What would our world look like if people relied less on medicines and pills, and more on social activities, personal growth and self-development? The UK is currently undergoing two major population health shifts: increasing health challenges amongst an ageing population, and deteriorating mental health in younger generations. Together, these trends are increasing pressure on the UK’s mental healthcare system and driving greater reliance on medication. Prescribed medications remain the standard treatment for many physical and mental health conditions. While effective in many cases, they are often seen as ‘treating symptoms rather than the cause’. As a result, if individuals return to the same social environment that contributed to their illness, they are likely to become unwell again.
For decades, research has shown that 70% of health outcomes are attributed to social, economic and behavioural causes (Hood et al., 2016). This has led to a gradual focus on social prescribing, a form of care that refers people to non-medical support and services (including arts and culture, heritage, natural environment, physical activity, and advice and information) in local communities, to address social needs and improve health and wellbeing. Although social prescribing has been rolled out for over 30 years, it was unclear whether social prescribing was provided equally across demographics. Studies by Fancourt & Steptoe (2025) and Bu et al. (2025) were designed to explore this.
Imagine a world where people relied less on medicines and pills, and more on social activities, personal growth and self-development.
Methods
Two recent studies, using different data sources, looked into the patterns and predictors of social prescribing:
- The first paper uses data from the English Longitudinal Study of Ageing (ELSA), a nationally representative cohort study of adults aged 50+ who reported their experience with social prescribing in 2021-23 (N=7,000) (Fancourt & Steptoe, 2025).
- The second paper uses data from the Clinical Practice Research Datalink (CPRD), a research data service that collects anonymised patient data from over 2,000 GP practices across the UK (in which the study focused on England – covering over 20% of GPs in 2009-2023; N=1.2m) (Bu et al., 2025).
While the ELSA study analysed data based on participants’ survey responses, the CPRD study analysed data aggregately using all data with a medical code related to social prescribing. Single-level regressions analysis, multi-level regression analysis and growth curve modelling were used to understand the patterns and predictors of social prescribing.
Results
Trends and patterns
The CPRD data show that social prescribing began to increase rapidly and steadily following its formal integration into the NHS Long Term Plan in 2019 (NHS England, n.d.). Growth trajectories suggested that the rise of social prescribing was almost linear since 2019, increasing by roughly 126 consultations or 61 patients per GP practice each year. In 2023, there were around 544 consultations and 278 patients related to social prescribing per practice. Based on what the researchers found in their sample, they estimated that around 1.6 to 1.9 million patients would have social prescribing recorded across all 6,311 GP practices in England in 2023. This equates to around 8.8 to 10 million consultations including any discussions of social prescribing from 2019 to the end of 2023, of which, 5.2-5.9 million had social prescribing codes specifically indicating that referrals took place.
Using data from ELSA, researchers found that exercise classes were the most frequent intervention of social prescribing, followed by arts groups, nature-based activities, and adult learning.
Who was most likely to be referred to social prescription?
Using data form ELSA, results from regression models suggested that referrals were more common for individuals who were older, those with diagnosed psychiatric conditions and above-threshold depressive symptoms, and those with cardiovascular conditions, diabetes, and chronic pain. People who were lonely were twice as likely to be referred, although martial status and frequency of social contact were not related. Similarly, those from the lowest wealth tertile were more likely to receive referrals, as were individuals receiving benefits. However, no association was found for the levels of area deprivation, educational attainment, urban dwelling, or working status. Behaviourally, individuals who were physically inactive were most likely to receive referrals. The social patterning of social prescribing uptake (rather than just referral) produced very similar findings.
When looking at referrals overtime between 2017 and 2023, the CPRD study showed that females were consistently overrepresented in social prescribing consultations, while patients from ethnic minority backgrounds showed a gradual increase since 2020, accounting for around 1 in 5 consultations in 2023. There was also a slight increase in the proportion of younger adults receiving social prescribing since 2020. Similarly, the proportion of people living in deprived areas who accessed social prescribing increased from 12.5% of referrals in 2017 to 25.7% in 2023 amongst the top 30% most deprived areas. Only around 13% of referrals were for individuals from rural areas.
Who was most likely to refuse social prescription?
Using data from CPRD, 22.2% of patients had refused social prescribing service, which declined to 10.6% in 2021 and remained stable since. Notably, in 2019, patient (rather than practice) characteristics appeared to play an increasingly important role in service refusal. Results from regression models suggested that social prescribing refusal was more common amongst older adults, males and those of white ethnicity. Over time, patients from less deprived areas had become less likely to refuse.
Researchers estimated that around 1.6 to 1.9 million patients would have access to social prescription across all 6,311 GP practices in England in 2023.
Conclusions
These two studies highlight the fast expansion of social prescribing, with referrals surpassing national targets of 900,000 patients by 2023-24 (NHS England n.d.). Such increase shows the clear appetite for social prescribing, especially in primary care, and that social prescribing has the potential to reach individuals from disadvantaged backgrounds. However, access to social prescribing remains unequal, with people from certain demographic groups more likely to receive referrals, to take part, or to refuse. Ensuring equitable access to social prescribing is essential for it to serve as a tool in addressing the social determinants of health.
