Introduction

In its simplest form, social prescribing (SP) refers to the process through which patients are referred to non-clinical services based within their local community. The services themselves encompass a wide range of activities that include, but are not limited to, sports, art, music, volunteering, storytelling, meditation, yoga, gardening, and cookery. Social prescribing is primarily, but not solely, for people with long-term health conditions that create or exacerbate psycho-social issues, for which medical or clinical intervention is not optimum. As a concept, SP makes use of voluntary and community sector organisations to improve and encourage self-care, facilitate health-creating communities, and reduce the number and intensity of costly interventions in urgent or specialist care services (Dayson and Bennett 2016; Moffatt et al. 2017).

As an approach to healthcare, social prescribing has been practiced informally for many years; however, in the last few years its benefits and potential have been formally recognised by the NHS through its inclusion in the NHS Long Term Plan (Alderwick and Dixon 2019), which means there is increased national focus on the success of social prescribing. Consequently, it is being widely promoted and adopted as a means of dealing with some of the pressures on general practice (Bickerdike et al. 2017). The concept behind social prescribing is about expanding the range of non-medical options available to general practitioners (GPs) and their patients, enabling greater patient involvement in their own care planning (South et al. 2008; Brandling and House 2009). While SP referrals themselves are made by professionals from various services (e.g., emergency, pharmacy, and social care), the majority come from GPs and are used to address issues connected with physical wellbeing, mental health, and social isolation. GPs are therefore a vital part of the social prescribing model, as they can utilise their status to encourage patients to attend appointments with link workers and social prescribers, and therefore help individuals get appropriate support for their needs (Bertotti et al. 2018).

A recent review of social prescribing effectiveness (Chatterjee et al. 2018) found several positive patient outcomes, including increase in self-esteem and confidence, improvement in mental well-being and positive mood, and reduction in anxiety, depression, and negative mood. Despite these positive findings, the review identifies a number of gaps in the evidence base and makes recommendations for future evaluation and implementation of referral pathways (Chatterjee et al. 2018). One of those gaps is the willingness of GPs and other medical professionals to adopt this type of paradigm shift into their practice (Husk et al. 2019); another gap is the lack of evidence to support reductions in GP workloads (e.g., Loftus et al. 2017), which is one of the main advantages often highlighted by proponents of the concept.

Against this background, there is limited data on the effectiveness of social prescribing as perceived by medical professionals (Elliot et al. 2022), as well as the extent to which GPs understand the concept and make use of it with patients. Literature focusing specifically on GPs' attitudes to SP is lacking; however, broader research on the perspectives of primary care healthcare professionals provides insight into the themes that need to be explored further with GPs. A study of 24 nurse practitioners, GPs, and practice managers demonstrated a general awareness of SP; however, 63% expressed the need for a deeper understanding of the practice (Moore et al. 2022). Additionally similar results were found during a cross-sectional study of pharmacists, in which 63.3% were entirely unaware of SP (Taylor et al. 2019). Despite the gaps in understanding and awareness, the general perspective on SP was positive, with many healthcare professionals keen to begin to engage more with SP when better informed (Moore et al. 2022). Pharmacists were especially interested in participating, suggesting that their involvement in SP might minimise the number of unnecessary GP appointments, reducing GP workload (Taylor et al. 2019).

As the results from both studies highlighted understanding of SP structure to be a barrier to its utilisation, this project explored the attitudes of GPs to the concept of social prescribing. Its aim was to collect a wide range of responses from across the UK to document GP experiences of social prescribing activities within their practice, and to further understand some of the barriers and facilitators that have been identified previously (e.g., Husk et al. 2019; Santoni et al. 2019).

Method

Design

This study was conducted using a mixed methods design consisting of an online survey and semi-structured interviews. The study was approved by the University of Nottingham Faculty of Medicine and Health Sciences Research Ethics Committee. The online survey consisted of a mix of multiple choice and Likert-scale questions, along with a free text box where participants were invited to provide additional narrative on their attitudes and experiences of SP. The semi-structured interviews used a range of broad themes to guide questioning, which followed closely to the themes in the survey.

Participants and recruitment

The survey was distributed through opportunity sampling, via social media, and through contacts within UK medical schools, between October 2021 and November 2022. The inclusion criteria for respondents were to be a GMC registrant and currently practising as a GP in the NHS, and this was verified via the GMC pin number that respondents were required to provide. There was an option for respondents to provide their email address to be contacted to take part in the interview section of the study. Survey responses were provided by 102 eligible participants, eight of whom consented to take part in semi-structured interviews via Microsoft Teams, held between October and December 2022.

