Background
There is convincing evidence supporting the role of physical activity in the prevention and treatment of non-communicable diseases. Globally, one in four adults report physical activity below recommended levels, and the proportion is even higher in Sweden: one in three adults [
1]. Numerous barriers have been identified for adult participation in physical activity, such as lack of social support [
2,
3], insufficient motivation and unfavourable health status [
4‐
6]. Physical activity on prescription (PAP) in routine health care has been implemented in several countries, including Sweden, to support patients who might benefit from increased physical activity [
7]. These programmes vary with regard to design, e.g. programme duration, reason for prescription and patient payments. Varying degrees of effectiveness in influencing patients’ physical activity have been observed [
7]. It is important that PAP programmes enhance recipients’ self-efficacy to participate in physical activity [
8] and address barriers to participation [
9,
10].
The Swedish PAP concept (FaR®) consists of motivational person-centred counselling, along with individualized written prescriptions of home-based activities (e.g. walking) and/or facility-based activities in local sports organizations and follow-up by health care professionals [
11]. Some health care organizations in Sweden have implemented multi-professional programmes for delivery of PAP in routine health care involving physical activity counsellors [
11]. Such programmes have been proposed as a possible means to facilitate delivery of PAP in routine health care and to improve the effectiveness of PAP to increase patients’ physical activity levels [
4,
12].
The Swedish PAP concept has been favourably evaluated in terms of the effect on patients’ change in physical activity levels and quality of life [
13‐
15], adherence to prescription [
16,
17] and the characteristics of prescribed patients and prescribers [
18]. Further, experiences of PAP among patients with chronic musculoskeletal pain have been described [
4]. However, the concept has not been investigated from a patient perspective to gain an understanding of how routine health care delivery of PAP and counsellor support may influence the recipients’ beliefs, attitudes and motivation to achieve increased physical activity. In general, there is a paucity of qualitative research on routine health care delivery of PAP from the patient perspective, including programmes that include access to counsellor support. Such research could potentially yield important insights into PAP recipients’ paths to behaviour change, i.e. increased physical activity. This knowledge is highly relevant for the development of improved PAP programmes. To address this gap in research on PAP, this study investigates the experiences of PAP recipients in a routine health care PAP programme in Sweden that offers the recipients support from physical activity counsellors. The aim was to explore influences on PAP recipients’ engagement in physical activity from a long-term perspective.
Methods
Study design
A qualitative explorative design was used, operationalized through individual semi-structured interviews with a purposively selected sample of PAP recipients. A combination of conventional (inductive) and directed (deductive) approaches to qualitative content analysis was used [
19,
20].
Study setting
Interviews were conducted with PAP recipients living in the County of Kronoberg, a small rural county in the southern part of Sweden, with approximately 190,000 inhabitants. The health care organization in this county is managed by the County Council of Kronoberg and consists of 22 public and 11 privately operated primary health care centres (all publicly funded) and secondary care provided in two public hospitals.
The Swedish welfare system is largely funded by taxes and provides equal access to health care for everyone, with elderly care and social services based on each person’s need for support. The system is characterized by shared welfare responsibility between 21 county councils and 290 municipalities. County councils provide primary and secondary health care, whereas municipalities are responsible for social services and providing health care for older people living at home or in nursing homes.
In 2009–2010, the County Council of Kronoberg implemented a multi-professional programme for health care delivery of PAP in both primary and secondary care. The PAP programme features the key elements of the Swedish PAP concept [
11]. The prescriptions are issued by licensed health care professionals, e.g. physicians, nurses and psychologists, during routine health care visits. In conjunction with issuing the PAP, the prescriber provides written information about the physical activity counsellors. There is no formal referral system arranged between prescribers and counsellors. PAP recipients who want this support contact a counsellor by e-mail or phone. This counselling support is free of charge and can be utilized for 1 year after the PAP is issued. The counsellors are licensed health care professionals (registered nurses and physiotherapists) who are trained in motivational interviewing (MI) counselling techniques. They are familiar with local facility-based activities. Any fees associated with participation in activities are paid for by the PAP recipients.
