Background
Osteoarthritis (OA) in the hip and knee is one of the main causes of disability, globally [
1]. The disease causes pain, stiffness, loss of function [
1] and decreased quality of life [
2]. Prevalence increases with age [
3,
4] and OA affects about 10% of men and 18% of women over the age of 60 [
4,
5]. However, many patients with OA in the hip and/or knee are of working age [
6]. With extended life expectancy, prevalence increases which leads to an increased burden of the disease for both the affected individual [
7], society and health care [
1]. In Sweden, estimations show that in 2032, 30% of adults over 45 years are expected to have OA [
8].
The evidence-based international non-pharmacological guidelines for treatment of OA include information, exercise and weight loss [
9,
10]. The Swedish national guidelines are in accordance with the international guidelines [
11,
12]. The guidelines highlight early diagnosis based on clinical examination and x-ray is only needed in certain circumstances such as if first line treatment has not been successful [
13,
14]. Swedish physiotherapists (PTs) are often primary assessors for patients with musculoskeletal disorders such as OA [
15]. This means that patients do not need a referral from a medical doctor (M.D.) to consult at PT [
15] and that PTs are qualified to diagnose OA.
In Sweden, patients with hip and/or knee OA are offered the opportunity to participate in a Supported OsteoArthritis Self-Management Programme (SOASP), which is generally led by a PT in primary health care (PHC) [
16]. The SOASP includes information and exercise and has been described elsewhere [
16] and is evaluated through a national quality register called Better Management of Patients with OA (BOA) [
16]. National quality registers contain data about patients concerning medical interventions, treatment and outcomes [
17]. The registers provide an opportunity to monitor, enable learning, improvement and research to ensure quality of the Swedish health care system [
17] and a vision is that the use of register data is integrated into daily clinical work [
17]. The intention of the BOA register is that all patients with OA should be offered treatment according to current guidelines [
18]. The BOA register contains patient reported outcome measures which enables systematic evaluation of the treatment for patients with OA and offers an opportunity for improvements of quality of care provided by PTs [
18].
Systematic reviews have shown that there is a divergence between evidence-based recommended care for OA and clinical practice [
19,
20] and less than 40% of patients with OA are offered the recommended first line treatment, i.e., education and exercise [
21]. In 2013, it was estimated that about 8.7% of patients 45 years and older seeking health care due to OA in Sweden also participated in a SOASP [
13]. A corresponding proportion for a region in southern Sweden, was almost the same (8.5%) [
13], which means that only a small number of patients with OA in the hip and/or knee were offered treatment in accordance with the guidelines.
In order for health care to meet future needs, regional guidelines were produced in a region in southern Sweden, based on the international and national guidelines, that also include a recommendation to report to the BOA register [
22]. According to the regional guidelines, PHC is primarily responsible for investigating, diagnosing, providing core treatment and additional treatment for patients with OA of the hip and/or knee [
22].
The regional guidelines were implemented in PHC in a region in southern Sweden between the years 2016 and 2019 with the intention to increase health care providers’ compliance with the guidelines, and furthermore, to make treatment more accessible for patients and increase the reporting rate to the BOA register [
22]. At the time of the implementation, there were about 150 PHC centres (tax-financed, public and private) in this region. All these PHC centres were given written information about the guidelines and about the implementation. Two coordinators with many years of clinical experience as PTs, of the SOASP and of reporting to the BOA register visited PHC centres to inform about the guidelines at workplace meetings where all categories of health care providers were present. The coordinators visited about 60% of the PHC centres at such workplace meetings between the years 2016 and 2019. Moreover, the coordinators offered the PTs individual support if needed and an education in how to organise a SOASP was provided for PTs on several occasions. In addition, an OA network for PTs responsible for offering the SOASP at their PHC centre was established to stimulate knowledge exchange, share experiences and news about OA treatment and from the BOA register. The OA network met twice a year during the implementation.