Procedure

Informed consent was obtaining via an online form prior to the start of the survey, and participants were free to withdraw at any point by prior to submission of the responses. Participants were required to complete a short (10-minute) survey, which asked about their approach and understanding of social prescribing. Participants were offered the opportunity to leave an email address if they would like to be contacted for interview. All participants interviewed were re-informed of the details surrounding the study and reminded of their right to decline to respond to a specific question or withdraw from the process at any time. Verbal consent was gained, evident in the interview transcripts, and the interviews commenced. Each interview was conducted on Microsoft Teams, ranging from 10–35 min.

Quantitative data analysis

Survey data was exported from Jisc Online Surveys and analysed using SPSS. Initial questions asked were whether respondents used SP in their practice, and their geographical location. These were followed by 20 statements about SP which were responded to on a Likert scale as follows: 1 = Disagree, 2 = Somewhat disagree, 3 = Neutral, 4 = Somewhat agree, 5 = Agree. The statements were designed to probe attitudes and experiences, and so there was some required overlap in the themes covered by the statements, to be able to assess response consistency. Responses to each of the 20 statements were analysed using a form of factor analysis, principal component analysis (PCA), which aimed to derive underlying links, or components, between the statements. Before this took place, re-coding of the scaled responses occurred to group together agreement and disagreement categories, to form three overall response categories for each statement: 'Agreement', 'Neutral' and 'Disagreement'.

Qualitative data analysis

The approach to thematic analysis outlined by Nowell et al. (2017) was employed to guide the analytical procedure. Firstly, the free text responses and interview transcripts were converted into a uniform format and read several times with the sole purpose of familiarisation. This process allowed for a greater understanding of the data to develop. Similarities and differences between accounts could begin to be identified along with re-occurring themes and possible codes.

Once familiar with the data, the coding procedure began. Codes were created to simplify and summarise concepts within the data and were recorded and adapted throughout the process. Transcripts and responses were revisited several times to ensure redundant codes were not created, and each code had explicit boundaries. Descriptive codes were later reviewed and grouped into lower-level themes. For example, the codes ‘Alternative to medical treatment’ and ‘An option when medication fails’ contributed to the formation of the lower-level theme ‘Position of SP within the treatment of patients'.

The next phase consisted of the creation of abstract, conceptual, higher-level themes. Lower-level themes were evaluated and connections between them were deduced, thus allowing a hierarchy of themes to emerge. Initial higher-level themes were joined together, leading to the final list of themes. During this process, new themes were held against those below them along with the raw data to ensure the final themes were well grounded in the opinions of the participants. Once the final list of themes was produced, a process of evaluation followed. The themes were reviewed as a collective to ensure the main concepts found in the source material were appropriately reflected. Semantic adaptions were made to ensure the final themes successfully conveyed the wider concepts they represented (Fig. 1).

Fig. 1
figure 1

Diagram of the high-level themes and the corresponding sub-themes derived from thematic analysis coding

Full size image

Results

Every GP respondent (n = 102) had heard of social prescribing as a method of treatment; however, not all of them reported that they used it in their current practice. Appendix 1 in the Supplementary Materials file shows where in the UK GPs practiced and whether they made use of social prescribing in their practice. In addition, 89 of the 92 GPs who reported using social prescribing also use a link worker as part of the scheme. All of the GPs (n = 10) who reported not using social prescribing at all agreed that this was because they don’t believe in it as a practice, and they have a lack of access to the resources needed to make it worthwhile.

Survey results

Initially, the data from the responses to the 20 Likert scale questions (n = 102) were examined for their suitability to be included in the factor analysis. Several well-recognised criteria for the factorability of a correlation were used. Firstly, it was observed that all 20 items correlated at least 0.3 with at least one other item, suggesting reasonable factorability. Secondly, the Kaiser–Meyer–Olkin measure of sampling adequacy was 0.75, above the commonly recommended value of 0.6, and Bartlett’s test of sphericity was significant [χ2 (190) = 654.15, p  < 0.001]. The diagonals of the anti-image correlation matrix were also all over 0.5. Finally, the communalities were all above 0.3, further confirming that each item shared some common variance with other items. Given these overall indicators, factor analysis was deemed to be suitable with all items.