Participants and recruitment
The participants in this study were purposively selected from PAP recipients in routine care between June 2013 and June 2014. The sample was selected to represent various age groups (18–29, 30–44, 45–64 and 65+ years), sex and municipality, prescribing health care setting and unit, and use of counsellor support (users and non-users). Criteria for participation were Swedish-speaking, willingness to share experiences with the researcher and having received a PAP 1.5–2.5 years earlier.
In total, 75 recipients were invited, 68 by postal invitation and 7 by telephone. Telephone invitations were used with the aim of reaching a younger and more geographically diverse sample. All participants received both written and verbal information about the study. A heterogeneous sample of 13 of the 75 invited recipients agreed to participate (Table
1).
Table 1
Characteristics of the PAP recipients who were interviewed
Sex |
Female | 9 |
Male | 4 |
Age |
30–44 years | 1 |
45–64 years | 3 |
65+ | 9 |
Prescribing setting |
Primary care | 11 |
Secondary care | 2 |
Visits to PAP counsellor |
Yes | 6 |
No | 7 |
Number of inhabitants in the municipality of residence |
< 15,000 | 3 |
15,000–30,000 | 5 |
> 80,000 | 5 |
Data collection
Data were collected from April 2015 to January 2017. A semi-structured interview guide was developed by the researchers in order to give the interviewees the opportunity and freedom to express their individual views on their engagement in physical activity. All questions were generated by the researchers informed by previous research about PAP and barriers and facilitators to physical activity behaviour.
The interview guide consisted of four main areas of questions: prescription and counselling; start of activity; maintenance of activity; and perceived benefits of PAP (see Additional file
1). A picture showing the topics was visible to the participant and the interviewer during the interviews. The interviews started with a key question for each main area followed by supplementary questions when necessary. Descriptive questions were used [
21], such as “Could you tell me about the consultation when you received PAP” followed by probing questions, e.g. “Can you explain a little further?”, and/or clarifying questions, e.g. “What advice did you receive from the prescriber?” At the end of each interview, the interviewer asked if there was anything that had not been elucidated.
The interview guide was tested in one interview to determine whether the questions were suitable for obtaining rich answers. The questions were found to be clear and informative. Consequently, no revision of the interview guide was done and the test interview was included in the analysis.
All interviews were conducted by the first author (PA), either in the participant’s home or another location selected by the interviewee. A maximum of 60 min was allocated for each interview. The interviews were recorded and transcribed verbatim, mainly by the interviewer (PA). Two were transcribed by an assistant. PA listened and read through the first version of all transcriptions and made corrections where necessary.
Theoretical framework
The interview questions were informed by the theoretical framework Capability-Opportunity-Motivation-Behaviour (COM-B) [
22]. COM-B was also used as a framework to analyse the data by means of categorizing the factors (influences on the behaviour) into the three behaviour change determinants, i.e. Capability-Opportunity-Motivation, in the deductive analysis.
The COM-B framework was developed with reference to existing theories of behaviour and a US consensus meeting of behavioural theorists, which considered the prerequisites for enacting various behaviours [
22]. The framework is intended to be comprehensive, parsimonious and applicable to all behaviours. COM-B has been widely applied to many different types of behaviours, including studies of audiologists’ behavioural planning [
23], adherence to swallowing exercise [
24] and promoters and barriers to being vaccinated and taking antiviral drugs [
25].
Capability is defined in COM-B as an individual’s psychological and physical capacity to engage in the behaviour concerned, encompassing having the necessary knowledge, ability and skills. Opportunity is defined as all the factors that lie outside the individuals that make the behaviour possible or prompt it. Motivation or willingness to enact behaviour is defined as all those brain processes that energize and direct individuals’ behaviour. Motivation includes habitual processes, emotional responding, as well as conscious, deliberate decision making [
22]. Behaviour in this study was defined as undertaking and maintaining some form of physical activity.
Data analysis
The analysis was grounded in qualitative content analysis according to Krippendorf [
20]. The interview transcripts were read through several times by PA and LL to obtain a sense of the whole. The text was then divided into meaning units (paragraphs) in relation to the aim (by PA and LL). The meaning units were condensed, with the purpose of reducing the text, but still preserving the core. The meaning units were then abstracted and labelled with a code. The codes were compared based on differences and similarities, and sorted into sub-categories, structured according to the predefined categories, i.e. the determinants in COM-B (by PA, LL and PN).
