Determinants of guideline use in primary care physical therapy · Determinants of guideline use in...
22
Determinants of guideline use in primary care physical therapy Susanne Bernhardsson Department of Medical and Health Sciences Linköping University Närhälsan Primary Care Rehabilitation, Region Västra Götaland GIN 2013 San Francisco Supervisors/co-authors: K. Johansson 1 , P. Nilsen 1 , B. Öberg 1 , M. Larsson 2 1 Department of Medical and Health Sciences, Linköping University 2 The Sahlgrenska Academy, Gothenburg University - a cross-sectional survey of attitudes, knowledge, and behavior
Transcript of Determinants of guideline use in primary care physical therapy · Determinants of guideline use in...
Determinants of guideline use in primary care physical therapy
Susanne Bernhardsson
Department of Medical and Health Sciences Linköping University
Närhälsan Primary Care Rehabilitation, Region Västra Götaland
GIN 2013 San Francisco
Supervisors/co-authors: K. Johansson1, P. Nilsen1, B. Öberg1, M. Larsson2
1Department of Medical and Health Sciences, Linköping University 2The Sahlgrenska Academy, Gothenburg University
- a cross-sectional survey of attitudes, knowledge, and behavior
Vorführender
Präsentationsnotizen
First of all, I want to thank the organizers for letting me present this study. My name is…, I’m a PT from Gothenburg, Sweden. I will present to you a survey on attitudes, knowledge and behavior related to EBP, and particularly, clinical practice guidelines among PTs in PC in W Sweden.
Disclosure of interest (last 3 years) Susanne Bernhardsson
I certify that, to the best of my knowledge, no aspect of
my current personal or professional situation might reasonably be expected to affect significantly my views on the subject on which I am presenting, other than the following:
Am currently employed by the Region Västra Götaland as physical therapist/process manager rehabilitation guidelines
Am currently registered as PhD student at Linköping University, Dept of Medical Health and Sciences
Vorführender
Präsentationsnotizen
I declare that I have no financial COI. I am employed by the county council of Västra Götaland as a PT and process mgr for the dev and impl of rehabilitation g-lines I am also a PhD student at Link U.
Evidence-based?
Photo courtesy of the library at The Swedish School of Sport and Health Sciences
Vorführender
Präsentationsnotizen
Since many of you may not be familiar with physical therapy and its origins – Physical therapy actually has its roots in Sweden! this pic from the world’s first PT school, which was founded in Sthlm in 1813. This was long before the era of EBM and one might wonder how evidence-based the treatments were…
Evidence-based!
Vorführender
Präsentationsnotizen
But over the past 200 years, physical therapy practice has undergone quite a bit of change - Today, we know a lot more about which treatments are effective! These pictures are examples of exercises and other PT treatments for back, shoulder and neck pain that have been proven effective in rand controlled trials.
Cumulative number of randomized controlled trials, systematic reviews, and evidence-based clinical practice guidelines in physical therapy, by year. Maher 2008/PEDro
2013: 20,351 RCTs
Physical therapy research
Vorführender
Präsentationsnotizen
The development in physical therapy research has also been dramatic. The number of RCTs published in the past 25 yrs or so has grown tremendously, today over 20000. This of course, creates the challenge of bringing all this research out to the clinicians, and bridging the research-to-practice gap.
Background • Guidelines are an increasingly used means to bridge
the research-to-practice gap and to facilitate EBP Grol 1999
• Use of guidelines in physical therapy contributes to EBP Overmeer 2005, Liddle 2009
• … and yields both better treatment outcomes and reduced costs Fritz 2007, Rutten 2010
• ... but availability of and access to guidelines vary Liddle 2009, Jette 2003, Turner 1997
• … use/adherence to guidelines also varies Jette 2003, Turner 1997, Heiwe 2011, Bernhardsson 2013
Vorführender
Präsentationsnotizen
A brief background to our study: G-lines are becoming a common tool to summarize evidence and to bridge this gap, thereby facilitating an ev-based practice. Studies have shown that the use of g-lines in PT contribute to EBP… and that it can improve treatment outcomes and reduce costs. … but there is quite a large variation in availability of, access to, and use of g-lines So what determines, or influences, the use of g-lines? Why do some people use them and some don’t?
Study aims • To investigate self-reported attitudes, knowledge,
behavior, prerequisites, and barriers related to EBP and, in particular, guidelines among PTs in primary care in western Sweden
• To explore associations between these factors and the use of guidelines
A better understanding of these factors can form the basis for the development of a strategy for the implementation of guidelines in primary care physiotherapy
Vorführender
Präsentationsnotizen
To better understand what drives the use of g-lines, it’s important to investigate possible determining factors. So the aims of this study were… And to explore associations between these factors and the use of g-lines A better understanding of these factors can help to develop implementation strategies, to introduce g-lines to clinicians.
