Group-based circuit training to improve mobility after stroke: a cross-sectional survey of German and Austrian physical therapists in outpatient settings

Abstract

BACKGROUND:

A contextual transferability analysis identified group-based circuit training (GCT) as an optimal intervention in German and Austrian outpatient physical therapy to improve mobility post-stroke. GCT incorporates task-oriented, high-repetitive, balance, aerobic and strength training and allows for increased therapy time without increasing personnel.

OBJECTIVE:

To determine the extent to which German and Austrian physical therapists (PTs) use GCT and its components in the outpatient treatment of stroke-related mobility deficits and to identify factors associated with using GCT components.

METHODS:

A cross-sectional online survey was conducted. Data were analyzed descriptively and using ordinal regression.

RESULTS:

Ninety-three PTs participated. None reported using GCT moderately to frequently (4–10/10 patients). The percentage of PTs reporting frequent use (7–10/10 patients) of task-oriented, balance, strength, aerobic, and high-repetitive training was 45.2%, 43.0%, 26.9%, 19.4%, and 8.6%, respectively. Teaching or supervising students, time for evidence-based practice activities at work, and working in Austria was associated with using GCT components frequently.

CONCLUSION:

German and Austrian PTs do not yet use GCT in outpatient physical therapy for stroke. Almost half of PTs, however, employ task-oriented training as recommended across guidelines. A detailed, theory-driven and country-specific evaluation of barriers to GCT uptake is necessary to inform implementation.

Keywords

Stroke physical therapists translational science biomedical circuit-based exercise guideline adherence

1 Introduction

In Germany and Austria, approximately 262,000 and 20,000 people, respectively, experience a first or recurrent stroke each year (Heuschmann et al., 2010). Ninety-four percent of people with stroke report limitations in their walking distance and speed (Brogårdh et al., 2012). Improvement of walking is therefore a major goal for rehabilitation of people with stroke (Paanalahti et al., 2018; Rice et al., 2017). To address stroke-related mobility deficits national German (Deutsche Gesellschaft für Allgemeinmedizin und Familienmedizin, 2020; Dohle et al., 2015) and international guidelines (Royal Dutch Society for Physical Therapy, 2014; Teasell et al., 2020) recommend physical therapy based on motor learning principles, i.e., task-oriented training with high numbers of repetitions.

People with stroke may not comprehensively benefit from scientific innovations in physical therapy, however, because these innovations are not consistently implemented into clinical practice (Lynch et al., 2018). Results from a survey using a case vignette suggest that unrecommended, traditional interventions, such as the Bobath approach, are predominantly used in Germany and Austria to treat stroke-related mobility deficits in the outpatient setting (Diermayr et al., 2021; Greisberger et al., 2022). These findings thereby confirm the worldwide knowledge to practice gap for German and Austrian physical therapy in stroke. Knowledge translation interventions are needed to narrow this gap. Such interventions are particularly important in outpatient physical therapy, because rehabilitation has increasingly shifted to the outpatient setting in both countries (Kaendler et al., 2022; Reiter et al., 2020; Reuther, 2011; Rollnik & Janosch, 2010).

Implementation of scientific innovations should be guided by knowledge translation (KT) methodology (Grol & Wensing, 2004). The “Knowledge-to-Action (KTA) cycle” is a widely used conceptual framework for KT that outlines activities required for sustainable implementation (Field et al., 2014; Graham et al., 2006). Initial steps of the KTA cycle include (1) the identification of a problem, (2) the identification, review and selection of knowledge and (3) the detailed description of the know-do gap (Graham et al., 2006). A subsequent step in the KTA cycle includes the evaluation of barriers and facilitators to the uptake of scientific innovations (Graham et al., 2006). This step is vital to develop knowledge translation interventions addressing context-specific barriers (Atkins et al., 2017; Graham et al., 2006).

In a recently published “contextual transferability analysis” (Diermayr et al., 2020) we addressed the first steps of the KTA cycle aiming to identify a specific evidence-based intervention addressing stroke-related mobility deficits that could be implemented into the German and Austrian outpatient settings. Based on a systematic literature search and an iterative consensus process, we identified group-based circuit training (GCT) as an intervention with the potential for implementation (Diermayr et al., 2020), where GCT is defined as an exercise intervention involving a group format with a therapist-to-patient ratio of 1 : 3 and a focus on practicing functional tasks with high repetitions (English et al., 2017). In most studies, components of GCT included task-oriented training, high-repetitive training, balance training, aerobic and strength training (Eng, 2010; English et al., 2015; English et al., 2017; Pang et al., 2005). English and colleagues found a high quality of evidence supporting associations between GCT and clinically important improvements of walking distance and speed (English et al., 2017). Implementing GCT in Germany and Austria was deemed particularly useful to introduce practice of functional tasks and to increase therapy time and the number of repetitions without increasing personnel given the growing shortage of physical therapists (PTs) in the outpatient setting (Physiodeutschland Deutscher Verband für Physiotherapie [ZVK], 2022; Rappold & Mathis-Edenhofer, 2020). Moreover, GCT does not require specific or expensive equipment.

In line with the KTA cycle, the know-do gap has to be addressed next (Graham et al., 2006). Describing the know-do-gap helps to understand current clinical practice of PTs in detail and to define what needs to be done differently (French et al., 2012) in order to implement GCT. To what extent PTs in Germany and Austria already use GCT or the components of GCT is not yet known.

Therefore, the primary objective was to determine the extent to which German and Austrian PTs are familiar with and use GCT and its components (task-oriented training, high-repetitive training, balance training, aerobic training, strength training) in the treatment of stroke-related mobility deficits in the outpatient setting. The secondary objective was to identify country-, organizational and therapist characteristics associated with the use of GCT and its components. Based on previous national and international survey results we hypothesized that higher educational level, teaching or supervising physical therapy students, having time set aside for evidence-based practice (EBP) activities at work, and employment at research-intensive institutions would be associated with PTs’ use of GCT and its components (e.g. Bernhardsson et al., 2014; Braun et al., 2018; Braun et al., 2022; Diermayr et al., 2015; Salbach et al., 2007; Salbach et al., 2010). Another secondary objective was to evaluate barriers to GCT uptake by country.

2 Methods

A cross-sectional study involving an online survey was conducted. We used the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) to guide study design and reporting (Eysenbach, 2004). The survey was first launched in Germany and then in Austria. The study was conducted according to the guidelines of the Declaration of Helsinki (2013), and the Ethics Committee of the German Physical Therapy Association (ZVK) e.V. (Submission number: 2018-01) and the Ethics Committee of the Danube University Krems (EK GZ 05/2018–2021) approved the study protocol. The German survey was registered with the German Clinical Trials Registry (DRKS00023729).

We have previously reported survey results on preferred assessment and treatment methods of German and Austrian PTs using open-ended questions (Diermayr et al., 2021; Greisberger et al., 2022) This paper focuses on items on knowledge and use of GCT and its components and barriers to GCT uptake.