In the next step, the highlighted text was read, coded and categorized separately by PA, LL and PN. The text was analysed individually by the three authors to ensure credibility, using a structured process to code and categorize the data according to the COM-B framework. Then, the authors discussed the interpretation of the data in relation to COM-B and compared their coding. Discussions in the group continued until no inconsistencies existed and a shared understanding was reached to prevent researcher bias and strengthen the internal validity [
26].
Representative quotations were identified to report the findings. Quotations were then translated from Swedish to English by PA and an assistant (native Swedish speaker and with good competence in verbal and written English), and thereafter controlled by PA and LL.
Discussion
This study investigated influences on PAP recipients’ engagement in physical activity. Analysis of the interview data yielded ten factors that were perceived to influence the interviewees’ engagement in physical activity in the 1.5–2.5 years after receiving the prescription. The factors were mapped onto COM-B; a framework that posits that behaviour (such as engaging in physical activity) depends on the capability, opportunity and motivation to undertake this behaviour. The three determinants of COM-B interact to influence behaviour, which in turn influences these components [
22]. COM-B was found to be useful to categorize factors found to influence physical activity.
Two of the factors concerned the PAP recipients’ capability to engage in physical activity: tailoring the PAP to the individual’s physical capacity and accounting for the individual’s earlier experiences of physical activity. These factors underscore the importance of adapting the physical activity to each PAP recipient’s physical and psychological capacity. An individual prescription of physical activity is considered one of the key components in the Swedish PAP concept [
5], and our findings lend credence to the importance of accounting for the individual characteristics of the recipients.
The PAP recipients’ previous experiences of physical activity seemed to influence their initiation of activity, whereas experiences during the activity period were important for sustained activity. We found that the existence of diseases influenced the PAP recipients’ capability to engage in physical activity, which is in accordance with previous research in which diseases were associated with fear and avoidance of activity [
6,
27,
28]. Diseases or bodily symptoms have been found to negatively influence the self-efficacy of PAP recipients, i.e. belief in one’s capability to complete tasks and reach goals in becoming more physically active [
29]. Matching the activity with the individual’s capacity and experience can be expected to enhance the individual’s self-efficacy with regard to physical activity [
30]. Mastery of experiences, i.e. previous successful performance of physical activity, is an important factor determining a person’s self-efficacy, and a higher level of self-efficacy can be achieved by means of feedback on past performance [
31].
Six of the ten factors influencing the PAP recipients’ physical activity were attributed to opportunity to engage in physical activity, which in COM-B refers to extraneous factors that enable or bring about behaviour. Receiving professional counselling and follow-up from a physical activity counsellor as well as having support and someone who can “push” for continued physical activity were considered important. The importance of social influences on behaviours is highlighted in many social cognitive theories on behaviour change. For example, Social Cognitive Theory [
30] assumes that an individual’s behaviours are influenced by the actions that the individual has observed in others. This was evident in our study. We found that PAP recipients believed that they were influenced by being “observed” by others, most importantly the counsellors and spouses. A previous study found that spousal support for physical activity was perceived to be important for regular exercise but shared participation in physical activity was uncommon; regular exercise appeared to be largely individual and independent of others [
32].
Our findings point to the importance of PAP counsellors in supporting the PAP recipients to find an appropriate physical activity and achieving sustainability of the activity over time. This is in line with previous studies of multi-professional PAP programmes, which have suggested that physical activity counsellors enhance physical activity in prescribed patients [
4,
33‐
35]. In this study, we found that the prescribers appeared to be the deciding factor for PAP recipients to seek counsellor support because they informed the patient about the option to receive counsellor support or referred directly to a counsellor. The flexibility in the duration of counsellor support found in our study seemed to support sustained physical activity for the PAP recipients. Counsellor visits were offered based on each recipient’s personal needs, and some PAP recipients seemed to require support for more than a year after the prescription. To our knowledge, this has not been described in previous studies of PAP from the patient perspective. There is no “one-size-fits-all” programme for PAP, a finding that is supported by other PAP studies from the patient perspective [
9,
29,
36‐
38].