Methods, summary • Study design: Cross-sectional survey • Participants: 271 primary care PTs • Setting: Primary care in Region Västra Götaland • Outcomes: Attitudes, knowledge, behavior,
prerequisites and barriers related to EBP and, in particular, guidelines
• Data analysis: Descriptive statistics and logistic regression analysis
Vorführender
Präsentationsnotizen
This slide gives an overview of the methods used in the study. It was a c-s survey of 271 primary care PTs in Region V Götaland Outcomes were att’s, knowledge, behavior, prereq & barriers related to EBP and g-lines We collected the data w/ a web-based questionnaire and we analyzed associations w log regr analysis
Questionnaire development • Literature search
• Selection of questionnaire
• Forward and backward translation and cross-cultural adaptation
• Further development to suit the purposes of this study
• Test of content and face validity (n=10)
• Transfer to web/survey software
• Test of reliability (n=42)
Bernhardsson & Larsson, Phys Ther 2013
Vorführender
Präsentationsnotizen
We developed a questionnaire by searching the literature and selecting a previously used questionnaire on EBP, from the US, which we translated and adapted using rigorous methodology. We translated it forward and back again, modified it and further developed it by removing some items and adding items. We then tested the questionnaire for validity and reliability, revised it again, and determined the final version to have acceptable psychometric properties.
Questionnaire design
• 34 items in 4 sections: • Demographic variables
• Variables related to EBP
• Variables related to guidelines
• Treatment methods used for LBP, neck pain and shoulder pain
• 5-point or 3-point Likert scales
• Open and m/c questions
Vorführender
Präsentationsnotizen
The Q comprised 34 items in 4 sections – One on demographics, one on EBP, one on g-lines and one on treatment methods used Questions were answered on 5-p or 3-point Likert scales, most commonly fr strongly disagree to strongly agree, and there were also some open and m/c questions
And now the results: We sent the survey, via e-mail, to 400 PTs. 271 PTs responded – a fairly good response rate of 67.8%. ¾ of the respondents were women, ¼ men Age was pretty much normally distributed As to education, 2/3 had a 3-yr bach degree and 28% had an older degree that was shorter than 3 years. Only 5 % had a postgrad degree (incl Master, PhD student or doctor’s degree)
Attitudes to EBP
0
10
20
30
40
50
60
EBPnecessary to
practice
EBP helpsdecisionmaking
Want tolearn/improve
skills
Createsunreasonable
demands
Strongevidence islacking for
mostinterventions
Pe
rce
nta
ge
strongly disagree
disagree
neutral
agree
strongly agree
89% 82%
55%
90%
54%
Vorführender
Präsentationsnotizen
Attitudes to both EBP and g-lines were in general very positive: -A large majority (almost 90%) agreed (or strongly agreed) that EBP is necessary. -Abt 80% agreed that EBP is helpful in clinical decision making. 90% wanted to learn or improve their EBP skills. A smaller majority, a little over half, disagreed that EBP creates unreasonable demands and that there is a lack of strong evidence for most interventions. These numbers are pretty much in line w previous studies, both in Sweden and abroad – so it seems safe to say that PTs do feel very positive about EBP.
Attitudes/awareness/access to guidelines
• 96% considered guidelines important
• 33% were aware of relevant guidelines 61% were aware to some extent
• 13% knew where to find guidelines 65% knew to some extent
• 9% perceived easy access to guidelines at their place of work 51% had access to some extent, 37% did not have access
Vorführender
Präsentationsnotizen
As to guidelines - a large majority, almost all of the respondents, agreed that g-lines important… But only 1/3 were aware of g-lines relevant to their work (61% stated were somewhat aware). Ony 13% knew where to find g-lines (65% knew this to some extent) And as little as 9% perceived that they had easy access to relevant g-lines at their place of work (Another 50% stated that they had access to some extent but as many as 37% reported not having easy access)
Use of guidelines
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
Almost never Rarely Sometimes Frequently Almost always
47%
12%
Vorführender
Präsentationsnotizen
In spite of this rather poor awareness of and access to g-lines, almost half of the resp stated that they used g-lines frequently Another 41% used g-lines sometimes, 12% used them rarely
Comparison to other studies
• United States: 40%
• Australia: 45%
• The Netherlands: 61%
• Sweden, hospital-based: 75%
Fritz 2007, Iles 2006, Rutten 2009, Heiwe 2011
Vorführender
Präsentationsnotizen
The proportion reporting frequent use of guidelines is similar to levels of guideline use reported by PTs in the United States (40%) and in Australia (45%), but lower than that in the Netherlands (61%). Another Swedish study conducted in a hospital setting reported a higher proportion, 75%.
