Abstract
Background A positive attitude toward evidence-based practice (EBP) has been identified as an important factor in the effectiveness of the dissemination and implementation of EBP in real-world settings.
Objective The objectives of this study were: (1) to describe dimensions of Icelandic physical therapists' attitudes toward the adoption of new knowledge and EBP and (2) to explore the association between attitudes and selected personal and environmental factors.
Design This study was a cross-sectional, Web-based survey of the total population of full members of the Icelandic Physiotherapy Association.
Methods The Evidence-Based Practice Attitude Scale (EBPAS) was used to survey attitudes toward EBP; the total EBPAS and its 4 subscales (requirements, appeal, openness, and divergence) were included. Linear regression was used to explore the association between the EBPAS and selected background variables.
Results The response rate was 39.5% (N=211). The total EBPAS and all of its subscales reflected physical therapists' positive attitudes toward the adoption of new knowledge and EBP. Multivariable analysis revealed that being a woman was associated with more positive attitudes, as measured by the total EBPAS and the requirements, openness, and divergence subscales. Physical therapists with postprofessional education were more positive, as measured by the EBPAS openness subscale, and those working with at least 10 other physical therapists demonstrated more positive attitudes on the total EBPAS and the openness subscale.
Limitations Because this was a cross-sectional survey, no causal inferences can be made, and there may have been unmeasured confounding factors. Potential nonresponse bias limits generalizability.
Conclusions The results expand understanding of the phenomenon of attitudes toward EBP. They reveal potentially modifiable dimensions of attitudes and the associated characteristics of physical therapists and their work environments. The findings encourage investigation of the effectiveness of strategies aimed at influencing various dimensions of attitudes toward EBP.
Professional behavior in health care involves keeping up to date with current research and using the evidence-based practice (EBP) approach in the treatment of all clients.1,2 Evidence-based practice has evolved among health care professionals, in both scope and definition, since Sackett et al3 presented the concept of evidence-based medicine. According to the Sicily statement on EBP,4 this approach “requires that decisions about health care are based on the best available, current, valid and relevant evidence. These decisions should be made by those receiving care, informed by the tacit and explicit knowledge of those providing care, within the context of available resources.” Therefore, embracing EBP as a cornerstone of professional behavior requires that physical therapists be willing to continually search for, appraise, and apply new knowledge in an ever-changing context. Such a dynamic way of practicing physical therapy is challenging because it is based on the complex interaction and integration of multiple personal and environmental factors.
At the personal level, a positive attitude toward EBP has been identified as an important factor in the effectiveness of the dissemination and implementation of EBP in real-world settings.5–8 This importance is based on the assumption that attitudes make up the psychological framework on which a health care professional bases decisions of appraising and applying evidence in practice.4 In the literature, several contextual factors have been identified as potential facilitators of positive attitudes. Among these are being a woman,5 being a man,9 younger age,8,9 a higher level of academic education,10 less professional experience,8,9,11 facilities offering good access to literature,8 larger facilities and teamwork,5,12 and working 31 to 40 hours per week rather than more than 40 hours.5
Despite numerous reports on physical therapists' positive attitudes toward EBP,5,6,8–10,13–17 research results have indicated that research evidence is not used as frequently as could be expected.5,9,10,15 Researchers have provided a few potential explanations for this gap between attitudes and behavior, including social desirability related to the concept of EBP,9 ambiguities in interpretation of the concept of evidence or EBP,12 and a simplistic way of defining and measuring the concept of attitude.13 Aarons7 responded to this simplistic view of EBP by presenting evidence for 4 distinct, potentially important, and measurable dimensions of provider attitudes toward the adoption of EBP. According to Aarons,7 although a health care professional may report a generally positive view of EBP, he or she can concurrently have somewhat opposing attitudes. For example, a physical therapist can be positively inclined to one dimension of attitudes and negatively inclined to another dimension. These data indicate that new approaches may be needed to improve understanding of the complex phenomenon of attitudes and their underlying dimensions in relation to evidence-based physical therapy.
