamic systems approach to lasting positive change: the synergis-tic change model. J Posit Psychol. 2018;13:406–418. [40] Harvey G, Loftus-hills A, Rycroft-Malone J, et al. Getting
evi-dence into practice: the role and function of facilitation. J Adv Nurs. 2002;37:577–588.
[41] Grol R, Wensing M, Eccles M, et al. Improving patient care the implementation of change in health care. 2nd ed. Chichester (UK): Wiley-Blackwell; 2013.
[42] Perepletchikova F, Kazdin AE. Treatment integrity and therapeutic change: issues and research recommendations. Clin Psychol Sci Pract. 2006;12:365–383.
