Abstract
Background Social prescribing link workers (SPLWs) have been based in general practices in Scotland since 2016 but the views of GPs on their work have not been quantified.
Aim To quantify GPs’ satisfaction with the work of SPLWs and belief that they can reduce health inequalities, and to explore GP and practice factors that influence these views.
Design & setting Secondary analysis of a cross-sectional survey of all qualified GPs’ working lives in Scotland conducted in 2023–2024.
Method Descriptive analysis of how satisfied GPs are with SPLWs, and whether they believe they can reduce health inequalities, and univariate and multivariate analysis of factors that influence these views.
Results In total, n = 836/1380 (60.6%) of GPs reported having a SPLW in their practice, and of these, 567 (67.8%) were satisfied with their work, and 587 (70.2%) thought they could reduce health inequalities. In multilevel multiregression analysis, three significant independent positive predictors of GP satisfaction with the work of SPLWs were identified: female GP sex (P = 0.017), high practice deprivation (P = 0.001), and a perceived reduction in GP workload (P<0.001). GPs’ belief that SPLWs can reduce health inequalities was predicted by high practice deprivation, and a perceived reduction in GP workload (both P<0.001).
Conclusion GPs in Scotland who have an SPLW in their practice are largely satisfied with link workers. GPs believe SPLWs can reduce health inequalities, especially those working in deprived areas and who perceive reductions in their own workload owing to the work of the SPLW.
How this fits in
Social prescribing through link workers based in general practice has been rolled out across the UK and in many other countries, as a way to help reduce health inequalities and improve non-clinical issues affecting health. However, the views of GPs on social prescribing link workers are not well known. In a national survey of GPs in Scotland, we found high levels of satisfaction with the work of SPLWs, and GPs believed they could reduce health inequalities. These views were especially positive in GPs working in high deprivation areas.
Introduction
Scotland has the largest disparities in both life expectancy and healthy life expectancy between the most and least deprived areas in the UK.1 It also has the widest health inequalities in Western Europe, with research showing the gap between high and low socioeconomic status is growing.2,3 On average, individuals in the most deprived 10% of Scotland’s population live 12 years less than those in the most affluent 10%. The difference in healthy life expectancy is even greater, with people in the most deprived areas spending more than 10 additional years in poor health before dying prematurely, compared with those in the least deprived communities.4
The Scottish Government (SG) has promoted social prescribing as one way of reducing or mitigating health inequalities. In 2016, the SG made a commitment to introduce 250 social prescribing link workers (SPLWs) to address health inequalities, particularly in deprived communities, and reduce pressure on general practice.5 SPLWs were to support patients with non-clinical issues affecting health, such as housing, financial insecurity, and social isolation, and by connecting them to community resources and services. Later in England, more than 3500 SPLWs were employed in primary care teams as part of the NHS Long Term Plan 2019.6 Between 2019 and 2024, they referred 2.7 million people.7
In 2018, SPLWs were embedded as one of six priority areas in the first-ever Scottish GP contract, which also saw the expansion of multidisciplinary teams (MDT) in primary care.7 Following the 2018 Scottish GP contract, funding and governance arrangements for SPLWs shifted centrally from the SG to Scotland’s 31 Health and Social Care Partnerships (HSCPs), which were formed to integrate health and social care services provided by Scotland’s 14 health boards and 32 local authorities, and came into effect following SG legislation in 2014.8 The rollout of SPLWs was delivered as part of the SG’s Primary Care Improvement Plans (PCIPs) and mainly funded through the Primary Care Improvement Fund (PCIF).9 Although most, but not all, of Scotland’s SPLWs are formally employed by community-based third sector voluntary organisations to facilitate their social prescribing role, they are allocated to GP practices. As of March 2024, 325 SPLWs were working in Scottish GP practices.9 Of approximately 900 GP practices in Scotland, around 700 have some form of access to an SPLW and while some are allocated to only one, others are spread across numerous practices.10
To date, there has been no national evaluation into the rollout of SPLWs at scale in Scotland’s GP practices. An early pilot study in Glasgow’s socially deprived Deep End practices found successful integration of SPLWs was associated with GP ‘buy-in’ and strong GP engagement, collaborative leadership, effective team dynamics, and robust support for SPLWs.11 An evaluation of five ‘early adopter’ SPLW programmes, which started in 2017 in four Scottish health boards, found facilitating effective SPLW working involved the following: having a lead GP with responsibility for the service; good relationships with practice managers and reception staff; SPLWs being attached to one practice full-time; and practice readiness for change.12 Currently, there is limited evidence on how GPs across Scotland view SPLWs. GP ‘buy-in’ is known to be a key factor in the successful implementation of SPLWs12 but this has not previously been quantified nationally. Thus, assessing GP satisfaction with the work of the SPLW is of key importance. Second, in the context of deprivation, GPs’ views on whether they believe SPLWs can reduce health inequalities is another key issue, given that this is a primary aim of the rollout of SPLWs in Scotland by the SG. GP and practice factors influencing these views are also important to understand, especially perceived impact on GP workload, given that less than 50% of GPs have reported a reduction in workload owing to the expansion of the MDT (including SPLWs).13
The aim of this study was to explore the following questions:
How many GPs in Scotland have access to an SPLW in their practice?