Two of the factors influencing the PAP recipients’ physical activity were related to their motivation: the desire to improve his or her health condition and finding appropriate activities that encourage continuation. Both factors suggest the relevance of having more autonomous motivation, i.e. engaging in physical activity for personal interests and enjoyment, rather than doing it because one feels compelled by pressures, i.e. controlled motivation [
39]. Self-Determination Theory posits that all behaviours lie along a continuum of relative autonomy, reflecting the extent to which a person endorses what he or she is doing. At one end of the continuum is behaviour that is intrinsically motivated and performed for its inherent satisfaction, e.g. for the fun, interest or challenge it offers. At the other end, is amotivation, which refers to a lack of intention to perform the behaviour. A considerable body of research exists that shows that more autonomously motivated behaviours’ are more stable, performed with greater care and quality, and accompanied by more positive experiences [
40]. Studies on physical activity have confirmed the relevance of achieving more autonomous forms of motivation [
41]. For example, Fortier et al. [
42] found that physical activity counsellors trained in Self-Determination Theory and MI facilitated change in recipients’ physical activity behaviour by fostering both the quantity and quality of motivation.
There was a great deal of interdependency among the different factors (i.e. sub-categories) and the three COM-B determinants (i.e. categories), with many of the factors related to opportunity having an impact on PAP recipients’ capability and motivation to engage in physical activity. The success of PAP programmes depends on viewing physical activity not merely as an endeavour or task for the individual PAP recipient; physical activity must be considered in the broader context of the recipient’s social circumstances.
This study has a number of methodological shortcomings that must be considered when interpreting the results. The factors influencing PAP recipients’ engagement in physical activity should not be considered as list of all possible factors because other studies may yield different factors or prioritize different factors. The results cannot be directly transferred to other county councils or international settings. We sought analytical generalization by comparing our findings with comparable studies.
The sample was purposively selected to account for a variety of characteristics. However, we were unsuccessful in recruiting PAP recipients in the 18–29 years age range. We lack information on why invited PAP recipients declined interviews, although some of the reasons that were reported by non-participants were “I did not receive a PAP” and “I could not participate in any activity due to spousal or their own health status,” indicating that many did not want to participate because they had not (or believed they had not) undertaken sufficient physical activity to be involved in the study. The interviews were conducted 1.5–2.5 years after the PAP was received, which suggests a potential risk of recall bias, i.e. an error caused by differences in the accuracy or completeness of the participants’ recollections regarding initiation of their physical activity. We believed this period was necessary to obtain information about the participants’ engagement in physical activity because PAP strives to reach longer-term maintenance of physical activity in everyday life. However, because of the length of time between PAP and the interview, the study does not allow for detailed description of time-specific influences.
The study applied COM-B as a framework. When using a preconceived framework, there is a risk of neglecting ways of analysing the data that may provide important insights [
43]. However, COM-B was found to provide a suitable framework for informing the data collection and analysing the data. One limitation with COM-B in this study is that the framework does not relate to a time perspective, which would have provided valuable information concerning the process of changing physical activity. In this study, the time perspective was addressed by asking the participants about factors associated with how they got started with an activity and how they maintained the activity. The selection of one specific theory, model or framework usually means that weight is placed on some aspects at the expense of others, thus offering only partial understanding. Regardless, we found COM-B to be broad enough to enable us to use an inductive approach. COM-B was not applied until the second phase of data analysis, which meant that the data had already been analysed inductively to arrive at the sub-categories, i.e. factors.
The study was conducted by a multidisciplinary team of researchers, which was strength of the study. When interpreting the data, this allowed for multiple perspectives and different pre-understandings concerning the issue of PAP, thus enhancing the rigor and credibility of the findings. The team consisted of two (female) nurses with experience in clinical patient work as well as work on miscellaneous research issues (PA, LL), a (female) primary care physician (SH) and an experienced (male) implementation researcher, who is an economist and organizational analyst (PN).
Acknowledgements
We would like to express our sincere thanks to the participants for giving their time and sharing their experiences. Thanks also to Professor Margareta Kristenson, Department of Medical and Health Sciences, Division of Community Medicine, Linköping University, for valuable input at the early stages of planning of the study.