Implementation of new work procedures often takes several years [
23]. Facilitating factors are, for example, involving personnel early in the process and also receiving support from the management [
24]. Studies show that changes in working procedures are possible when implementation strategies are well planned and well performed [
25]. Using an implementation theory or framework when planning, realising and evaluating implementation is recommended [
26‐
28]. It is of great importance to evaluate an implementation in order to identify barriers and areas for improvement to increase knowledge for future implementation strategies [
29‐
33]. Therefore, the aim of the study was to explore PTs’ experiences of the regional guidelines for treatment of OA of the hip and/or knee and their experiences of the implementation of the guidelines in a region in southern Sweden.
Methods
An interview study was performed, analysed with qualitative content analysis (QCA).
Participants and recruitment
A purposive sampling was used. Forty-three PHC centres a region in southern Sweden that had been visited by a coordinator between 2017 and 2018 during the implementation were contacted by email. The heads of departments (HOD) were informed about the study and were asked for permission to contact the PT responsible for the SOASP at the PHC centre to participate in the study. In total, 20 HODs did not reply despite a reminder being sent. One PHC centre had no PT at the time for the study.
The HODs at 22 PHC centres gave their consent and an email was sent with written information about the study to the PT responsible for the SOASP at these PHC centres. Approximately a week later, two co-authors, AS and KS who are both experienced registered PTs and PhDs trained in qualitative research, contacted the PTs by email to set a date for the interview. All participating PTs were given written and verbal information about the study and gave their written informed consent to participate in the study. In total, 18 PTs were interviewed once. Four were not included in the study due to the PTs not answering (two), declining to participate (one) and being on parental leave (one).
The characteristics of the participating PTs are described in Table
1.
Table 1
Characteristics of the participants
Sex |
Male | 5 |
Female | 13 |
Work experience in PHCa, mean (SD) | 10.5 (7.2) |
Work experience at this PHCa centre, mean (SD) | 4.4 (2.6) |
Work experience with SOASPb, mean (SD) | 4.25 (2.5) |
Type of PHCa centre |
Public | 16 |
Private | 2 |
Education in providing SOASPb | 13 |
Data collection
Data were collected between February and September 2019. Eighteen semi-structured, individual interviews were conducted by AS (
n = 10) and KS (
n = 8), either at the participant’s workplace (
n = 17) or at the Health Sciences Centre (
n = 1) in Lund, Sweden, according to the participant’s preferences and with no one else present. All interviews were conducted in Swedish. An interview guide (see Additional file
1) was prepared, with inspiration from the framework Promoting Action on Research Implementation in Health Services (PARIHS) [
34‐
36]. The PARIHS suggests that implementation is most likely to be successful when evidence is viewed as aligning with professional and patient beliefs, health care context is receptive to implementation and mechanisms are in place to facilitate implementation [
37,
38].
The participants were interviewed once and the interviews lasted between 22 and 60 min, with a mean duration of 40 min (SD 13.2). All interviews were recorded using a tape recorder and were transcribed verbatim by the first author (KSÅ). There were no field notes made during or after the interviews. The transcripts were checked against the audio files twice. After the transcription, the record of participants was separated from the transcribed interviews ensuring anonymisation throughout the process of analysis.
Data analysis
The transcribed interviews were analysed using QCA, with an inductive approach as described by Graneheim and Lundman [
39,
40]. The method was suitable to gain a deeper understanding of PTs experiences and since it enabled us to explore opinions, ask open questions and to be able to approach data with focus on different and similar experiences. The transcripts as a whole were considered as units of analysis.
All 18 interviews were first re-read several times by three authors (AS, KS and KSÅ) to obtain an overview and an overall sense of the data. The text was then divided into meaning units, identified as “words, sentences or paragraphs containing aspects related to each other through content and context” [
39]. In order to provide coherence, three authors (AS, KS and KSÅ) analysed two of the interviews separately regarding meaning units and two authors (KS and KSÅ) analysed two additional interviews regarding meaning units. The separate analysis was followed by a discussion by three authors (AS, KS and KSÅ). The remaining 14 interviews were divided into meaning units by the first author (KSÅ). Then, the meaning units in all 18 interviews were condensed and labelled with codes that were sorted in subcategories based on similar manifest content by the first author (KSÅ). The sorting and labelling of subcategories was followed by a discussion by three authors (AS, KS and KSÅ) until consensus was reached which led to some subcategories being re-sorted and re-labelled. The analysis process went back and forth. Finally, after further abstraction, subcategories were sorted into categories reflecting their content. Three authors (AS, KS and KSÅ) were engaged in labelling the categories. All authors (AS, KS, EEH and KSÅ) were involved in discussing the results and the conclusion. The participants did not provide feedback on the findings. However, when closing each interview, the interviewers gave a summary of the interview, and the participants were able to modify. Examples of meaning units, condensed meaning units, codes, subcategories and categories are provided in Table
2.