2.1 Participants and recruitment

Individuals meeting the following criteria were considered eligible: (1) PTs licensed in Germany or Austria, (2) working in outpatient physical therapy settings, and (3) treating at least three patients with stroke per year. Therapists were recruited via physical therapy Facebook groups, physical therapy websites and newsletters of professional associations. In Germany, additionally, email addresses were extracted from the therapist search engine of the German Physical Therapy Association (ZVK) e.V. and the German Federal Association of Self-employed Physiotherapists (IFK) e.V. websites; email addresses of physical therapy outpatient clinics were identified via the Yellow Pages register. Recruitment in Germany was carried out in two exemplary states: Baden-Württemberg and Thuringia.

The survey link and a short text advertising the survey was posted in Facebook groups, on websites and newsletters or included in emails. A multi-step email contact procedure was used to increase the response rate (Dillman, 2007). We completed Facebook posts that included the survey link up to three times. Physical therapists who clicked on the survey link were informed about the study (purpose of study, length of time of questionnaire, data handling) and asked for electronic consent. Moreover, their eligibility for inclusion was screened. The survey was open from September 14 to October 18, 2018 in Germany and from March 11 to 30, 2019 in Austria. The survey was not password-protected, participation was voluntary and there were no financial incentives.

2.2 Development of the questionnaire

The questionnaire was developed using a multi-step process. First, questionnaire items were compiled based on previous EBP surveys (Diermayr et al., 2015; Salbach et al., 2010), psychological constructs for behavior change (Michie et al., 2005), the prescription catalogue for physical therapy outpatient care in Germany (Heilmittel-Richtlinie und Heilmittelkatalog, 2022), and current German (Dohle et al., 2015) and Dutch (Royal Dutch Society for Physical Therapy, 2014) stroke guidelines. The Dutch guideline was included, as it provides exclusively recommendations for physical therapy interventions post stroke. Austrian stroke guidelines were not available for questionnaire development.

Expert interviews were conducted to evaluate face and content validity of the questionnaire. Five PTs with at least five years of professional experience in stroke treatment and/or related teaching activities were interviewed using a semi-structured interview guide. Authors discussed comments of the experts and revised the questionnaire accordingly. The second version was subsequently tested for comprehensibility and user-friendliness using a focus group discussion with five PTs who represented the target group. Their comments were then used to finalize the questionnaire. The final version was reviewed by two independent research assistants to evaluate its usability and completion time (for detail see our previous study (Diermayr et al., 2021)).

We adapted the questionnaire to the Austrian context by removing items only pertinent to the German context and making minor language adaptations. Cognitive interviews were conducted with eight PTs varying in age, gender, type of educational degree (diploma, BSc, MSc), state, and location (urban, rural) to evaluate comprehensibility. As a result, items pertaining to insurance issues were deleted as the insurance system in Austria was in transition during the time of the study (for detail see our previous study (Greisberger et al., 2022)).

2.3 Questionnaire

The questionnaire was created using the online tool “SoSci Survey”. The final version consisted of 39 items in the German version and 35 items in the Austrian version. For the current study 15 items were used. The questionnaire included adaptive questioning (e.g., only participants who indicated no or low GCT use were presented with items on barriers to GCT use). The individual questions were not randomized. The items relevant for this study asked participants about familiarity and frequency of use of task-oriented training, high-repetitive training, balance training, aerobic training, strength training and group-based circuit training (Dohle et al., 2015; Royal Dutch Society for Physical Therapy, 2014). Response categories were “do know” and “do not know” for familiarity with recommendations and “0 times out of 10 eligible patients” (no use), “1–3/10 patients” (low use), “4–6/10 patients” (moderate use) and “7–10/10 patients” (frequent use) for frequency of use. One item in this section addressed barriers to GCT use using a semi-open response format with multiple responses being allowed. Moreover, items on resources as well as sociodemographic items with either open-ended questions or the response categories “yes”, “no” and “do not know” were used.

Participants were able to review and change their responses using the “back” button. The survey was conducted anonymously. Participants’ IP addresses were not stored, and no cookies were set. Therefore, the number of unique website visitors could not be calculated.

2.4 Data analysis

Participants with complete data on all questionnaire items were included in the analysis. Data were checked for multiple entries using the participants’ responses to open-ended questions in the first section of the questionnaire. Completion rates were determined by the ratio of the number of participants who reached the final questionnaire page to the number that met the inclusion criteria and gave informed consent.

To address our primary objective, frequencies and percentages were used to summarize data from items on familiarity with and use of GCT and its components (task-oriented training, high-repetitive training, balance training, aerobic training, strength training) by country and with countries combined. We used a chi-square test to compare knowledge and use between Germany and Austria. Response categories of the five GCT components were dichotomized into “no to low use” (using the component 0–3/10 patients) and “moderate to frequent use” (using the component 4 – 10/10 patients). Because none of the participants reported moderate to frequent GCT use, “GCT use” was dichotomized into “no use” (0/10 patients) and “low use” (1–3/10 patients).

To address our secondary objective, we used ordinal regression evaluating determinants of the use of GCT components. The use of GCT was not included in the ordinal regression, because of its low use across participants. We used a multi-step process to create the dependent variable that captured “frequent use of GCT components”. First, we created a variable that represented frequent use (yes/no) of each GCT component, where “yes” was defined as use 7–10/10 patients, and “no” was defined as use 0–6/10 patients. Second, we created the dependent variable “frequent use of GCT components” by summing the number of GCT components used frequently. GCT use had a 6-point scale that ranged from 0 (uses 0 of the 5 GCT components frequently) to 5 (uses 5 out of 5 GCT components frequently). Because few participants used 4 and 5 components frequently, these two response categories were collapsed, yielding a final 5-point ordinal scale corresponding to the frequent use of 0, 1, 2, 3, or 4-5 GCT components.

Independent variables included country (Germany/Austria), organizational characteristics (research is undertaken in work setting (yes/no); time for EBP is provided at work (yes/no)) and therapist characteristics (academic degree (yes/no), certificate in traditional neurophysiological treatment approach (yes/no), being involved in teaching or supervising physical therapy students (yes/no), learned EBP in PT training (yes/no), clinical specialization in neurological physical therapy (yes/no), work experience (0 – 15 years/>15 years), age (20–39 years/≥40) and using research evidence≥10% in clinical practice (yes/no) among different sources of evidence such as external evidence, clinical experience or patient preferences. Independent variables were screened to determine inclusion in the regression model using Mann Whitney U-Tests where “frequent use of GCT components” was the dependent variable, and each predictor variable was separately entered as independent variable. Variables yielding a p-value <0.25 were carried forward to the ordinal regression model (Hickey et al., 2015). Prior to running the regression analysis, collinearity of the included variables was checked by determining possible correlations using the Phi coefficient for nominal data. Correlations did not exceed Phi = 0.4 except for the relationship of the variables “age” and “years of work experience” (Phi = 0.76). Work experience and not age was entered into the final model given its conceptual relevance (i.e., work experience more accurately reflects years since graduation than age given that people may (re-) enter university/school later in life).