For those who have an SPLW:
How satisfied are they with the work of the SPLW?
Do they believe that SPLWs can reduce health inequalities?
How do these views vary by GPs’ individual and practice characteristics?
Method
A survey of GPs’ working lives and views was posted to all qualified GPs in Scotland in October 2023, with two postal follow-ups between December 2023 and January 2024.13 The focus of the survey was to assess GPs’ working lives, future work intentions, and views on the Scottish GP contract. The response rate was 31% and the GPs who responded were broadly nationally representative in terms of individual and practice characteristics.13
The following three questions on SPLWs were included in the survey:
Q1. ‘Do you have access to a link worker in your practice?’ (Answered as ‘yes’, ’no’, or ‘don’t know’).
For those who answered yes:
Q2. ‘Are you satisfied with the work of the link worker?’ (Answered on a five-point Likert scale, from ‘very satisfied’ to ‘very dissatisfied’).
Q3. ‘Do you think link workers can reduce health inequalities?’ (Answered ‘yes, definitely’, ‘yes, to some extent’, ‘no’, or ‘don’t know’).
GP information was collected on demographics and employment details. GP variables included age, sex, ethnicity, if they were a partner in the practice, clinical sessions worked per week, and (if they had access to a link worker) whether they had felt the SPLWs had reduced their own workload (‘some access to a link worker but insufficient to reduce my workload’ scored as 0, and ‘sufficient access to reduce my workload’ scored as 1). Practice variables included practice list size (number of registered patients), deprivation level of patients registered in the practice measured by the Scottish Index of Multiple Deprivation (SIMD)14 and rurality, based on the proportion of registered patients per practice living in urban or rural areas.15
Analysis
Descriptive (univariate analysis) was first carried out on the data. The two key variables of interest were: (a) satisfaction with the work of the SPLW (as measured on the five-point Likert scale, from 1–5); and (b) belief that the SPLW can reduce health inequalities (scored as 1 ‘yes, definitely’, 2 for ‘yes, to some extent’, and 3 for both ‘no’, and ‘don’t know’). The distributions of these two variables were relatively normal (see supplementary file Figures S1 and S2, with information on skew and kurtosis) and thus linear multiregression analysis was then conducted with these two questions as the dependent variables. Because several GPs within the same practice would be commenting on the same SPLW, we determined the extent of clustering at practice level and found intraclass-correlation coefficients for satisfaction with SPLW of 0.428, and belief they can reduce health inequalities of 0.160. In light of this, multilevel modelling was used with adjustment for GP practice as a random effect. The other independent variables were entered as fixed variables into the models and included age, sex, ethnicity, whether a partner in the practice or not, years in current practice, clinical sessions worked per week, and whether the SPLW had reduced their workload. Practice characteristics were list size, deprivation (SIMD score as quintiles), and rural or urban locality. Based on our previous work, all of these variables could feasibly influence GPs’ views on link workers16 and many were correlated with each other (Supplementary file, Table 1).
There was no imputation of missing data. All analysis was carried out using SPSS (version 27) and R Studio.
Results
Regarding the question ‘Do you have a link worker in your practice’, 99.6% (1380) of GPs in the survey responded, of which 60.6% (836) replied ‘yes’, 34.6% (477) replied ‘no’, and 4.9% (67) replied ‘don’t know’. As shown in Table 1, SPLWs in practices in the most deprived areas (quintile 5) were more than twice as likely to have an SPLW than those in the least deprived (quintile 1) (P<0.001), and SPLWs were more common in urban areas than in rural areas (P<0.001).
GP satisfaction with the work of the social prescribing link workers
Overall, 67.8% (n = 567/836) of GPs who had an SPLW in their practice were satisfied with their work (30.4% ‘very satisfied’; 37.5% ‘satisfied’; 24.2% ‘neutral’; 6.0% ‘dissatisfied’; 1.9% ‘very dissatisfied’). In univariate analysis, satisfaction was higher in female GPs (P = 0.005), urban GPs (P = 0.026), and those working in deprived areas (P<0.001) (Table 2). Satisfaction was also significantly higher (P<0.001) in GPs who reported that they had sufficient access to an SPLW to reduce their own workload, in both high and low deprivation practices (Figure 1). However, satisfaction was significantly higher in GPs in deprived practices than affluent, irrespective of the effects on workload (P<0.001, Figure 1).