Therefore, the first objective of this study was to explore, in depth, physical therapists' attitudes toward changing practice and the adoption of EBP in the form of 4 concepts: the likelihood of adopting EBP given the requirement to do so, the intuitive appeal of EBP, openness to new practices, and the perceived divergence of usual practices from new and research-based interventions. The second objective was to explore the associations between these distinct dimensions of attitudes and selected personal characteristics of physical therapists (including sex, age, postprofessional education, and amount of professional experience) and the associations between these distinct dimensions of attitudes and the physical therapists' practice environments (defined as type of practice or facility, main content of work, percentage of working time in the practice, and number of physical therapists at the facility).
Method
Study Design
The study was a cross-sectional survey of Icelandic physical therapists.
Setting
This study took place in Iceland, where health care is publicly funded, meaning that residents are insured by the state with equal access for everyone. In 2013, there were approximately 550 licensed physical therapists in the country, and the majority worked in health care organizations such as hospitals, rehabilitation centers, and nursing homes or in the private sector. Physical therapists in Iceland are autonomous practitioners to whom patients or clients have direct access, affording an entry point into the health care system.
In Iceland, the Department of Physical Therapy at the University of Iceland is the only one to graduate physical therapists. The standards for the Icelandic program, including an emphasis on EBP, follow the standards for higher education institutions in Europe18 and guidelines from the World Confederation for Physical Therapy19,20; influence from the Commission on Accreditation in Physical Therapy Education standards in the United States21 also is apparent.
Participants and Procedure
Physical therapists eligible to participate in this study were all licensed and full members of the Icelandic Physiotherapy Association (IPA). Additional criteria for inclusion were a valid email account and an understanding of Icelandic. A list of members was generated by the IPA and submitted to the researchers. The final sample consisted of 534 members of the IPA.
An email was sent in May 2013 to each of the eligible 534 IPA members with a request to participate in the study by answering a Web-based survey. The email included a link to a SurveyMonkey online survey tool (SurveyMonkey, Palo Alto, California) administering the standardized Evidence-Based Practice Attitude Scale (EBPAS)7 and background questions on personal characteristics and practice environments. Two reminder notices were emailed to nonrespondents, 1 week apart. The respondents were not reimbursed for their participation in the study.
EBPAS Measurement of Dependent Variables
A Web-based Icelandic version of the EBPAS was used to measure the dependent variables in this study. The EBPAS is a standardized questionnaire that was developed in 2004 by Aarons7 to assess mental health providers' global attitudes toward the adoption of innovation and EBP. Since its development, EBPAS has been applied in many studies both in the United States and internationally and, recently, in research on allied health care professionals, including physical therapists.22 The EBPAS was designed to explore, in depth, more complex dimensions of attitudes than prior research had done and to tap into health care providers' attitudes that are likely to be related to elements of practice and that may facilitate or hinder the adoption of EBP.7 Furthermore, Aarons7 included standard descriptive terms, such as research-, clinical guideline–, or manual-based approaches, rather than the phrase “evidence-based practice.” These descriptive terms are central to the EBP approach, and using them was expected to decrease biases engendered by ambiguities or ideas of social desirability related to the concept of EBP. Additionally, the questionnaire requires little of a respondent because it takes only 3 to 5 minutes to complete.
The EBPAS contains 15 questions used to identify 4 dimensions of health care providers' attitudes toward the adoption of EBP.7 The response to each of the 15 questions is rated on a 5-point scale ranging from 0 (“not at all”) to 4 (“to a very great extent”). The following EBPAS scores were computed as the mean of corresponding item ratings: the requirements subscale contains 3 items that assess the likelihood of adopting EBP given a requirement to do so; the appeal subscale contains 4 items that measure the intuitive appeal of EBP; the openness subscale contains 4 items that measure openness to new practices; and the divergence subscale contains 4 items that assess perceived conflicts between research-based interventions and current practice. In addition, a total (mean) score was computed for the 15 items in the total scale. A higher score indicates a more positive attitude toward EBP on all of the scales, except for the divergence subscale, on which a lower score indicates a more positive attitude. Therefore, calculation of the mean score for the total EBPAS included reverse scoring for the divergence subscale. In cases of missing data, calculations were carried out to allow for one fewer item than make up the scale.