We applied a proportional odds ordinal regression model, wherein we assume that the odds of a unit increase in the dependent variable is the same across categories (Scott et al., 1997) suggesting that the odds of using 1 versus 4 and 5 GCT components frequently is considered to be the same as the odds of using 2 versus 3 GCT components frequently. This assumption of proportional odds was met as shown in a non-significant test of parallel lines (p = 0.470). Model fit was evaluated using the Pearson chi-square and Deviance chi square tests (O’Connell, 2006). Non-significant results point towards a well-fitting model. Odds ratios and associated 95% confidence intervals (CIs) were reported. An odds ratio > 1 indicates that the probability of falling into a higher category of the dependent variable (i.e., using more GCT components frequently) increases. When the confidence interval excluded 1, the association between dependent and independent variable was considered significant.

We described sociodemographic and practice characteristics of participants using frequencies and percentages for categorial variables and median and interquartile range for continuous variables by country and with countries combined. We used a chi-square test to compare categorical variables and a Mann Whitney U-test to compare medians between countries. Prior to running chi-square tests, response categories of categorical variables with≥3 response categories were combined to create binary variables.

Percentages were reported for GCT barriers by country. When responses in the open-ended “other” category aligned with pre-existing categories of barriers, those were assigned to the pre-existing categories.

The level of significance for all statistical tests and procedures was set at α <0.05, if not explicitly stated otherwise. SPSS Statistics version 27 (IBM Corp., Armonk, NY, USA) was used for data analysis.

3 Results

The survey link was clicked 1446 times. Of the 365 persons who agreed to participate, 230 met the inclusion criteria. Eighty-six participants discontinued the questionnaire following the items on inclusion criteria, 25 participants discontinued the questionnaire during/after the items using the case scenario in the first section (see Fig. 1). The final sample size for the current data analysis was therefore 93 participants with a corresponding completion rate of 39.9% .

Fig. 1

Process of recruitment in survey.

Table 1 shows the therapist and organizational characteristics of participants by country and all participants combined. Participants were between 26 and 71 years of age (mean 48.4 (standard deviation (SD) 11.6)). The majority of the participants (79.6%) were female, 24.7% had an academic degree and 87.1% of participants held a certificate in a neurophysiological approach (68/93 PTs in the Bobath approach). The median percentage of sources of evidence used in clinical practice for continuing education, clinical experience and external evidence was 30%, 30%, and 5%, respectively. The median percentage of work time was highest for private practice (median 70%) and home health care (median 20%). Employers provided time for EBP to 36.6% of participants. Of the 15 therapist and organizational characteristics, seven were associated with country. Compared to PTs in Austria, PTs in Germany were more likely to be older (χ² = 12.85, p = 0.000), have more years of work experience (χ² = 6.38, p = 0.012), a lower level of education (χ² = 11.45, p = 0.001), and spend a higher percentage of worktime in a private practice setting (U = 332.0, p = 0.000).

Table 1

Participant characteristics

Characteristic level		All	Germany	Austria	χ² (p-value)
		(n = 93)	(n = 63)	(n = 30)
		n (%)	n (%)	n (%)
Sociodemographic data
Age	20–29 years	10 (10.8)	2 (3.2)	8 (26.7)	12.85 (0.000)
	30–39 years	22 (23.7)	12 (19.0)	10 (33.3)
	40–49 years	18 (19.4)	10 (15.9)	8 (26.7)
	50 + years	43 (46.2)	39 (61.9)	4 (13.3)
Gender	female	74 (79.6)	45 (71.4)	29 (96.7)	7.96 (0.005)
Work experience	<5 years	6 (6.5)	3 (4.8)	3 (10.0)	6.38 (0.012)
	5–10 years	14 (15.1)	5 (7.9)	9 (30.0)
	11–15 years	10 (10.8)	7 (11.1)	3 (10.0)
	>15 years	63 (67.7)	48 (76.2)	15 (50.0)
Education and clinical practice
Highest degree	Certificate/diploma	70 (75.3)	54 (85.7)	16 (53.3)	11.45 (0.001)
	Bachelor’s degree	12 (12.9)	6 (9.5)	6 (20.0)
	Master’s degree	11 (11.8)	3 (4.8)	8 (26.7)
Learned EBP in PT training		52 (55.9)	33 (52.4)	19 (63.3)	0.99 (0.320)
Certificate in neurophysiological approach ^*)		81 (87.1)	58 (92.1)	23 (76.7)	0.051^‡)
% of different sources of evidence used in clinical practice (median (IQR))	CE courses	30 (20 – 40)	30 (20 – 40)	30 (20 – 33.8)	U = 643.5 (0.143)^{^)}
	My own clinical expertise	30 (22.5 – 40)	30 (20 – 40)	30 (26.3 – 40)	U = 782.0 (0.878)^{^)}
	Prescription	1 (0 – 10)	1 (0 – 10)	3 (0 – 10)	U = 771.0 (0.788)^{^)}
	External evidence	5 (0 – 10)	5 (0 – 10)	7.5 (0 – 13.8)	U = 794.0 (0.969)^{^)}
	Patient preferences	20 (10 – 30)	20 (10 – 27,5)	20 (10 – 30)	U = 704.5 (0.372)^{^)}
	Other	0 (0 – 0)	0 (0 – 0)	0 (0 – 0)	U = 761.5 (0.542)^{^)}
Involved in supervising/teaching PT students		38 (40.9)	25 (39.7)	13 (43.3)	0.11 (0.738)
Work Setting
Clinical specialty	Neurological PT	70 (75.3)	45 (71.4)	25 (83.3)	1.55 (0.214)
	Other	23 (24.7)	18 (28.6)	5 (16.7)
% of time spent in different work settings (median (IQR))	Private practice^¶)	70 (25 – 82.5)	80 (50 – 90)	10 (0 – 50)	U = 332.0 (0.000^{^)}
	Hospital	0 (0 – 0)	0 (0 – 0)	0 (0 – 27.5)	U = 663.5 (0.000)^{^)}
	Rehabilitation center	0 (0 – 0)	0 (0 – 0)	0 (0 – 0)	U = 855.0 (0.169)^{^)}
	Home health care	20 (10 – 50)	20 (10 – 50)	30 (0 – 72.5)	U = 876.5 (0.570)^{^)}
	Community work	0 (0 – 0)	0 (0 – 0)	0 (0 – 0)	U = 944.5 (0.989)^{^)}
	University	0 (0 – 0)	0 (0 – 0)	0 (0 – 0)	U = 882.0 (0.039)^{^)}°⁾
Work location	Rural	31 (33.3)	21 (33.3)	10 (33.3)	0.88 (0.349)
	Small city	28 (30.1)	21 (33.3)	7 (23.3)
	Medium city	19 (20.4)	15 (23.8)	4 (13.3)
	Urban	15 (16.1)	6 (9.5)	9 (30.0)
Time for EBP at work		34 (36.6)	21 (33.3)	13 (43.3)	0.88 (0.349)
Research activities at work		11 (11.8)	5 (7.9)	6 (20.0)	0.166^‡)
Support for CE at work		82 (88.2)	62 (98.4)	20 (66.7)	0.000^‡)
Frequency of using timed walking tests	0/10 patients	34 (36.6)	30 (47.6)	4 (13.3)	6.16 (0.013)
	1–3/10 patients	26 (28.0)	16 (25.4)	10 (33.3)
	4–6/10 patients	11 (11.8)	7 (11.1)	4 (13.3)
	7–10/10 patients	22 (23.7)	10 (15.9)	12 (40.0)