PCA was used because the primary purpose was to identify the factors underlying the relationship between the statements, all of which probed attitudes and experiences. Initial eigen values indicated that the first six factors explained 26.3%, 9.9%, 8.1%, 7.3%, 5.9%, and 5.2% of the variance respectively. Factors 7 to 19 had eigen values under one, and cumulatively explained 38% of the variance. Solutions for four-, five- and six-factor models were each examined using varimax rotations of the factor loading matrix. The six-factor solution, which explained 62.9% of the variance, was preferred because of: (a) the questions included in this solution being similar to one another in terms of themes, (b) the ‘levelling off’ of eigen values on the scree plot after six factors, and (c) the questions included in the two final components of the six-factor solution touched on themes that were not covered in the first four components, and so it was deemed important to include them to be fully representative of the range of attitudes measured. See Table 1 for results and Appendix 2 in the Supplementary Materials file for original survey statements.

Table 1 Factor loadings and communalities based on a principal component analysis with varimax rotation for 20 items (mean response scores used)
Full size table

This analysis revealed that the best fitting model included six separate component themes to describe the underlying attitudes and experiences of the GP respondents. Scrutiny of the statements contributing to each factor (see Table 1), and post-hoc cross-referencing with the themes derived from the interview data for sense checking, indicates that the six component themes could be labelled as follows:

  • Component 1 (comprised of survey statement numbers 1, 2, 7, 9, 14, 17): SP is effective, and GPs are confident in its use and implementation.

  • Component 2 (comprised of survey statement numbers 8 and 10): SP ethos is important and there is enough guidance to support its usage.

  • Component 3 (comprised of survey statement numbers 4, 5, 18): There is enough training available and there are enough external organisations available for referral.

  • Component 4 (comprised of survey statement numbers 11 and 12): Patients lack awareness and education on SP.

  • Component 5 (comprised of survey statement number 19): GP willingness to implement SP but need further support to do so.

  • Component 6 (comprised of survey statement numbers 6 and 13): SP has a perceived stigma and patients prefer clinical interventions.

These themes were further explored via the qualitative data collection and interpretation.

Qualitative results and interpretation

Responses from GPs who participated in semi-structured interviews (n = 8), and free-text comments provided in the survey (n = 55) were also included in this thematic analysis. Three core themes emerged from the thematic analysis of the participants perspectives on SP: ‘Attitudes towards and comprehension of Social Prescribing’; ‘Structure’ (with the sub-themes ‘Funding’ and ‘Delivery’); and ‘Perceived Efficacy’. The Supplementary Materials File Appendix 3 outlines the Author Reflexivity statement, which highlights the awareness of the subjectivity of response analysis. Supplementary Materials File Appendix 4 contains the linked supporting quotations referred to throughout this section.

Theme 1: attitudes towards and comprehension of social prescribing

All of the GPs interviewed and those who provided free-text responses, demonstrated a sound understanding of the concept of SP. Frequently the respondents recognised the benefits of the practice, and most had examples of patients who were positively impacted by SP. The majority of positive accounts involved the use of a link worker where GP involvement was not extensive, and patients were perceived to receive substantial support (Q1). The narrative was echoed by other respondents, and together showed agreement that the GP’s responsibility in the SP process was the identification of a patient’s needs and the following referral to a link worker.

However, throughout the responses, negative attitudes towards SP and poor comprehension of the practice were presented as two of the greatest barriers to its success. Several respondents reported their patients were completely unaware of SP or were unsure what it entailed. This lack of understanding was also reported to be shared amongst GP colleagues, and respondents frequently offered this as a primary reason for the lack of engagement exhibited by some of their colleagues (Q2).

For the respondents who were undecided on the benefits of using SP, concerns with regard to patient attitudes towards the intervention were said to further dissuade them from using it (Q3; Q4). The suggested resolution to the negative reactions exhibited by patients was substantial advertising with educational campaigns. GPs theorised that this would consequently lead to an improvement in patients’ attitudes to SP, and help foster a culture where SP is seen as suitable an intervention as medication (Q5).

However, before large efforts are made to educate the public on SP several respondents felt the term ‘social prescribing’ itself needed to be revisited (Q6; Q7). The term SP was suggested to be a misleading description of the aims of SP and the interventions available within it. Some believed the phrase did not provide a clear picture of the ethos of SP, making it harder for patients to grasp and potentially harder to implement. Additionally, it was said to contribute to the medicalisation of non-medical issues, as the term prompted non-clinical determinants of well-being to be discussed using medical vocabulary.