Table 2
Examples of the analysis process with meaning unit, condensed meaning unit, code, subcategory and category
Then it is the younger ones, those who are still more active and simply are unable to leave their place of work (ID 1) | It is the younger ones who are more active and unable to leave their work | Younger are unable to leave their place of work | The Supported OsteoArthritis Self-Management Programme does not suit all patients with OA | The Supported OsteoArthritis Self-Management Programme is overall a well-functioning part of the regional guidelines but there is room for improvement |
I actually don’t really know what the nurses know about osteoarthritis. I mean when they meet patients and on the phone and what they might say to the patients (ID 2). | Don’t know what nurses know about osteoarthritis and what they say to patients | What do nurses know about osteoarthritis | Guidelines must be easy to follow | Management plays a key role when it comes to guideline compliance |
The meaning units, codes, subcategories, categories, and quotes were translated into English by the first author (KSÅ). Thereafter, a professional English reviser was consulted who edited the manuscript which was also proofread by a bilingual reviser (English and Swedish). The language editing was done in cooperation with the first author (KSÅ) so that the English translation remained close to the Swedish text in the interviews since this is a focus in QCA.
This study was conducted by four female PTs and researchers: one professor (EEH), one associate professor (KS), one PhD (AS) and one PhD student (KSÅ). Two researchers have on-going clinical practice (EEH, KSÅ) and two are experienced in interviewing and doing qualitative studies (KS, AS). The study was reported in accordance with the checklist of Consolidated Criteria for Reporting Qualitative Research (COREQ) [
41] (see Additional file
2). The software programme NVivo 12 was used in the analysis.
Discussion
In this interview study, we found that the SOASP was generally a well-established and functioning part of the regional guidelines and that the PTs saw a great need for the treatment. Moreover, the PTs were confident in their professional role and as a primary assessor for patients with OA in the hip and/or knee. However, the SOASP was not seen as appropriate for all patients with OA. Moreover, management support was considered important when it came to compliance with the guidelines, and to enable evaluation, development and sustaining knowledge among the health care personnel.
In our study, PTs described compliance with the guidelines on several points. The PTs expressed confidence in being the primary assessor and when clinically diagnosing patients with OA of the hip and knee. Our result is in line with another Swedish study where PTs in PHC centres self-reported to be confident in managing patients with OA regarding assessment, treatment and education [
42]. However, international studies have shown that PTs wanted x-rays to support a diagnosis [
43,
44]. Since an x-ray is not needed when diagnosing OA of the hip and knee according to international guidelines [
9‐
11], it is encouraging that the PTs in our study felt confident in clinically diagnosing OA.
The SOASP was first introduced in Sweden in 2008 [
45] and was generally already established in PHC. Furthermore, many PTs in Sweden have undertaken an education in order to offer the SOASP in accordance with the guidelines for the treatment of OA [
46]. In addition to knowledge about OA, the education provides the PTs with digital material to be used when offering the SOASP to patients [
46]. The majority of the participating PTs in our study had undertaken the education and this might be one reason why PTs clearly expressed confidence in assessing and diagnosing OA and to their compliance with the guidelines when it comes to offering the SOASP. One other study has shown that PTs in general show compliance with the guidelines for the treatment of OA in PHC when it comes to exercise and education [
42]. This is not surprising since there is evidence that exercise is beneficial for patients with OA and exercise is included in the guidelines [
11].
The PTs saw the SOASP as being efficient, useful, and beneficial for patients with OA in the hip and/or knee. Therefore, they might also be motivated to provide the treatment. Studies show that compliance with guidelines is positively affected by professional beliefs [
37,
38,
47].