^*)Certificate in the Bobath approach, Proprioceptive neuromuscular facilitation or Vojta therapy. ^‡)Results of Fisher’s exact test (p-value) was reported when the expected cell frequency fell below 5. ^{^)}Mann Whitney U-Test. ^¶)including 3 Austrian participants that worked 2–25% at a daycare center. ^∘)2 Austrian PTs worked at university setting with 30% and 55% of their total work time. EBP = evidence-based practice, PT = physical therapy, IQR = interquartile range, CE = continuing education.

Table 2 shows results on familiarity with GCT and its five components. Overall, 68.8% of participants knew of GCT. Among GCT components, most participants knew of balance training (90.3%) and least participants of high-repetitive training (54.8%). Compared to PTs in Germany, a significantly higher percentage of PTs in Austria knew of aerobic training (χ² = 5.82, p = 0.0014) and GCT (Fisher’s exact test: p = 0.016).

Table 2

Familiarity with GCT and GCT components

Knowledge of intervention n (%)
Recommended intervention	All	Germany	Austria	χ² (p-value)
	(n = 93)	(n = 63)	(n = 30)
	n (%)	n (%)	n (%)
Task-oriented Training	75 (80.6)	48 (76.2)	27 (90.0)	2.483 (0.115)
High-repetitive Training	51 (54.8)	32 (50.8)	19 (63.3)	1.290 (0.256)
Balance Training	84 (90.3)	54 (85.7)	30 (100.0)	0.054^*)
Aerobic Training	77 (82.8)	48 (76.2)	29 (96.7)	5.82 (0.014)
Strength Training	81 (87.1)	52 (82.5)	29 (96.7)	0.095^*)
GCT	64 (68.8)	38 (60.3)	26 (86.7)	0.016^*)

^*)Results of Fisher’s exact test (p-value) was reported when the expected cell frequency fell below 5. GCT = group-based circuit training.

Table 3 presents use of GCT and its five components. While none of the participants reported using GCT moderately (i.e., 4–6/10 patients) or frequently (i.e., 7–10/10 patients times), the percentage of participants reporting frequent use of task-oriented training, balance training, strength training, aerobic training, and high repetitive training was 45.2%, 43.0%, 26.9%, 19.4%, and 8.6%, respectively. A significantly higher percentage of PTs in Austria used strength (χ² = 11.691, p = 0.001) and aerobic training (χ² = 9.145, p = 0.002) moderately to frequently.

Table 3

Frequency of use of GCT and GCT components

Use of intervention n (%)
Recommended intervention	All	Germany	Austria	χ² (p-value)
Frequency of Use	(n = 93)	(n = 63)	(n = 30)
	n (%)	n (%)	n (%)
Task-oriented Training
0/10 patients	20 (21.4)	15 (23.8)	5 (16.6)	1.176 (0.278)
1–3/10 patients	12 (12.9)	9 (14.3)	3 (10.0)
4–6/10 patients	19 (20.4)	16 (25.4)	3 (10.0)
7–10/10 patients	42 (45.2)	23 (36.5)	19 (63.3)
High-repetitive Training
0/10 patients	58 (62.4)	40 (63.5)	18 (60.0)	0.222 (0.637)
1–3/10 patients	13 (14.0)	9 (14.3)	4 (13.3)
4–6/10 patients	14 (15.1)	9 (14.3)	5 (16.7)
7–10/10 patients	8 (8.6)	5 (7.9)	3 (10.0)
Balance Training
0/10 patients	7 (7.5)	7 (11.1)	0	1.392 (0.238)
1–3/10 patients	19 (20.4)	13 (20.6)	6 (20.0)
4–6/10 patients	27 (29.0)	19 (30.2)	8 (26.7)
7–10/10 patients	40 (43.0)	24 (38.1)	16 (53.3)
Aerobic Training
0/10 patients	23 (24.7)	20 (31.7)	3 (10.0)	9.145 (0.002)
1–3/10 patients	26 (28.0)	20 (31.7)	6 (20.0)
4–6/10 patients	26 (28.0)	15 (23.8)	11 (36.7)
7–10/10 patients	18 (19.4)	8 (12.7)	10 (33.3)
Strength Training
0/10 patients	18 (19.4)	18 (28.5)	0	11.691 (0.001)
1–3/10 patients	14 (15.1)	11 (17.5)	3 (10.0)
4–6/10 patients	36 (38.7)	24 (38.1)	12 (40.0)
7–10/10 patients	25 (26.9)	10 (15.9)	15 (50.0)
GCT
0/10 patients	81 (87.1)	56 (88.9)	25 (83.4)	0.558 (0.455)
1–3/10 patients	12 (12.9)	7 (11.1)	5 (16.7)
4–6/10 patients	0	0	0
7–10/10 patients	0	0	0

GCT = Group-based Circuit Training.

We entered six independent variables (country, time for EBP at work, research activities at work, being involved in supervising/teaching physical therapy students, work experience and using external evidence in clinical practice) into the final ordinal regression model. In the final model Pearson chi-square (χ² (118) = 130,784, p = .0.199) and Deviance chi-square [χ² (118) = 108,260, p = 0.729] tests were not significant. In the final model country, having time for EPB at work and being involved in teaching/supervising students remained significant predictors of frequent use of GCT components. The odds that participants from Germany fall into a higher category of frequent GCT use (i.e., that they use more components frequently) are 4 times less likely compared to participants from Austrian (OR = 0.26, 95% CI 0.11 – 0.64). PTs who reported being involved in teaching or supervising students compared to those who were not are 3.4 times more likely to use more GCT components frequently (OR = 3.36, 95% CI 1.39–8.15). Similarly, PTs who reported to have dedicated time for EBP at work are 2.7 times more likely to use more GCT components frequently compared to those who are not provided time for EBP (OR = 2.74, 95% CI 1.11–6.77).

Figure 2 shows barriers for GCT use. In total 59 participants (33 German participants) reported barriers (total n = 113 responses; 70 responses from German participants). The two top barriers were the notion that GCT does not allow for individualized treatment (25.7% and 16.3% of barriers cited by German and Austrian participants, respectively) and logistic constraints (21.4% and 41.9% of barriers cited by German and Austrian participants, respectively).