Overall, the respondents were positive about the benefits that could be gained from using SP; however, lack of general awareness and comprehension of SP was highlighted as a major barrier to its success. GPs asserted that very few of their patients knew about the pathway and that some GP colleagues had a poor understanding of it. This was theorised to decrease SP usage and possible successful outcomes. Moreover, the phrase ‘social prescribing’ was said to display a lack of clarity, unhelpful when attempting to promote a new initiative. However, with substantial awareness raising campaigns, many were hopeful to see a change in attitudes that would drive SP’s success.

Theme 2: structure

The delivery and funding structures underpinning SP was a major topic of discussion amongst respondents. The implementation models used were viewed positively when operating as intended, with many praising the contributions link workers made to primary care. However, criticism arose when discussing the day-to-day realities of SP, with many stating poor funding left SP prone to structural breakdowns. Moreover, the expected involvement of GPs in the model of SP caused a significant divide in opinions.

Delivery of SP

It was unanimously agreed that the non-medical factors such as socioeconomic status, social networks, and educational background impacted health, creating a need for a formal pathway to help combat such issues. However, opinions diverged when discussing the degree to which the onus of such problems fell on GPs, and the NHS more broadly. Although in the minority, some GPs expressed the belief that SP inappropriately passed the failings of society on to GPs and subsequently was not their responsibility (Q8). Social and community care were repeatedly referenced as the appropriate domain for SP to operate within, this highlighting the conceptualisation of SP as a social programme that should mainly exist outside their boundaries as providers of medical expertise. Despite this, the proximity of SP to medicine was appreciated, with many identifying SP as a programme that allowed for non-medical and mild clinical issues presented at general practice to be combatted while also ensuring their time is used wisely (Q9).

The ability of SP to help GPs utilise their time and clinical skills more effectively is a significant potential benefit, as many respondents suggested general practice as it is currently structured does not allow the time for GPs to use SP. The primary reason given was the administrative role GPs felt SP placed upon them, which was described as frustrating and not an appropriate use of time (Q10–12). GPs repeatedly reported frustrations with referral processes, with many stating it was lengthy and time-consuming. However, this was not the case for a minority of respondents (n = 2), who described a quick and simple process. This discrepancy highlights the irregularities within the implementation structure of SP nationally; however, it also illustrates that administrative tasks could be adapted to better suit GPs.

Perspectives regarding the role of GPs in SP were divided; however, respondents shared similar opinions on the role of link workers within primary care. The link worker model was viewed as the most conceptually sound, leading to the best patient outcomes, with manageable levels of GP involvement when done correctly. The use of a link worker increased the likelihood of a patient receiving adequate support and subsequently led to better engagement, unlike direct signposting, done frequently due to ease, but which yields mixed results. Those who had worked with link workers expressed appreciation for the current knowledge of SP interventions which link workers possessed, something GPs did not have the time to maintain (Q13). Removing the barrier of lack of knowledge of relevant services and how to access them for both GPs and patients was noted as a crucial factor to the smooth running of SP. Aside from this, GPs also valued the time which link workers were able to commit to support patients throughout their intervention, something they themselves were unable to do within the confines of short consultations and full patient lists (Q14).

The additional social and emotional support link workers can offer was deemed integral to the success of SP. This aligns with the groups for whom SP was reported to be most appropriate, namely patients who lacked social support, adequate education, confidence, or motivation. However, many reported the link worker model to be prone to structural breakdown. The main causes of these breakdowns included poor link worker retention and unmanageable caseloads suggested to be a result of underfunding. This often led to long waiting times, and unsatisfied patients left without support. Moreover, some GPs expressed dissatisfaction with the training link workers received. However, one GP recognised and appreciated link workers who appeared to have knowledge beyond that expected for their role (Q15).

Experience working with link workers with a degree of clinical knowledge was unique to one respondent; however, their account highlights the possible benefits. Link workers with medical background may aid the selection of the most suitable interventions for patients, helping them support the patients holistically. Additional barriers to the effectiveness of the link worker model discussed were poor communication and inappropriate positioning within practices. GPs expressed poor relationships between link workers and GPs as well as link workers and patients at times, causing reluctance to engage in the service (Q16–17).