According to the PTs, the SOASP was not appropriate for all patients with OA. Patients of younger age, with native languages other than Swedish and with lower levels of education were more difficult to reach within the SOASP. Studies have highlighted the importance for all patients to understand information in order to be empowered and be active participants in their own health care [
48,
49]. The PTs in our study emphasised the importance of providing equal care to all patients with OA but also the need to individualise treatment. Other studies have previously suggested a more person-centred approach in treatment [
49‐
51]. Criticism of guidelines is, for example, that they are not patient-centred but disease focused [
51,
52]. Furthermore, guidelines do not consider comorbidities which make them difficult to follow in clinical practice [
51]. Our result might indicate that the SOASP could be further developed to be more individualised.
The support from management was found to be important both regarding compliance with the guidelines for all health care personnel and also regarding knowledge, evaluation and development. Such findings on management support are not surprising or new but it rather confirms previous research [
32,
53‐
55]. Even so, it is important to study the ways in which management is important since this can differ between different contexts.
The PTs in our study experienced that there was often more attention on the production of care and budget at the PHC centre, rather than on the quality of care. This is in line with another Swedish study where physiotherapy managers experienced more focus on budgets and production of care than on evidence-based practice from higher management levels [
56]. In our study, the PTs wished for more management support to be able to evaluate and develop the work process. The PTs continued to report in the BOA register even though it was time consuming, and HODs did not ask for results. Having time to analyse and use knowledge based on local data reported in the BOA register could be a useful way to ensure quality of care for patients with OA of the hip and knee and would be in line with the vision for how the national quality registers could be integrated and used in the daily clinical workflow [
17].
According to the interviewed PTs, not all health care personnel followed the guidelines. Some patients were not referred to a PT, some came to the PT late in their disease development and after having an x-ray. This is in line with other studies showing that M.D.s were unaware of guidelines for OA [
47] and overused x-ray in diagnosing OA [
57]. Guidelines should be easy to follow [
47] and in an attempt to facilitate compliance and make it easier for M.D.s, flowcharts have been created [
58]. Since the PTs in our study already followed the guidelines, perhaps the implementation should have focused more on including other health care personnel, i.e., M.D.s and R.N.s and also HODs. This could possibly have a greater impact on providing evidence-based care for patients with OA in the long run.
Strengths and weaknesses of the study
Several steps were taken to ensure trustworthiness [
59]. A purposive sample was used i.e., PHC centres and participants were selected based on their experience of having had a coordinator at a workplace meeting informing all health care personnel about the regional guidelines. The sample of participants was geographically spread and included PTs with longer and shorter clinical work experience as PTs and experience of the SOASP. Quotes were presented in a way which prevented identification of the participants. Several researchers were involved in the analysis process and all authors were involved in discussing the results and the conclusion.
The first author, KSÅ, has experience from working as a PT in PHC and of working according to the guidelines. KSÅ was actively involved in the implementation as being one of the coordinators. However, the interviews were conducted by two co-authors that had not been involved in the implementation in order to reduce the risk of bias. Our intention with the study was to explore the PTs’ experiences of the regional guidelines for treatment of OA of the hip and/or knee and their experiences of the implementation of the guidelines with an inductive approach, not to evaluate an implementation process in relation to a specific framework. However, using the PARIHS framework as inspiration when preparing the interview guide ensured that relevant topics were illuminated during the interview, for example the PTs’ and patients’ beliefs about the guidelines and the PTs’ experiences of barriers and facilitators when working in accordance with the guidelines. Hence, our findings correspond to parts of the PARIHS framework as would be expected.
The description of the setting, recruitment, characteristics of the participants and process of analysis has been presented in detail both in the Method section and in Table
1 and Table
2 to enable evaluation of transferability as well as credibility.
We assessed that 20 participants would be sufficient to provide a variation in experiences and yet be a manageable amount of data. Eighteen interviews were conducted which we considered to be close to the desired sample.
A limitation with the study is that a pilot interview was not conducted. A pilot interview was planned but was cancelled. However, we believe this was compensated to some extent since potential changes in the interview guide were discussed by three of the authors after the first two interviews were conducted. No changes were made after these first interviews. Another limitation is the difference in the length of the interviews that might have affected the amount of data. Moreover, there were only two private PHC centres represented which somewhat reflects the fact that fewer private PHC centres invited the coordinators to their PHC centre during the implementation.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.