Social prescribing has been rapidly expanding over the years, surpassing national targets, yet access remains unequal.
Strengths and limitations
The two studies, drawing on different data sources, have nicely complemented each other. For instance, while ELSA focused on general social prescription (so not just those referred by GPs) reported by participants and only currently has one wave of data, CPRD had been longitudinally tracking social prescribing records using primary care data (so providing a more objective measure). ELSA allows for comparisons between those who received social prescribing and those who did not, whereas the CPRD study was restricted to patients with social prescribing medical codes, making comparisons not feasible. Further, ELSA contains data on prescribed interventions, which are not captured in the CPRD. Yet, CPRD has a much larger sample for researchers to identify patterns and predictors of social prescribing refusal, which is not possible in ELSA due to a small sample size reporting social prescribing referral. Such research triangulation through different data sources has enabled researchers to explore the trends, patterns, and predictors of social prescribing more comprehensively, while mitigating the shortcomings inherent in individual datasets.
Despite this, both studies have some common limitations. For instance, while they helped monitor and track the implementation of social prescribing at both local and national levels, using large and nationally representative samples, both ELSA and CPRD data currently lack detail in social prescribing on referral reasons and patient outcomes. This has prevented researchers from gaining a nuanced understanding of social prescribing’s reach and effectiveness. Particularly, it would be interesting to compare effectiveness both across different interventions (e.g. arts groups vs exercise) within social prescribing and between social prescribing and medical prescriptions, and how long the impacts can be observed for.
Further, several key questions remain unanswered:
- How many times have patients been referred to social prescribing?
- How many sessions have they attended?
- Why have some people declined the social prescribing offer?
- Do social prescribing referrals made by GPs have higher uptake rates?
Answers to these questions will further enhance our understanding on the scalability and feasibility of social prescribing.
These two studies, drawing on different data sources, provide a comprehensive picture of patterns and predictors of social prescribing, yet some key questions remain to be answered.
Implications for practice
Social prescribing has been long recognised as an alternative or complementary approach to medical treatment, offering a personalised approach aimed at addressing the social determinants of health. Evidence from these two studies highlights the rapid expansion and broad reach of social prescribing across England, indicating that people from less advantaged backgrounds are also increasingly accessing and engaging with this approach. This expansion reflects two emerging developments: first, a growing emphasis on integrating psychological, social and biomedical approaches to improving health; and second, a shift in treatment focus, from recovery to fostering social connectedness and self-empowerment. Nevertheless, the findings from these studies also highlight two key considerations that require further attention.
Firstly, although GP practice is not the only referral pathway for social prescribing, patients are most likely to visit their GPs when they feel unwell. In this context, it would be valuable to explore whether GP referrals might lead to higher uptake of social prescribing compared to other types of referrals, and why some GPs are more likely than others to refer patients to social prescribing. Future research should investigate the effectiveness of social prescribing, as well as individual- and neighbourhood-level factors influencing its uptake and long-term adherence amongst patients.
Secondly, although social prescribing reaches some less advantaged individuals, access remains unequal. To democratise social prescribing access and engagement, reforms in health and cultural policies are needed. Policy initiatives could include increasing financial support for third-sector community groups to enhance the availability and quality of diverse local activities prescribed to patients, as well as raising public awareness of the benefits of social prescribing.
Healthcare models continue to evolve, but their ultimate goal should be to enable individuals to live independently and lead fulfilling lives without relying on medication. Social prescribing may be a way to support this.
The ultimate goal of healthcare models should be to enable individuals to live independently and lead fulfilling lives without relying on medication. Social prescribing may be a key way to support this.
Statement of interests
The author of this blog works with some of the researchers involved in the studies, but was not involved in this piece of work.
Links
Primary papers
Bu, F., Burton, A., Launders, N., Taylor, A. E., Richards-Belle, A., Tierney, S., Osborn, D., & Fancourt, D. (2025). National roll-out of social prescribing in England’s primary care system: a longitudinal observational study using Clinical Practice Research Datalink data. The Lancet Public Health, 0(0). https://doi.org/10.1016/S2468-2667(25)00217-8
Fancourt, D., & Steptoe, A. (2025). Can social prescribing reach patients most in need? Patterns of (in)equalities in referrals in a representative cohort of older adults in England. Perspectives in Public Health, 145(4). https://doi.org/10.1177/17579139251330767
Other references
Hood, C. M., Gennuso, K. P., Swain, G. R., & Catlin, B. B. (2016). County Health Rankings: Relationships between Determinant Factors and Health Outcomes. American Journal of Preventive Medicine, 50(2), 129–135. https://doi.org/10.1016/j.amepre.2015.08.024
NHS England » Social prescribing. (n.d.). Retrieved 24 January 2020, from https://www.england.nhs.uk/personalisedcare/social-prescribing/