This illustrates the poor communication many GPs experienced when attempting to engage with and use SP. Difficulty contacting and maintaining a rapport with link workers left some GPs feeling discouraged, as it made continuity of care increasingly difficult. This is made worse as link workers were said to fail to build relationships with patients sufficiently well for them to receive maximum benefit for SP interventions, due to heavy caseloads.

Overall, respondents recognised the potential benefits of SP’s placement in primary care. However, opinions on the ownership of the SP framework, as well as the appropriate level of GP involvement, were divided. While many respondents were willing to participate in addressing non-medical factors that may affect their patients’ health, others felt SP was inappropriate use of NHS resources and GP time. However, GPs holding both opinions demonstrated frustration at the structural implementation of SP, which was often prone to breakdown. This produced doubt concerning the sustainability of the link-worker model, which was highly favoured by many GPs due to the knowledge link-workers held and the personal support which patients are able to receive.

Funding

GPs identified funding issues as a significant hurdle to SP’s success. The two most pressing areas of concern were the current funding structure through NHS budgets and a general lack of funds needed to sustain the link worker model (Q18). The funding structure behind SP was perceived as inadequate by many GPs; however, viewpoints on a more appropriate structure varied. Those unconvinced of SP’s efficacy in concept were frustrated by its reliance on primary care network (PCN) budgets and NHS finances more widely. The description of SP as a programme that inhibits service in some responses indicates that for some, SP does not make a notable impact on patient outcomes, it reduces GPs ability to offer the best medical patient care. The view that SP is an inappropriate part of budgetary expenditure suggests that SP’s impact has not been substantial enough to be recognised as a valuable service in primary care or produces costs too great to offset any perceived benefits.

For those who deemed SP a valuable use of NHS finances, increased funding for the link-worker model was called for. This model was deemed the most successful in theory; however, it also seemed to suffer the most from inadequate funding (Q19).

Responses from GPs supported the decision to implement link workers as the primary model of SP, evident in the NHS Long Term Plan (Alderwick and Dixon 2019); however, an underestimation of the funds needed for its success was expressed many times. GP reports of link workers overstretched due to high caseloads and responsibility over an unsustainable number of practices were often attributed to the lack of funds needed to hire more link workers. This proved a major cause of structural breakdown, as link workers were said to resign regularly due to excessive caseloads and were subsequently difficult to replace (Q20–21).

Following inadequate link worker funding, lack of financial support for the third-sector organisations SP relies upon was highlighted as a major failing in funding structure. The support SP can offer patients relies almost exclusively on the intervention provided by organisations often surviving on donations and inadequate government funding. Lack of funding was recognised as a major flaw of SP, as it applies further pressure to such organisations without the appropriate increase in support needed to meet the rising demand (Q22). Appreciation for the support offered by the third sector was shown by many. However, the dependence on these organisations was deemed to be unreliable or unsustainable. Many GPs reported such organisations to be ever-changing and available for short periods of time. Furthermore, the availability of interventions varied greatly with region, leaving many areas with a lack of options (Q23–24).

The vast amount of discussion surrounding funding highlights GPs frustration and concerns surrounding the financial backing of primary care and the strain SP will impose on their practices. SP was discussed as an inappropriate allocation of resources by some, and a promising yet underfinanced area by many. Initially, these opposing opinions may be seen to elucidate the difference in perspectives on SP’s efficacy. However, if SP received the increase in funding needed in the opinion of many GPs, perspectives on its usefulness might change as it was able to better fulfil its role.

Theme 3: perceived efficacy

The efficacy of SP is primarily based upon patient experience with their care and final outcomes. Many GPs deemed SP promising in theory; however, SP often fell short in practice. Notably, all respondents felt unable to comment on SP’s ability to improve patient outcomes due to a lack of knowledge on the interventions received or the pathway itself. The most probable reason for this appeared to be the lack of feedback GPs received once a patient had been referred. Some offered anecdotal evidence to support the possibility of its efficacy when the programme ran as intended, but respondents were cautious to make claims on its success as a whole (Q25).