Fig. 2

Percentage of barriers to the use of group-based circuit training (GCT).

4 Discussion

The study showed that PTs in Germany and Austria rarely use GCT in the outpatient setting for people with stroke. Knowledge and use of GCT components varied. Across countries, about two thirds of PTs reported to either moderately or frequently (4–10/10 patients) use task-oriented training and balance training, whereas high-repetitive training is infrequently used. Austrian PTs showed higher use of aerobic and strength training compared to German PTs. Being involved in teaching, having time for EBP activities at work and working in Austria are associated with using more GCT components frequently. Barriers to GCT use at the therapist, organizational and system levels were identified.

In two separate papers drawing from the same sample, we reported results of open-ended questions using a case vignette in the outpatient stroke rehabilitation setting (Diermayr et al., 2021; Greisberger et al., 2022). Compared to the current study, the vignette-based results showed a lower use of recommended practices: 39% and 47% of the participants’ open responses were categorized as recommended in clinical guidelines (e.g., strength training, endurance training) in Germany and Austria, respectively. Rather, two-thirds of participants reported preferred use of traditional approaches such as Bobath, and none of the German and only five Austrian participants self-reported the use of task-oriented training (Diermayr et al., 2021; Greisberger et al., 2022). Given that closed-ended questions presented in the current study may introduce bias towards the response categories given (Züll, 2015), the adherence reported in the current study may indeed be lower in both countries. In line with the two previous papers (Diermayr et al., 2021; Greisberger et al., 2022) German PTs show lower guideline adherence in the current study.

Results from the current study reveal that recommendations in national (Dohle et al., 2015) and international (Royal Dutch Society for Physical Therapy, 2014; Teasell et al., 2020) guidelines to implement GCT and GCT components, such as high-repetitive training, have not yet been fully implemented into clinical practice in Germany and Austria. Compared to Germany and Austria, the level of implementation of GCT and GCT components in other countries is higher (Ajimsha et al., 2019; Boyne et al., 2017; Doyle & Mackay-Lyons, 2013; Kristensen et al., 2016). For example, a retrospective chart review of 216 case files conducted in Qatar showed that circuit class training was adhered to in 95% of the case files for gait and cardiovascular training (Ajimsha et al., 2019). Moreover, 77% of Canadian PTs surveyed reported using aerobic training for people post-stroke (Doyle & Mackay-Lyons, 2013). It is important to note that compared to Germany, the discrepancy between Austria and the international data on the use of recommended practices is slightly lower.

A recent study surveying PTs treating patients in the subacute stroke stage confirms the rather low use of recommended practices for Germany (Scheffler et al., 2022). However, the study’s participants who worked primarily in inpatient rehabilitation appear to use recommendations at a higher rate compared to our data suggesting a more substantial research-practice gap for the outpatient setting.

Several reasons may explain the rather low use of GCT and its components in both countries. First, the discrepancy may be a function of differences in degree levels. Research use has been associated with degree level of PTs treating stroke (Salbach et al., 2010). An academic degree is held by a greater number of participants in these international studies (Ajimsha et al., 2019; Boyne et al., 2017; Doyle & Mackay-Lyons, 2013; Kristensen et al., 2016; Silva et al., 2015) compared to both the German and the Austrian sample. In line with this interpretation, PTs in our survey reported to use external evidence on average for less than 10% in their clinical practice. Similarly, a survey addressing PTs across settings and clinical specialties in Germany showed that only a third of PTs are frequently searching and using research literature (Braun et al., 2022). In contrast, it appears that PTs in countries that have offered university–level PT education for longer than Germany or Austria such as Canada (Salbach et al., 2010) integrate research into clinical practice more frequently. It appears, therefore, that academic physical therapy degrees serve as catalysts for engagement in recommended practices.

While degree level was not associated with implementation of GCT components in clinical practice in the current study, more Austrian PTs held academic degrees and were using more GCT components frequently than their German counterparts lending further weight to this interpretation.

Differences in participant characteristics between countries may contribute to differences in GCT use and use of its components between German and Austrian PTs. Compared to German PTs, Austrian PTs held higher degrees, were younger and had less work experience. These therapist characteristics are important predictors of EBP engagement (Bernhardsson et al., 2014; Braun et al., 2018; Diermayr et al., 2015; Jette et al., 2003; Salbach et al., 2007). The difference in academic degree may be a function of country differences in the distribution of participants across age groups. Moreover, it may represent the differing models of transitioning from purely vocational to an academic PT training. While Austria currently offers only bachelor degree programs for physical therapy education (Mériaux-Kratochvila, 2021), in Germany, vocational and academic programs are still being offered (Hochschulverbund Gesundheitsfachberufe e.V., 2018). Subsequently, an academic degree (versus a certificate or diploma) is currently held by about 22% and 3% of all Austrian and German PTs, respectively (Deutscher Bundestag, 2021; Mériaux-Kratochvila, 2021).

Second, the lack of organizational and system support to use guideline recommendations in Germany and Austria may serve as a major barrier to guideline uptake. A survey evaluating factors influencing guideline use in international stroke rehabilitation revealed the lack of a mandate to deliver recommended interventions at the system level as an important barrier (Lynch et al., 2021). In line with this a qualitative meta-synthesis summarized, that guidelines can only be implemented when systems support its adoption (Halls et al., 2021). In contrast, a German and an Austrian survey confirm that most clinical institutions do not yet prioritize or mandate external evidence in clinical decision making (Braun et al., 2022; Diermayr et al., 2015), thereby impeding its uptake. Moreover, currently, reimbursement for group-based physical therapy for people with stroke is low in Germany (Verband der Erstatzkassen e.V., 2022) substantiating the lack of support by the health care system. Reimbursement issues were also reported as a barrier to GCT by German participants.

Third, participants in both countries reported predominantly using knowledge gained from continuing education (CE) and their clinical practice but not external evidence. At least two-thirds of PTs in Germany and Austria held a CE certificate in a neurophysiological approach; most of them were Bobath-trained. This treatment approach did not traditionally focus on repetitive training, aerobic and strength training or group-based training (Horak, 1991; Mayston, 2008). One may argue that holding a CE certificate in a traditional neurophysiological approach impedes uptake of best practices. The almost significant finding that more German compared to Austrian PTs hold a certificate in a traditional neurophysiological approach and fewer German than Austrian PTs use GCT components supports this notion. Moreover, a qualitative study embedded in a clinical trial in Canada reported that PTs following the Bobath approach were less prone than PTs not following the Bobath approach to adopt high-repetitive interventions (Connell et al., 2018).

The perceived barrier that GCT cannot be individualized to patients’ needs may be associated with using predominantly knowledge from CE and one’s own clinical experience in decision making processes. Interestingly, Canadian PTs who conducted a group-based 7-day circuit class with stroke survivors also believed that GCT may impede individualizing treatment to patients’ needs (van Kessel et al., 2017). These perceptions are supported by findings from qualitative studies where PTs perceive that guideline recommendations do not sufficiently take into account variability among stroke patients (Halls et al., 2021; McCluskey et al., 2013).

