Participants
Eighty-four participants took part in this longitudinal study at the first measurement point (December 2011; t1). At that time, they were all students at a school of physiotherapy in Tübingen, Germany (PT Academy). Twenty-nine students were first-year students and 55 were advanced students (second and third year). At t1, 80 students (95%) were between 21 and 30 years old, only 4 students (5 %) were between 31 and 40. Fifty-eight were women, 26 were men. Forty-six physiotherapists (55% of those who had participated in the first measurement) filled in the questionnaire at the second measurement point (January 2015; t2). At that time, they were working in different places (multiple answers were possible): Six were working in a hospital (four in an emergency hospital, two in a rehabilitation center), 37 were employed in a physiotherapy practice, two were self-employed in a physiotherapy practice, and three were working in the sports sector. Two of the participants were not working as physiotherapists at t2. The following analysis is based on those 41 participants who filled in the questionnaire at both measurement points and were working as physiotherapists at the time of t2 (see Figure 1).
Procedure
The data collection at t1 took place at the PT Academy. After informing the students about privacy protection, their right to terminate their participation at any time without any disadvantage, and about the general purpose of the study, they gave written informed consent. Then they were asked to fill in the questionnaire (see below).
The follow-up measurement was conducted online three years later. All participants who were students at the PT Academy at t1 were contacted by e-mail from the secretary’s office of the PT Academy. They were asked to fill in an online questionnaire. Participation took about 15 minutes and was compensated with the opportunity to take part in a lottery, where participants could win ten vouchers worth 15 Euros each. The datasets were linked by a code that the participants had created individually following a fixed procedure (the first letter of their mother’s first name, the first letter of their father’s first name, and their mother’s year of birth).
Measurements
As described in the procedure of the cross-sectional analysis [29], the health-related EBs were measured with the Connotative Aspects of Epistemological Beliefs (CAEB) scale [30]. This scale was adapted to the domains of physiotherapy and medicine respectively. The participants assessed the knowledge in the field of physiotherapy and in medicine on seven-point semantic differential scales. Analogously to the cross-sectional study the CAEB-subscale texture was used, in which both domains were described with the same ten pairs of adjectives respectively. Sample items of this scale are “sorted – unsorted” or “absolute – relative”, whereby “unsorted” and “relative” represented sophisticated EBs (see Table 1).
Table 1 - The texture subscale of the Connotative Aspects of Epistemological Beliefs (CAEB) scale
Knowledge in the domain of physiotherapy [or medicine respectively] is …
|
1
|
objective
|
ò-ò-ò-ò-ò-ò-ò
|
subjective
|
2
|
confirmable
|
ò-ò-ò-ò-ò-ò-ò
|
unconfirmable
|
3
|
superficial
|
ò-ò-ò-ò-ò-ò-ò
|
profound
|
4
|
exact
|
ò-ò-ò-ò-ò-ò-ò
|
vague
|
5
|
absolute
|
ò-ò-ò-ò-ò-ò-ò
|
relative
|
6
|
sorted
|
ò-ò-ò-ò-ò-ò-ò
|
unsorted
|
7
|
precise
|
ò-ò-ò-ò-ò-ò-ò
|
imprecise
|
8
|
definite
|
ò-ò-ò-ò-ò-ò-ò
|
ambiguous
|
9r
|
negotiated
|
ò-ò-ò-ò-ò-ò-ò
|
discovered
|
10
|
structured
|
ò-ò-ò-ò-ò-ò-ò
|
unstructured
|
Seven-point semantic differential scales for measuring EBs about physiotherapy and medicine (texture subscale); reversely coded items are marked with a superscript ‘r’.
The bm and bps health concepts were measured following the procedure presented previously [17,29]. Participants rated the importance of five representative bm terms, such as “diagnosis”, and five characteristic bps concepts, such as “functionality”, on six-point Likert scales ranging from 1 (not important) to 6 (very important) (see Table 2).
Table 2 - Therapeutic health concepts scale
How important are the following subjects for your therapeutic thinking and acting?
|
1
|
Functionality
|
2*
|
Diagnosis
|
3*
|
Science
|
4*
|
Evidence-based methods
|
5
|
Limited activity of a patient
|
6*
|
Standardized tests
|
7
|
Limited participation of a patient (in the social environment)
|
8*
|
Medical guidelines
|
9
|
Mental health of a patient
|
10
|
Requirements of the patient’s everyday life
|
Six-point Likert scales for measuring the bm and bps concepts; bm items are marked with an asterisk (*).
Statistical analysis
Data analysis was performed using IBM SPSS 20.0 for Windows [31]. Cronbach’s Alpha was calculated to determine internal consistency for all scales. All data are reported as means (M) ± standard deviations and median values. For testing the hypotheses, paired sample t-tests were applied. Paired sample t-tests were also applied for comparing EBs regarding physiotherapy to EBs regarding medicine at t1 and t2. As it is discussed whether t-tests should be used for Likert scales [32], Wilcoxon signed-rank tests and Mann-Whitney U tests were also calculated, as nonparametric test equivalents to the t-tests.
The level of significance was set at P < 0.050. Cohen’s d scores were calculated as effect sizes of mean differences.
Ethics statement
This research was performed in accordance with the Declaration of Helsinki. The PT Academy’s administration provided ethical approval for the participation of its students (due to legal specifications, the school administration was responsible for checking and approving the participation of its students). Regarding the follow-up measurement, this study had full approval of the ethics committee of the Leibniz-Institut für Wissensmedien (approval number: LEK 2013/035). All participants took part voluntarily and anonymously.