The face validity and content validity of the EBPAS have been described, and its 4-factor structure (subscales) has been supported in a few studies involving researchers, clinicians, and mental health service providers.7,23,24 These studies reported good reliability of internal consistency, with Cronbach alpha values ranging from high to moderate for the requirements (α=.90–.93), appeal (α=.74–.80), openness (α=.78–.84), and divergence (α=.59–.66) subscales and for the total EBPAS (α=.76–.79).7,23,24
Translation and back-translation25 were used to produce a standardized Icelandic version of the EBPAS. To our knowledge, this study is the first to apply the Icelandic version of the EBPAS with a sample of physical therapists. Therefore, we included testing of the reliability of internal consistency and the factor structure of the EBPAS with our sample. Cronbach alpha values ranged from high to moderate for the requirements (α=.93), appeal (α=.68), openness (α=.80), and divergence (α=.68) subscales and for the total EBPAS (α=.84). Factor analysis with oblique rotation was performed on the 15 questions, and the results supported the 4-factor structure of the EBPAS.
Personal Characteristics and Practice Environments (Independent Variables)
The physical therapists self-reported their personal characteristics and practice environments. The personal characteristics of the therapists included sex, age, postprofessional education, and professional experience. The practice environment variables included type of practice or facility, main content of work, percentage of working time in the practice, and number of physical therapists at the facility.
Ethics
All questionnaires were filled out anonymously. The study was registered at the Icelandic Data Protection Authority.
Data Analysis
All analyses were performed with IBM SPSS Statistics for Windows, version 22.0 (IBM Corp, Armonk, New York). Because of the exploratory nature of this study, there was no power calculation before it began; to avoid underpowering of our results, we selected a relatively conservative ratio of 26 participants per independent variable. No corrections were made for multiple statistical tests, and significance was set at a P value of less than or equal to .05 for all regression calculations.
The characteristics of the physical therapists and their practice environments were described by analyzing central tendencies, distributions, and frequencies. The EBPAS scores were assumed to be reasonably close to real numbers so that they could be considered for parametric statistics.7 Because linear regression is a powerful method for detecting multivariable associations, it was used on the EBPAS data after creation and inspection of residual plots to verify assumptions of normality and homoscedasticity. Moreover, before regression analyses, the dependent variables (total EBPAS and subscales) were screened to confirm that they met the assumptions for linear regression. Multicollinearity was examined by calculating and examining variance inflation factors (VIF) and tolerance for each independent variable. A mean VIF close to 1 represents low collinearity, whereas a mean VIF of 10 or greater is very poor and reflects high collinearity.26 In our analyses, a VIF of 15.2 revealed a problem with 2 variables: professional experience and age. This problem was resolved by dichotomizing the professional experience variable, after which all VIF values were less than 2.0. For 6 of the independent variables, categories were collapsed (variables dichotomized) to build theoretically sound and stable regression models. The age variable, however, was kept in its original, continuous form.
Linear regression was used to explore the univariate association between the EBPAS (total EBPAS and each subscale) (dependent variables) and the 8 independent variables. Then, we used a full multivariable linear regression model to determine the independent association of all respondent and practice characteristics with the EBPAS (total EBPAS and each subscale) (using the enter method), while simultaneously controlling for the potential confounding effects of other variables in the model. All analyses focused on associations within the total sample, and interaction effects were not included in the regression models. In the multivariable linear regression, the nonstandardized coefficient, t statistic, and P value were calculated for each independent variable, and R2, F, and P values were calculated for each model.
Results
Participants and Their Practice Environments
Of the 534 eligible IPA members who were invited to participate, 211 members replied (response rate=39.5%). Table 1 shows respondents' personal and practice characteristics. The respondents' average age was 43.4 years, and 76% were women. On average, they had 17.5 years of professional experience, 56.1% had practiced for more than 15 years, and 24.6% had an advanced postprofessional degree (MS, MA, or PhD). More than half of the respondents worked full time, and the majority described themselves as clinicians with direct treatment as the main content of their work. The most frequently cited type of facility was a private outpatient practice, and the average number of physical therapists at each facility was 10.4.