Figure 1 is a grouped bar chart showing the association between satisfaction with social prescribing link workers and the perceived effect on GP workload, stratified by practice deprivation quintile.The y-axis shows the percentage of responders who were “very satisfied” with the work of the link worker. The x-axis has two categories describing whether the link worker reduces workload: “some access but insufficient to reduce workload” and “access sufficient to reduce workload”. Green bars represent the least deprived practices, and red bars represent the most deprived practices.In practices reporting some access to link workers but insufficient access to reduce workload, 10.9% of responders in the least deprived practices were very satisfied, compared with 25.6% in the most deprived practices. The difference between deprivation groups is statistically significant, with p = 0.005.In practices reporting access sufficient to reduce workload, 45.8% of responders in the least deprived practices were very satisfied, compared with 67.5% in the most deprived practices. The difference between deprivation groups is statistically significant, with p = 0.006.Overall, satisfaction is higher when link worker access is sufficient to reduce workload, and within both workload-reduction categories, the percentage of responders who were very satisfied is higher in the most deprived practices than in the least deprived practices.
Multilevel multiregression analysis was conducted to control for possible confounding by the GP and practice characteristics and thus identify the true independent predictors of GP satisfaction with the work of the SPLWs. All of the GP and practice characteristics shown in Table 1 were included as independent variables, plus the variable on whether GPs perceived that SPLWs had reduced their own workload.
As shown in Table 3, three independent variables were found to be significantly associated with higher GP satisfaction with the work of the SPLWs in linear regression analysis: female sex (P = 0.017), higher deprivation (P = 0.001), and perceived reduction in GP workload owing to the work of the SPLWs (P<0.001). Tolerance and Variance Inflation Factor (VIF) scores indicated that multicollinearity was not a problem (Table S2 supplementary file). Additionally, the P-P plot of the dependent variable residuals showed only minor deviations from the diagonal, supporting the assumption of normality for the residuals (Figure S3, supplementary files).
GPs’ belief that social prescribing link workers can reduce health inequalities
Overall, 70.2% (n = 587/836) of GPs who had an SPLW in their practice thought they (SPLWs) can reduce health inequalities (18.0% ‘yes definitely’; 52.4% ‘yes to an extent’; 17.3% ‘no’; 12.3% ‘don’t know’). In univariate analysis, this belief was higher in GPs who were not partners in the practice (P = 0.016) and those working in deprived areas (P<0.001) (Table 4). Belief that the SPLWs can reduce health inequalities was also higher in GPs who perceived that SPLWs had reduced their own workload (P<0.001).
In multilevel multiregression analysis, the GP and practice characteristics shown in Table 4 were included as independent variables, plus the variable on whether GPs perceived that SPLWs had reduced their own workload. The analysis indicated that two independent variables were significantly associated with GPs’ belief that SPLWs can reduce health inequalities (dependent variable): working in a deprived area (P<0.001); and the perception that SPLWs reduce the GPs’ own workload (P<0.001) (Table 5). Again, multicollinearity was not an issue, and the P-P plot of the dependent variable residuals showed no major deviations from the diagonal (Figure S3, supplementary files).
Are the findings influenced by GPs’ own work satisfaction or life satisfaction?
There was no significant correlation between GP satisfaction with SPLWs and GPs’ own job satisfaction (Spearman’s rho -0.013, P = 0.710) nor life satisfaction (Spearman’s rho 0.018, P = 0.601). Similarly, there was no significant correlation between GPs’ belief that SPLWs can reduce health inequalities and GP job satisfaction (Spearman’s rho -0.002, P = 0.951) nor life satisfaction (Spearman’s rho 0.042, P = 0.277). Adding either of these two GP variables into the regression models had no effect on the findings (results not shown).
Discussion
Summary
Of the 60.6% of GPs who reported having an SPLW in their practice, more than two-thirds were satisfied with their work, and seven out of 10 believed that they can reduce health inequalities. After controlling for potential confounders, the key explanatory factor for satisfaction with the work of the SPLW was a perceived reduction in the GP’s own workload. Working in a deprived areas and female sex were also significant independent predictors. Belief that SPLWs can reduce health inequalities was also mainly related to perceived reduction in GP workload, but also to working in a deprived area.
Strengths and limitations
A main strength was that the survey was national and included all qualified GPs in Scotland, and the responders were broadly nationally representative in terms of a range of individual and practice characteristics. Other strengths included using many validated variables across many workforce factors such as GP job satisfaction.
Potential weaknesses include the cross-sectional design of the study, from which causality cannot be inferred, and the use of bespoke items to measure GP satisfaction with the work of the SPLWs, and their belief that SPLWs can reduce health inequalities. However, this was necessary as no validated questions on these items exist, as far as we are aware. Including more questions around these two areas could have given additional useful information, but the survey’s main focus was about GP working life and job satisfaction.13 We were thus limited in the number of additional questions we could include.