4.1 Implications for implementation

Our findings suggest that KT interventions aimed at implementation of GCT and its components in Germany and Austria should target foremost the use of GCT itself and the component high-repetitive training.

Importantly, our results show that KT interventions need to differ between Germany and Austria: use of aerobic training and strength training need to be addressed primarily in Germany. Also, therapist-level, organizational and system level barriers appear to differ across countries warranting individual KT implementation interventions for Germany and Austria, respectively. Our preliminary evaluation of barriers shows that therapist-level barriers such as believes about consequences (e.g., participants’ notion that GCT cannot be individualized to patients) or lack of knowledge are more pronounced in German physical therapy. Prior to designing KT interventions, a detailed analysis of barriers and facilitators will need to be undertaken (Atkins et al., 2017).

The finding, that PTs involved in teaching or supervising students are more likely to use more GCT components frequently, suggests that students in clinical internships can represent a low threshold strategy for knowledge to disseminate into clinical practice; particularly for older and more experienced PTs who are more likely than young PTs to supervise students.

4.2 Limitations

The generalizability of our results is limited due to the relatively small sample size (n = 93). The most important determinant for the small sample size compared to more general surveys on EBP in physical therapy in Germany (Braun et al., 2018; Braun et al., 2022) and Austria (Diermayr et al., 2015) was most likely our inclusion criteria that PTs had to treat stroke patients in the outpatient setting. Moreover, we collected data in only two states in Germany. Baden-Württemberg can be seen as representative of the western federal states with higher salaries for PTs and a lower utilization of physical therapy services. Thuringia is an eastern federal state with lower salaries and a higher utilization of physical therapy services typical of the eastern federal states (Rommel & Prütz, 2017). We are therefore confident that the results are generalizable to other German states.

The majority of participants held a diploma or certificate in PT; therefore, our results primarily apply to PTs without academic education. However, the percentage of PTs with an academic degree is higher in our sample compared to the PT population in Germany (ZVK, 2022) and Austria (Mériaux-Kratochvila, 2021), further limiting generalizability. Moreover, the participants in our study may be more interested in EBP and guideline recommendations than the general population of PTs in Germany and Austria. Also, social desirability may have influenced the responses. This potential bias in our sample together with the lower guideline adherence in our two previous studies using open-ended questions (Diermayr et al., 2021; Greisberger et al., 2022), suggest that the results of the current study need to be interpreted with caution.

Despite our results of a relatively low use of recommended practices in German and Austrian PTs, we cannot conclude that the overall quality of physical therapy care for people with stroke in the German and Austrian outpatient setting is affected. However, our data suggest that the potential of physical therapy may not yet be fully exploited in Germany and Austria.

5 Conclusion

While GCT is recommended in clinical practice guidelines, PTs in Germany and Austria do not yet use this intervention to treat mobility deficits in people with stroke in the outpatient setting. Study findings suggest that there is a substantial know-do-gap. Implementation efforts need to address foremost the adoption of group-based training and high-repetitive training. In line with the KTA cycle, a detailed, theory-driven evaluation of barriers and facilitators to GCT uptake – separately in Germany and Austria – is necessary in a subsequent step. These results together with data of the current study can be used to design context-specific implementation strategies for GCT uptake.

Footnotes

Conflict of interest

The authors declare that there are no conflicts of interests.

Acknowledgments

The authors would like to thank the experts and PTs for their expertise and time during questionnaire development as well as the PTs who participated in the survey.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

Ajimsha,

M. S.

, Smithesh,

, Al-Mudahka,

(2019) Adherence of physical therapy with clinical practice guidelines for the rehabilitation of stroke in an active inpatient setting, Disability and Rehabilitation, 41(15), 1855–1862.

Atkins,

, Francis,

, Islam,

, O’Connor,

, Patey,

, Ivers,

, Foy,

, Duncan,

E. M.

, Colquhoun,

, Grimshaw,

J. M.

, Lawton,

, Michie,

(2017) A guide to using the Theoretical Domains Framework of behaviour change to investigate implementation problems, Implementation Science, 12(1), 77.

Bernhardsson,

, Johansson,

, Nilsen,

, Öberg,

, Larsson,

M. E. H.

(2014) Determinants of guideline use in primary care physical therapy: A cross-sectional survey of attitudes, knowledge, and behavior, Physical Therapy, 94(3), 343–354.

Boyne,

, Billinger,

, MacKay-Lyons,

, Barney,

, Khoury,

, Dunning,

(2017) Aerobic Exercise Prescription in Stroke Rehabilitation: A Web-Based Survey of US Physical Therapists, Journal of Neurologic Physical Therapy, 41(2), 119–128.

Braun,

, Ehrenbrusthoff,

, Bahns,

, Happe,

, Kopkow,

(2022) Adherence to and influencing factors of evidence-based practice in physiotherapeutic care in Germany: a cross-sectional study [Adhärenz und Einflussfaktoren evidenzbasierter Praxis in der physiotherapeutischen Versorgung in Deutschland – eine Querschnittstudie], Zeitschrift fuer Evidenz, Fortbildung und Qualitaet im Gesundheitswesen, 168, 8–20.

Braun,

, Rieckmann,

, Weber,

, Grüneberg,

(2018) Current use of measurement instruments by physiotherapists working in Germany: A cross-sectional online survey, BMC Health Services Research, 18(1), 810.

Brogårdh,

, Flansbjer,

U.-B.

, Lexell,

(2012) Self-reported walking ability in persons with chronic stroke and the relationship with gait performance tests, PM & R: the Journal of Injury, Function, and Rehabilitation, 4(10), 734–738.

Connell,

L. A.

, Klassen,

T. K.

, Janssen,

, Thetford,

, Eng,

J. J.

(2018) Delivering intensive rehabilitation in stroke: Factors influencing implementation, Physical Therapy, 98(4), 243–250.

Deutsche

Gesellschaft für Allgemeinmedizin und Familienmedizin.

(2020), Schlaganfall: AWMF-Register_N. DEGAM-Leitlinie Nr., 8, Berlin. https://www.awmf.org/uploads/tx_szleitlinien/053-011l_S3_Schlaganfall_2021-03.pdf.

10.

Diermayr,

, Schachner,

, Eidenberger,

, Lohkamp,

, Salbach,

N. M.

(2015) Evidence-based practice in physical therapy in Austria: Current state and factors associated with EBP engagement, Journal of Evaluation in Clinical Practice, 21(6), 1219–1234.

11.

Diermayr,

, Schomberg,

, Barthelmes,

L.-S.

, Greisberger,

, Elsner,

, Salbach,

N. M.

(2021) Current practice of physical therapists in outpatient stroke rehabilitation: a cross-sectional survey in Baden-Württemberg and Thuringia / Arbeitsweise von Physiotherapeut*innen in der ambulanten Schlaganfallrehabilitation: eine Querschnittsumfrage in Baden-Württemberg und Thüringen, International Journal of Health Professions, 8(1), 48–59.