Respondents' Personal Characteristics and Practice Environments
EBPAS
Table 2 shows a description of the scores on the total EBPAS, subscales, and all 15 items. Scores on all scales and items indicated relatively positive attitudes toward change in practice and EBP. Table 3 shows univariate associations between the EBPAS and 7 (dichotomous independent variables) of the 8 respondents' personal characteristics and practice environments. The only continuous independent variable, age, was not significantly associated with the EBPAS (data not shown).
Scores on Total Evidence-Based Practice Attitude Scale (EBPAS), 4 Subscales, and 15 Items
Univariate Associations Between Evidence-Based Practice Attitude Scale (EBPAS) and Respondents' Personal Characteristics and Practice Environmentsa
Table 4 shows the results of 5 separate multivariable linear regression analyses. Sex, postprofessional education, and the number of physical therapist coworkers were the independent variables that were significantly associated with some dimensions of attitudes toward EBP. Compared with men, women had more positive attitudes, as measured by the total EBPAS and the requirements, openness, and divergence subscales. Therapists who had completed postprofessional education (diploma, MS, MA, or PhD) scored higher on the openness subscale than did those with no postprofessional education. Working in a group of at least 11 physical therapists was associated with more positive attitudes, as measured by the total and openness subscales.
Multivariable Linear Regression Analyses of Independent Variables Associated With Evidence-Based Practice Attitude Scale (EBPAS)a
Three of the multivariable regression models explained a significant proportion of the variance in the EBPAS scores (Tab. 3). These were the models built to explain the variance in scores for the total EBPAS (14.3%, F8,164=3.408, P=.001), the openness subscale (14.5%, F8,167=3.544, P=.001), and the divergence subscale (12.4%, F8,164=2.90, P=.005). Only 6.5% of the variance was explained by the model for the requirements subscale, and only 6.7% was explained by the model for the appeal subscale (F8,167=1.44, P=.182, and F8,167=1.48, P=.164, respectively).
Discussion
Main Findings
The physical therapists who participated in the present study had positive attitudes toward the adoption of new knowledge and EBP. Being a woman was independently associated with more positive attitudes, as measured by the total EBPAS and the requirements, openness, and divergence subscales. Physical therapists with postprofessional education had more positive attitudes, as measured by the openness subscale, and working with at least 10 other physical therapists was associated with more positive attitudes, as measured by the total EBPAS and the openness subscale.
Attitudes Toward EBP
By selecting the EBPAS questionnaire, we were able to obtain a more comprehensive assessment of physical therapists' attitudes toward EBP than have most studies. The existing measures of EBP have been criticized for including a limited number of items for the assessment of attitudes and for lacking a theoretical framework.27 The EBPAS is based on theoretical models and captures dimensions of attitudes that are likely to be important for understanding the dissemination and implementation of EBP.7 Our examination of more complex dimensions of attitudes confirmed and expanded the results of other studies on physical therapists' attitudes toward EBP. The physical therapists in our study continue to report positive attitudes toward new knowledge and EBP.5,8,9,13,14,28 Moreover, they showed more favorable scores on all parts of the EBPAS than did the mental health providers in the original EBPAS study in the United States,7 medical doctors in later research in Greece,27 and allied health care professionals in Australian research.22 In this comparison, it is important to recognize that health care disciplines may have different needs and experience different barriers in the application of EBP. More research is needed to learn about the potential effects of professional backgrounds on willingness to adopt EBP.
Interestingly, the use of EBPAS may minimize the much-discussed effects of the potential social desirability of the EBP concept. The positive attitudes toward EBP in the present study were less likely to have been based on the social appeal of EBP, because no question in the EBPAS questionnaire included the phrase “evidence-based practice.” Furthermore, the present study demonstrated how different dimensions of attitudes are associated with the selection of the personal characteristics of physical therapists and their practice environments.