Many GPs shared feelings of discouragement, as they were often unaware of the outcomes of their referrals. Although the use of link workers was felt to be more effective than sign posting or direct referrals, GPs recognised that poor communication frequently left them ‘out of the loop’, which was less likely to occur with other models. This is potentially more concerning when discussing the link worker model, as GPs stated they were more likely to use it for those with more complex needs. The lack of short-term resolution associated with SP was off-putting to GPs, who had become accustomed to patients presenting to them with a complaint and leaving the practice with a resolution or clear next step. However, this was not the experience of every GP, with some supporting the efficacy of SP (Q26). When considering patient experience, the consensus view suggested that SP can offer good quality care; however, structural breakdowns often led to inconsistencies in the quality patients received. This is exacerbated by the reliance on voluntary sector organisations and a lack of standardisation in the implementation of SP pathways (Q27). This was said to be the case even between GPs in the same practice (Q28).

Secondary to patient outcomes, SP has been proposed as a means by which to decrease the strain on general practice, GPs' workloads, and reduce the demand for NHS resources in the long term. The potential for SP to reduce this demand is one of the major benefits underpinning respondents’ optimism for the programme. Generally, GPs were knowledgeable of the theoretical impact SP may have; however, most GPs expressed either no change, or an increase in their workload. Some reported a decrease in workload during periods where SP was running effectively; however, a sustained positive effect was scarcely reported. As with the anticipated improvements in patient outcomes, the desired decrease to GPs' workload was seen to be stifled by issues of the implementation and structure, which frequently meant GPs had to fulfil the tasks of link workers and take on administrative roles which were overly time-consuming and burdensome (Q29–30).

Disappointment was expressed by some GPs who had been hopeful that SP would decrease their case load and increase the time available to treat patients. However, the experiences relayed by one GP illustrated SP’s capacity to do so when running effectively (Q31).

Aside from a decrease in GP workload, the most commonly anticipated outcomes for SP included a reduction in prescriptions, hospitalisations, and repeat or unnecessary GP appointments. Some GPs also believed it had the potential to empower patients to play a greater role in their own health. Those who perceived that SP empowered patients suggested that the support of link workers helped patients gain the necessary confidence to address non-clinical issues themselves, while others felt SP fostered further reliance on GPs for issues that ought to be self-solved. With the impact on patient outcomes variable and dependent on effective implementation, the longevity of SP was questioned. Some respondents interpreted poor funding and inconsistent outcomes as an indicator of a programme that will not stand in the long run. It was considered that this lack of confidence in SP’s longevity may contribute to some doctors’ reluctance to engage with the process. Nevertheless, many GPs do not believe it has been given the necessary time, planning, and funding for confidence to grow.

General discussion

This study used a mixed methods approach to explore the attitudes and experiences of UK general practitioners (GPs) using social prescribing as a healthcare intervention. The quantitative and qualitative analyses derived six and three component themes respectively. These themes showed agreement between the two methodologies in eliciting the key issues felt by GPs in terms of their attitudes and experiences of social prescribing. The labels given to these themes across the two data sets can be merged into the following overarching themes:

  • 1. Perceptions of SP effectiveness and efficacy

The data shows that the extent to which GPs perceive SP to be effective is a key driver for its successful use as an approach to healthcare. The success of the practice was largely measured in terms of patient outcomes, impact on NHS resources, and its overarching conceptual strength. The survey data is supported by the qualitative responses, where multiple positive accounts of SP successes were provided, demonstrating the various ways it has the potential to improve patient outcomes, especially where psycho-social issues are exacerbating, or being exacerbated by, a medical or health issue. This is consistent with literature showing that other healthcare professionals recognise the benefits of SP when it is working as intended (Moore et al. 2022; Taylor et al. 2019). However, inconsistencies in the care SP provided was a frequent criticism which acts as a caveat to the experiences had by respondents. There was consensus that SP as an approach to healthcare often struggled to provide consistency for patients, and that the intricacies of non-medical problems meant that a ‘one size fits all’ approach to referrals was not optimal, which is in line with previous literature on the topic (Husk et al. 2019).

  • 2. Training and awareness

GPs identified lack of awareness of SP from both doctors and patients as a significant barrier to ongoing expansion of this as a model of healthcare provision. However, responses unanimously pointed toward one solution: education. The need for further training and awareness raising was a theme drawn out in the initial survey analysis, and one that become much clearer throughout the interview data. The perception from respondents was that patients’ scepticism towards SP was rooted in the fear of the unfamiliar, which often prompted them to seek the treatment pathways they knew and understood. However, if a significant effort was made to increase the public’s awareness and understanding of SP, a shift in attitudes could be expected, leading patients to view SP optimistically. Furthermore, educational campaigns would allow the opportunity for patients to become more proactive in their healthcare, which may increase engagement (Vahdat, et al. 2014).