Future qualitative work is required to explore GPs’ views on SPLWs and their possible effect on health inequalities in more depth. Additionally, GPs’ views on whether SPLWs can reduce health inequalities is not a proxy of measurement of actual reduction. This would require direct measurement and quantification on large datasets and/or randomised controlled trials.
Comparison with existing literature
Our study highlighted high satisfaction with SPLWs, especially for those GPs who perceived a reduction in their own workload owing to the SPLW. In a survey of 208 GPs across 33 countries, 56% were aware of social prescribing, and GPs who utilised social prescribing reported enhanced job satisfaction.17 In contrast, our study did not find a link between GP job satisfaction and views on link workers. The reason for this discrepancy is not clear but could relate to many changes that have happened in Scottish general practice since the introduction of the new GP contract in 2018, and the rapid expansion of the multidisciplinary team that has brought advantages and disadvantages, but has not improved GP job satisfaction.13 Regarding workload, a recent National Academy for Social Prescribing report in England suggested that SPLWs can decrease GP consultations, particularly among frequent users.7 However, this should be viewed with caution owing to a lack of control group and small sample size.
More than 70% of GPs in the present study who had access to an SPLW also believed that SPLWs could contribute to reducing health inequalities. Overall, there is little direct evidence to support this view.18 In previous work in Scotland in deprived areas, the effectiveness of SPLWs in improving patient outcomes (and thus potentially reducing health inequalities) was only seen after three or more contacts with the SPLW, at which point uptake of suggested community resources increased dramatically.19 Recent work on a large population sample suggests that the rollout of link workers in England is associated with slightly better outcomes for population groups with long-term conditions (specifically targeted for social prescribing)20 but the same authors also found that the rollout of link workers has not been sufficiently targeted at areas of high deprivation.21 There is a lack of specific theoretical models on SPLWs and health inequalities, but our previous work using self-determination theory has been used to illuminate how SPLWs can help motivate patients in deprived areas towards healthier behaviours.22 However, the role of SPLWs in reducing or mitigating health inequalities needs to be considered within the broader context of the distribution of wealth and power in society and the social determinants of health such as employment, income, education, and housing.23 Additionally, the way that SPLWs are distributed, employed, and managed, including the extent to which they are integrated into the primary care team, differs between Scotland and England, which may also affect the impact of SPLWs.24
Implications for research and practice
We found that GPs working in deprived areas were more satisfied with the work of SPLWs even when their own workload was not reduced (although the effect was larger when their workload was perceived to be reduced) (Figure 1). This may be understood in the context of the inverse care law, and the fact that GPs in deprived areas are dealing with a high volume of patients with complex needs relating to psychosocial problems compared with GPs working in affluent areas.4 This implies that distributing SPLWs by deprivation, as has happened in Scotland5 but not currently in England,20 is regarded as generally beneficial by GPs but further qualitative research is required to more fully understand GPs’ views on SPLWs in deprived and less deprived areas. Further quantitative research is needed to determine the effect of link workers on GP workload. Research is also required to quantitatively demonstrate any effect of SPLWs on health inequalities.
In conclusion, GPs in Scotland who have an SPLW in their practice are largely satisfied with link workers and believe they can reduce health inequalities, especially those who perceive reductions in their own workload owing to the work of the SPLW. Further research is required to confirm or refute these perceptions.
Notes
Funding
This survey was funded through a research grant from the Economic and Social Research Council (reference: ES/T014164/1). Paul Wilson and Helen Frost’s input to this paper was funded by the National Institute for Health Research (grant number: 134066).
Ethical approval
Ethical approval was obtained from the Wales REC 6 research ethics committee (reference: 21/WA/0078), and research and development approval from participating Scottish Health Boards.
Provenance
Freely submitted; externally peer reviewed.
Data
The dataset relied on in this article is available from the corresponding author on reasonable request.
Acknowledgements
We would like to thank our Patient and Public Involvement group: Colin Angus (Chair), Morag Cullen, Mary Hemphill, Anne Marie Kennedy, who gave valuable feedback throughout the research programme. Special thanks to our Patient and Public Involvement Coordinator: Jayne Richards. We would like to thank all the GPs who contributed to the survey.
Competing interests
The authors declare that no competing interests exist.
- Received July 2, 2025.
- Revision received September 17, 2025.
- Accepted October 1, 2025.
- Copyright © 2026, The Authors
This article is Open Access: CC BY license (https://creativecommons.org/licenses/by/4.0/)
References
LinkedIn
Jump to section
Keywords
More in this TOC Section
Related Articles
Cited By...