Association Between Personal Characteristics and Attitudes
The results of the present study demonstrated differences in attitudes toward EBP between the sexes. Women reported more positive attitudes than did men on the total EBPAS and the requirements, openness, and divergence subscales. Other studies have examined differences in attitudes and beliefs regarding EBP between the sexes. In a small study of attitudes among workers in community-based mental health organizations, women scored higher on both the total EBPAS and the openness subscale.29 A study of physical therapists providing service to adults with stroke revealed that women had more positive attitudes toward EBP than did men.5 In contrast, Bernhardsson et al9 demonstrated that men were more likely than women to consider EBP helpful in decision making, indicating more positive attitudes for men. Although we expect that male and female physical therapists receive the same education and encouragement to use evidence-based practice, our results stress the importance of further research in this area.
Respondents with more postprofessional education were more open to new practices in the field than were those with no postprofessional education. This finding should not be a surprise because graduate university programs should be expected to emphasize research and critical appraisal skills to a greater extent than should undergraduate programs, both of which form the basis for the implementation and adoption of EBP. Therefore, physical therapists with postprofessional education should have the skills to critically evaluate innovations and accept or reject them accordingly. A systematic review of barriers, enablers, and interventions of EBP in physical therapy included studies reporting a positive association between a higher academic level and knowledge of EBP, more positive attitudes, and fewer perceived barriers toward the use and adoption of EBP.10
Although professional experience was not significant in any of the multivariable models, the univariate analysis revealed that having 1 to 5 years of professional experience was associated with more positive attitudes on the total EBPAS, a result consistent with those of previous studies.5,8,9,13 In recent years, entry-level education in physical therapy in Iceland has highlighted the importance of EBP, and students learn various EBP skills. Therapists with even a small amount of experience, therefore, may be more confident about EBP concepts and critical appraisal than those who graduated earlier.
Association Between Practice Environments and Attitudes
Physical therapists who work in a group of at least 11 physical therapists were more positive, as measured by the total EBPAS and openness subscale, than were those who worked in small clinics. Salbach et al5 found that small institutions were disadvantaged in implementing EBP because of a lack of human resources. They and other researchers28 also found that a lack of peer support is an organizational barrier to EBP. Physical therapists have expressed how they see their colleagues as important in extending experience beyond their own and how the opportunity to converse with colleagues is especially valued by those who work in seclusion.12 In smaller clinics, social interactions may be lacking, with diminished opportunities to share tools, routines, and stories and to negotiate meaning related to the complexities of the real-life application of evidence in practice.30 Physical therapists working in smaller settings therefore may have the greatest need for help in developing positive attitudes toward EBP and in implementing the approach.5
The type of facility was not significant in any of the multivariable models, yet there was a univariate association between working in a private outpatient practice and more negative attitudes toward applying EBP if required to do so. In contrast, in another univariate comparison, physical therapists in an Australian private outpatient practice and hospitals rated their EBP skills equally and practiced EBP with equal frequency.14 In Iceland, a high percentage of physical therapists work in private outpatient practices (>50% of the respondents in the present study). Further research is needed to explore whether the culture and infrastructure of outpatient clinics are less supportive of clinicians' effort to implement EBP and will eventually lead to more negative attitudes or whether those who are less willing to follow professional guidelines and requirements prefer to be self-employed.
Implications for Practice
The findings of the present study should be of interest for physical therapist clinicians, administrators, and educators. Knowledge about various dimensions of attitudes toward EBP provides a potentially important foundation for design interventions to implement EBP in physical therapy. If attitudes form an individual's psychological framework for building decisions about applying evidence in practice, as Dawes et al4 proposed, our results support the conclusion that physical therapists have the background to apply evidence in practice. However, implementation of EBP in practice must account for the complex dimensions of attitudes, dynamic influences, and interplay between the individual and the practice setting.
We support the idea that strategies directed at the individual physical therapist's skills and attitudes toward the use of research must be considered in a wider context of influences on clinical behavior.31 When designing interventions to implement EBP in practice, managers must be aware of potential differences in attitudes between men and women. Our results imply that men may need strategies aimed at affecting their attitudes toward EBP.