Barriers to using guidelines
0% 10% 20% 30% 40% 50% 60% 70%
Lack of time
Don't know where to find
Are too general
Take too long to read
No/too few guidelines exist
Too much "recipe"
Lack of support from colleagues
Lack of interest
Other
Vorführender
Präsentationsnotizen
The largest barriers to using g-lines were lack of time (reported by 2/3), not knowing where to find them (45%), that they are too general (40%), and that they take too long time to read (38%)
Associations between EBP and demographic variables
unreasonable demands (OR 0.3) • <5 yrs experience – confident to find relev. research
(OR 1.9) • Men – EBP helps in decision making (OR 2.0) • Postgrad degree – read articles (OR 6.4), search
databases (OR 20.6), confident to find relev. research (OR 11.1)
Vorführender
Präsentationsnotizen
In general we didn’t find many associations betw demographic factors and the outcome variables. We did find that those with the least exp and in the youngest age group had more pos attitudes. For example were those with <5 yrs exp 3 times more likely to consider EBP necessary and less likely to feel that it creates unreasonable demands. The only gender diff we found was that men were twice as likely to find EBP helpful in dec making The strongest association was that between having a postgrad degree and EBP behavior, where those with postgrad degrees were 6-20 times more likely to search databases, read sci articles and also to be confident to find relevant research. This relationship is also consistent with other studies.
Associations between guideline use and EBP variables
Important that guidelines exist Agree 7.87 (1.77–34.96) 0.001 0.06 268
Self-efficacy to treat according to evidence Agree 3.53 (2.00–6.22) <0.001 0.10 267
Self-efficacy to find research Agree 2.46 (1.47–4.11) 0.001 0.06 265
Knowledge how to integrate patient preferences Agree 7.89 (4.14–15.04) <0.001 0.22 269
Awareness that guidelines exist Yes 3.32 (1.95–5.65) <0.001 0.10 269
Knowledge where to find guidelines Yes 4.48 (1.94–10.31) <0.001 0.07 269
EBP is encouraged in the workplace Agree 1.86 (1.12–3.07) 0.015 0.03 268
Easy access to guidelines Yes 3.78 (1.45–9.84) 0.003 0.04 268
Guidelines are important to facilitate practice Agree 11.17 (3.32–37.57) <0.001 0.12 265
Guidelines are important to provide best treatment Agree 4.15 (1.36–12.68) 0.005 0.04 266
Guidelines are important to provide equal treatment Agree 3.25 (1.41–7.48) 0.003 0.04 266
Vorführender
Präsentationsnotizen
As for the use of g-lines, we found several associations. In a first step of uni-variate logistic regression analyses we found 15 variables to be associated with the use of g-lines, with OR:s between 1.9 and 11.2. Without spending much time on this slide – the variables most strongly associated with guideline use were Knowing hot to integrate pat prefs and considering g-lines important to facilitate practice.
Determinants of guideline use final multiple logistic regression model (n=258)
Independent Variable Level B (SE) Odds Ratio (95% CI)
P value
Guidelines are important to facilitate practice
Agree Disagree
2.31 (0.72) 10.11 (2.47–41.33) Reference
0.001
Knowledge how to integrate patient preferences Agree Disagree
1.72 (0.42) 5.58 (2.47–12.58) Reference
<0.001
EBP places unreasonable demands Disagree Agree
1.18 (0.34) 3.25 (1.68–6.28) Reference
<0.001
Awareness that guidelines exist Yes No
1.08 (0.35) 2.95 (1.49–5.86) Reference
0.002
Most treatments lack strong evidence Disagree Agree
1.06 (0.32) 2.89 (1.53–5.46) Reference
0.001
Model P =<0.001, Nagelkerke R2 =0.46, Overall percentage correctly predicted 75.6%
Vorführender
Präsentationsnotizen
Of the 15 variables, only 5 remained significant in the final multiple model, together explaining 46% of the variation in the outcome. The strongest associations remained the same - considering guidelines important to facilitate practice and knowing how to integrate patient preferences, with OR 10.1 and 5.6, respectively. This seems to indicate a perceived need for guidelines, and also a need for further knowledge about patient preferences, and how to integrate them in clinical decision making.
Implications for guideline developers and implementers • The positive attitudes found toward EBP and
guidelines are an important prerequisite for EBP in primary care physical therapy, but need to be accompanied by increased availability of and access to clinical guidelines.
• Identified barriers and facilitators can be used both in guideline development and to tailor a guideline implementation strategy.
Vorführender
Präsentationsnotizen
There are some implications for guideline implementers: - The pos att’s are imp prereq’s for EBP, but this study shows that it’s not enough. We also need to increase availability of g-lines, meaning produce more guidelines, and ensure that they are easily accessible. - And the identified barriers and facilitating factors could be used both in g-line development and to tailor a g-line impl strategy
Conclusions • Attitudes to EBP and guidelines were very positive
• Use of guidelines was not as frequent as could be expected in view of highly positive attitudes
• Awareness of and perceived access to guidelines were rather limited
• The identified barriers and facilitators should be addressed when developing guideline implementation strategies
Vorführender
Präsentationsnotizen
To conclude, the results of our study showed that attitudes to EBP & g-lines were very positive but that the use of g-lines was not as frequent as could be expected with such pos attitudes which is most likely explained by the low awareness of and access to guidelines And that the identified barriers & facilitators should be addressed when developing g-line impl strategies
Vorführender
Präsentationsnotizen
And with this non-evidence based exercise suggestion of what to do tonite while you wait for the bus or cable car – I thank you very much for yr attention!