12.

Diermayr,

, Schomberg,

, Greisberger,

, Elsner,

, Gronwald,

, Salbach,

N. M.

(2020) Task-Oriented Circuit Training for Mobility in Outpatient Stroke Rehabilitation in Germany and Austria: A Contextual Transferability Analysis, Physical Therapy, 100(8), 1307–1322.

13.

Dillman,

D. A.

(2007) Mail and internet surveys: The tailored design method (2nd ed.). John Wiley & Sons Inc.

14.

Dohle,

, Quintern,

, Saal,

, Stephan,

, Tholen,

, Wittenberg,

(2015) Rehabilitation der Mobilität nach Schlaganfall (ReMoS), Neurologie & Rehabilitation, 21(7), 355–494.

15.

Doyle,

, Mackay-Lyons,

(2013) Utilization of aerobic exercise in adult neurological rehabilitation by physical therapists in Canada, Journal of Neurologic Physical Therapy, 37(1), 20–26.

16.

Eng,

J. J.

(2010) Fitness and Mobility Exercise (FAME) Program for stroke, Topics in Geriatric Rehabilitation, 26(4), 310–323.

17.

English,

, Bernhardt,

, Crotty,

, Esterman,

, Segal,

, Hillier,

(2015) Circuit class therapy or seven-day week therapy for increasing rehabilitation intensity of therapy after stroke (CIRCIT): A randomized controlled trial, International Journal of Stroke, 10(4), 594–602.

18.

English,

, Hillier,

S. L.

, Lynch,

E. A.

(2017) Circuit class therapy for improving mobility after stroke, The Cochrane Database of Systematic Reviews, 6, CD007513.

19.

Eysenbach,

(2004) Improving the Quality of Web Surveys: The Checklist for Reporting Results of Internet E-Surveys (CHERRIES), Journal of Medical Internet Research, 6(3), e34.

20.

Field,

, Booth,

, Ilott,

, Gerrish,

(2014) Using the Knowledge to Action Framework in practice: A citation analysis and systematic review, Implementation Science, 9, 172.

21.

French,

S. D.

, Green,

S. E.

, O’Connor,

D. A.

, McKenzie,

J. E.

, Francis,

J. J.

, Michie,

, Buchbinder,

, Schattner,

, Spike,

, Grimshaw,

J. M.

(2012) Developing theory-informed behaviour change interventions to implement evidence into practice: A systematic approach using the Theoretical Domains Framework, Implementation Science,, 7, 38.

22.

Graham,

I. D.

, Logan,

, Harrison,

M. B.

, Straus,

S. E.

, Tetroe,

, Caswell,

, Robinson,

(2006) Lost in knowledge translation: Time for a map? Journal of Continuing Education in the Health Professions, 26(1), 13–24.

23.

Greisberger,

, Mayer,

, Barthelmes,

L.-S.

, Diermayr,

(2022) Arbeitsweise von Physiotherapeut*innen in der ambulanten Schlaganfallrehabilitation: eine Querschnittsumfrage in Österreich, International Journal of Health Professions, 9(1), 151–163.

24.

Grol,

, Wensing,

(2004) What drives change? Barriers to and incentives for achieving evidence-based practice, Medical Journal of Australia, 180(S6), 57–60.

25.

Halls,

, Murray,

, Sellar,

(2021) Why allied health professionals use evidence-based clinical guidelines in stroke rehabilitation: A systematic review and meta-synthesis of qualitative studies, Clinical Rehabilitation, 35(11), 1611–1626.

26.

Heilmittel -Richtlinie und Heilmittelkatalog: Das Standardwerk für die regelkonfrome Verordnung von Heilmitteln. (2022) Urban & Fischer Verlag/Elsevier GmbH.

27.

Heuschmann,

, Busse,

, Wagner,

, Endres,

, Villringer,

, Röther,

, Kolominsky-Rabas,

, Berger,

(2010) Schlaganfallhäufigkeit und Versorgung von Schlaganfallpatienten in Deutschland, Aktuelle Neurologie, 37(07), 333–340.

28.

Hickey,

G. L.

, Dunning,

, Seifert,

, Sodeck,

, Carr,

M. J.

, Burger,

H. U.

, Beyersdorf,

, on behalf of the EJCTS and ICVTS Editorial Committees. (2015) Statistical and data reporting guidelines for the European Journal of Cardio-Thoracic Surgery and the Interactive CardioVascular and Thoracic Surgery, European Journal of Cardiothoracic Surgery, 48(2), 180–193.

29.

Hochschulverbund Gesundheitsfachberufe e.V. (2018) Notwendigkeit und Umsetzung einer vollständigen hochschulischen Ausbildung in den Therapieberufen (Ergotherapie, Logopädie und Physiotherapie) – Strategiepapier. https://www.hv-gesundheitsfachberufe.de/wp-content/uploads/Strategiepapier-2018_11_08.pdf.

30.

Horak,

F. B.

(1991) Assumptions Underlying Motor Control for Neurologic Rehabilitation. In Contemporary management of motor control problems. Proceedings of the II STEP Conference, (pp. 11–27).

31.

Jette,

D. U.

, Bacon,

, Batty,

, Carlson,

, Ferland,

, Hemingway,

R. D.

, Hill,

J. C.

, Ogilvie,

, Volk,

(2003) Evidence-Based Practice: Beliefs, Attitudes, Knowledge, and Behaviors of Physical Therapists, Physical Therapy, 83(9), 786–805.

32.

Kaendler,

, Ritter,

, Sander,

, Elstner,

, Schwarzbach,

, Wagner,

, Meisel,

(2022) Positionspapier Schlaganfallnachsorge der Deutschen Schlaganfall-Gesellschaft – Teil 1: Nachsorge nach einem Schlaganfall: Status quo der Versorgungsrealität und Versorgungsdefizite in Deutschland [Position paper on stroke aftercare of the German Stroke Society-Part 1: long-term care after stroke: status quo of the reality and deficits of care in Germany], Der Nervenarzt, 93(4), 368–376.

33.

Kristensen,

H. K.

, Ytterberg,

, Jones,

D. L.

, Lund,

(2016) Research-based evidence in stroke rehabilitation: An investigation of its implementation by physiotherapists and occupational therapists, Disability and Rehabilitation, 38(26), 2564–2574.

34.

Lynch,

E. A.

, Chesworth,

B. M.

, Connell,

L. A.

(2018) Implementation-The Missing Link in the Research Translation Pipeline: Is It Any Wonder No One Ever Implements Evidence-Based Practice? Neurorehabilitation and Neural Repair, 32(9), 751–761.

35.

Lynch,

E. A.

, Connell,

L. A.

, Carvalho,

L. B.

, Bird,

M.-L.