Additionally, interventions may include raising the educational level of physical therapists as well as increasing both the practical value of research and access to databases with peer-reviewed publications. It is important to create a research-friendly climate involving cooperation with colleagues and academic programs that are available to support clinicians in their understanding and implementation of research results.10,12
It is also important to recognize that physical therapists working in different settings and with different backgrounds may have different needs and encounter different barriers to applying EBP.10 Small practices may need to seek help from colleagues in larger institutions and academic settings to develop positive attitudes to EBP and to ensure successful implementation. Another solution is for small practices and geographically dispersed physical therapists to engage in a joint online course to further EBP.
Strengths and Limitations
The results of the present study should be interpreted with several limitations in mind. This was a cross-sectional observational study; therefore, no causal inferences can be made, and there may have been unmeasured confounding factors. The data were based on self-reports, which are inherently subjective and rooted in individual personality, outlook, and context. The population approach and the use of a Web-based survey without any reimbursement may explain the response rate of only 39.5%. This response rate makes the study prone to nonresponse bias, with respondents potentially holding more positive views toward research than the general population of Icelandic physical therapists. Yet, the association between attitudes and background variables should be less influenced by the response rate. Furthermore, our response rate is above the 33% average for surveys delivered via email and reflects meta-analysis results showing that among professionals, email surveys generally have a 23% lower response rate than mail surveys.32
Despite the low response rate, the generalizability of our findings is strengthened by the fact that the age and sex distributions of the respondents were comparable to those of the total population of IPA members. According to the IPA, in the first quarter of 2013, the mean age of full members was 45.6 years (SD=10.4), and 74.1% were women.
Our sample included only physical therapists who were licensed in Iceland and who understood Icelandic. Therefore, generalization to other groups must be made by logical inference on the basis of close similarities between physical therapist education and practice in Iceland and, for example, northern Europe, North America, and Australia. With no interaction terms included in the regression models, the explanatory power of the present study was restricted to overall (main) effects of the 8 independent variables in the model. A larger and more powerful study is needed to analyze subgroups and various interaction terms, along with additional variables, such as information on workplace and organizational culture. Finally, because no corrections were made for multiple statistical tests, the significance of the findings may have been inflated.
The strengths of the present study include the use of a short standardized questionnaire with established psychometric properties and the potential to improve understanding of the complex phenomenon of attitudes and to minimize social desirability bias related to EBP. Moreover, the factor structure of the EBPAS was confirmed in our sample of physical therapists. The Web-based questionnaire was time and cost efficient and easy to administer. These benefits can facilitate replication of the present study in other contexts. Finally, there is convincing evidence that the EBPAS can be applied in the health care sector in general7,23,24; therefore, the use of the EBPAS can facilitate cross-disciplinary comparisons.
The present study confirms that physical therapists have generally positive attitudes toward changing practice and the adoption of EBP. The results expand understanding of the phenomenon of attitudes toward EBP and reveal an association between dimensions of attitudes and personal factors. These findings may help educators and managers understand which types of physical therapists are more likely to implement evidence in practice; for example, women and, especially, physical therapists with postprofessional education may be able to lead the implementation of EBP. Moreover, the findings may help managers understand how to create an environment facilitating EBP. Different practice settings may require different strategies for the implementation of EBP. Collaboration between academic settings and institutions can be beneficial, especially for small practice settings.
The results should be used in continuing research on personal and environmental factors that foster a dynamic way of practicing evidence-based physical therapy. Finally, the findings encourage investigation of the effectiveness of strategies aimed at influencing various dimensions of attitudes toward EBP in physical therapy curricula and continuing education courses for physical therapists.
Footnotes
Both authors provided concept/idea/research design, writing, data collection and analysis, project management, participants, facilities/equipment, and institutional liaisons.
The authors express their gratitude to the physical therapists who responded to the survey and to Dr Thorarinn Sveinsson for advice and assistance in data analysis.
The study was registered at the Icelandic Data Protection Authority (No. S6203).
- Received July 24, 2015.
- Accepted May 5, 2016.
- © 2016 American Physical Therapy Association