In addition, educating clinicians on the potential benefits of SP along with local guidance on the expectations around use may also reduce GP reluctance. It is important to note that most UK medical school curricula do not include SP as a core component (although it is acknowledged that some do include optional components which cover this topic), and so it is likely that SP presents a conceptual change from current practice for many doctors, who will need support and guidance to implement it effectively. Santoni et al. (2019) showed that after raising awareness of SP as a conceptual approach to healthcare, medical students were more inclined to make clinical decisions using SP as a pathway. Further integration of SP as a concept into clinical education and training is one way of ensuring a more consistent understanding and use, which may overcome some of the current barriers to effective use of SP previously identified (e.g., Husk et al. 2019).

We also have to acknowledge the role of raising patient awareness of SP. This includes making patients aware of the validity of this as a pathway, and also the types of conditions that might benefit from signposting to a link worker. What is also clear is that patients need to remain confident that any medical needs would still be addressed by the GP and that this can happen alongside an SP referral.

  • 3. Structural and operational delivery

This study shows that fundamentally it is the model of delivery used to implement SP that determines its success or not, in any one area. Throughout this data we can see the positive impact of link work schemes as a most effective means of referral into SP services, which is consistent with existing evidence (Moffatt et al. 2017; Bickerdike et al. 2017; Pescheny et al. 2020). However, we can also see that these need to be properly implemented, resourced, and embedded into standard practices and pathways in order for a positive GP and patient experience to result (Bickerdike et al. 2017). There also needs to be focus put onto the regional resources and model of delivery available to fully implement SP as a viable alternative to psycho-social care offered currently, as primary care and GP practices cannot do this effectively in isolation. Indeed, this data shows that the need to be aware of well-established community pathways and suitable referral options, including those that would suit specific types of patient need, drives the operational success of SP within a community. This is where consistent links with voluntary and third-sector organisations are crucial, especially those that provide bridging services for multiple community groups. This is perhaps one of the biggest barriers, which is to maintain accurate mapping of the facilities available on a regional basis for equity and ease of access, which could be overseen by integrated care boards or other commissioning bodies.

Strengths and limitations

This study provides one of the largest published set of responses from GPs on the topic of attitudes to using SP in their practice. Therefore, it provides a unique perspective on the facilitators and barriers to effective implementation of SP, in the context of the ongoing debate around lack of high-quality evaluative evidence in this field. By collecting thematic responses from clinicians, we are able to further understand the nuances that impact on their SP referral decisions, and gain insight into what enables SP to be used effectively.

Reflection on this study should consider its methodological limitations. The analysis of both interview data and free text responses allowed a sufficient sample size to reach saturation. Participants spoke freely and at length at interview; however, it is acknowledged that those with stronger opinions regarding SP may be more receptive to taking part in the study. This was partially combatted by the inclusion of free text responses in the thematic analysis, which could be argued to include a wider range of opinions due to the anonymised nature, although it is still plausible that the same bias exists. A larger study on this topic could employ stratified sampling to overcome this. This sampling approach could also overcome another limitation, which is the variation in regional locations of respondents. Whilst we feel that we do have representation from across the UK, there are areas with more respondents, which may reflect a more established SP network or set of pathways in that area, hence a potential skew in the data arising from this.

Conclusions and future directions

To date, GPs attitudes to social prescribing have not been extensively explored; therefore, this study provides new data which enhances our current understanding. The topic of efficacy and effectiveness was discussed at length, indicting that longitudinal data following the impact of SP on patient outcomes would be a particularly informative piece of evidence. This would provide a greater basis for the judgement of SP as a cost–effective care pathway. It may also be beneficial to investigate the attitudes and opinions of link workers as well as the public. Given the assumptions and beliefs GPs presented regarding their views of and experiences with SP, it would be beneficial to also explore the degree to which GPs accurately interpreted the perspectives of both parties. Such research would allow the evaluation of SP from the perspectives of link-workers, who are integral to the model deemed most effective by GPs. Moreover, the opinions of the public could help identify misconceptions and gaps in understanding of SP, this would greatly help in the education of the subject.

While previous research has focused on the effectiveness of social prescribing, emphasising its aims and barriers, this study has focused on the attitudes and experiences of GPs towards using social prescribing as a treatment option. This data contributes a new insight into the relationship between GPs opinions on social prescribing and their intended future use of it in their practice.