(2021) Do clinical guidelines guide clinical practice in stroke rehabilitation? An international survey of health professionals, Disability and Rehabilitation, 44(15), 4118–4125.

36.

Mayston,

(2008) Bobath Concept: Bobath@50: Mid-life crisis–what of the future? Physiotherapy Research International, 13(3), 131–136.

37.

McCluskey,

, Vratsistas-Curto,

, Schurr,

(2013) Barriers and enablers to implementing multiple stroke guideline recommendations: A qualitative study, BMC Health Services Research, 13, 323.

38.

Mériaux-Kratochvila,

(2021) The academization of the health professions in Austria: facts and figures/Akademisierung der Gesundheitsberufe in Österreich: Zahlen und Fakten, International Journal of Health Professions, 8(1),, 141–145.

39.

Michie,

, Johnston,

, Abraham,

, Lawton,

, Parker,

, Walker,

(2005) Making psychological theory useful for implementing evidence based practice: A consensus approach. Quality & Safety in Health Care, 14(1), 26–33.

40.

O’Connell,

A. A.

(2006) Logistic regression models for ordinal response variables Sage University papers series Quantitative applications in the social sciences: Vol. 146. Sage Publications.

41.

Paanalahti,

, Berzina,

, Lundgren-Nilsson,

Å.

, Arndt,

, Sunnerhagen,

K. S.

(2018) Examination of the relevance of the ICF cores set for stroke by comparing with the Stroke Impact Scale, Disability and Rehabilitation, 41(5), 508–513.

42.

Pang,

M. Y. C.

, Eng,

J. J.

, Dawson,

A. S.

, McKay,

H. A.

, Harris,

J. E.

(2005) A community-based fitness and mobility exercise program for older adults with chronic stroke: A randomized, controlled trial, Journal of the American Geriatrics Society, 53(10), 1667–1674.

43.

Physiodeutschland Deutscher Verband für Physiotherapie. (2022), Zahlen, Daten, Fakten zur Physiotherapie. https://www.physio-deutschland.de/fileadmin/data/bund/Dateien_oeffentlich/Beruf_und_Bildung/Zahlen__Daten__Fakten/Zahlen_Daten_Fakten14.1.22.pdf.

44.

Rappold,

, Mathis-Edenhofer,

(2020) MTD-Personal prognose für Österreich bis 2030. https://jasmin.goeg.at/1789/1/MTD-Prognose_Endbericht_bf.pdf.

45.

Reiter,

, Fülöp,

, Pochobradsky,

, Röthlin,

, Stoppacher,

(2020) Rehabilitationsplan 2020. Wien. https://www.sozialversicherung.at/cdscontent/load?contentid=10008.742311&version=1611835415.

46.

Reuther,

(2011) Versorgungsbedarf und tatsächliche Versorgung neurologischer Patienten, Neuroreha, 3(03), 120–124.

47.

Rice,

D. B.

, McIntyre,

, Mirkowski,

, Janzen,

, Viana,

, Britt,

, Teasell,

(2017) Patient-Centered Goal Setting in a Hospital-Based Outpatient Stroke Rehabilitation Center, PM & R: the Journal of Injury, Function, and Rehabilitation, 9(9), 856–865.

48.

Rollnik,

J. D.

, Janosch,

(2010) Current trends in the length of stay in neurological early rehabilitation, Deutsches Aerzteblatt International, 107(16), 286–292.

49.

Rommel,

, Pruetz,

(2017) Inanspruchnahme physiotherapeutischer Leistungen in Deutschland, Journal Health Monitoring, 2(4), 101–108.

50.

Royal

Dutch Society for Physical Therapy.

(2014), KNGF Clinical Practice Guideline for Physical Therapy in patients with stroke. https://www.dsnr.nl/wp-content/uploads/2012/03/stroke_practice_guidelines_2014.pdf.

51.

Salbach,

N. M.

, Guilcher,

S. J.

, Jaglal,

S. B.

, Davis,

D. A.

(2010) Determinants of research use in clinical decision making among physical therapists providing services post-stroke: A cross-sectional study, Implementation Science, 5, 77.

52.

Salbach,

N. M.

, Jaglal,

S. B.

, Korner-Bitensky,

, Rappolt,

, Davis,

(2007) Practitioner and organizational barriers to evidence-based practice of physical therapists for people with stroke, Physical Therapy, 87(10), 1284–1303.

53.

Scheffler,

, Schimböck,

, Schöler,

, Rösner,

, Spallek,

, Kopkow,

(2022) Current physical therapy practice and implementation factors regarding the evidence-based ‘Rehabilitation of Mobility after Stroke (ReMoS)’ guideline in Germany: A cross-sectional online survey, BMC Neurology, 22(1), 284.

54.

Scott,

S. C.

, Goldberg,

M. S.

, Mayo,

N. E.

(1997) Statistical assessment of ordinal outcomes in comparative studies, Journal of Clinical Epidemiology, 50(1), 45–55.

55.

Silva,

T. M.

, Costa,

L. C. M.

, Costa,

L. O. P.

(2015) Evidence-Based Practice: A survey regarding behavior, knowledge, skills, resources, opinions and perceived barriers of Brazilian physical therapists from São Paulo state, Brazilian Journal of Physical Therapy, 19(4), 294–303.

56.

Teasell,

, Salbach,

N. M.

, Foley,

, Mountain,

, Cameron,

J. I.

, Jong,

, de, Acerra,

N. E.

, Bastasi,

, Carter,

S. L.

, Fung,

, Halabi,

M.-L.

, Iruthayarajah,

, Harris,

, Kim,

, Noland,

, Pooyania,

, Rochette,

, Stack,

B. D.

, Symcox,

,... Lindsay,

M. P.

(2020) Canadian Stroke Best Practice Recommendations: Rehabilitation, Recovery, and Community Participation following Stroke. Part One: Rehabilitation and Recovery Following Stroke; 6th Edition Update 2019, International Journal of Stroke, 15(7), 763–788.

57.

van Kessel,

, Hillier,

, English,

(2017) Physiotherapists’ attitudes toward circuit class therapy and 7 day per week therapy is influenced by normative beliefs, past experience, and perceived control: A qualitative study, Physiotherapy Theory and Practice, 33(11), 850–858.

58.

Verband der Erstatzkassen e.V. (Ed.). (2022) Preisvereinbarung und Vergütungslisten bei Heilmitteln. https://www.vdek.com/vertragspartner/heilmittel/preisvereinbarungen.html.

59.

Wissenschaftliche Dienste – Deutscher Bundestag. (2021) Zur Akademisierung der Physiotherapie, Logopädie und Ergotherapie. https://www.bundestag.de/resource/blob/876440/89106532b0796a8f70e18b2ad2ff62fc/WD-9-091-21-pdf-data.pdf.

60.

Züll,

(2015), Offene Fragen. https://www.gesis.org/fileadmin/upload/SDMwiki/Offene_Fragen_Zuell_08102015_1.1.pdf.