Comparative effectiveness of nature-based social prescribing to alleviate loneliness in Spain: a decision analysis informed by the RECETAS trial

Background Social prescribing appears to be an effective intervention to reduce loneliness. More evidence is needed on the long-term benefits of nature-based social prescribing (NBSP) to alleviate loneliness. Objective We aimed to evaluate the long-term effectiveness, in terms of life expectancy (in life years: LY), quality-adjusted life expectancy (in quality-adjusted life years: QALYs), and loneliness-free years (LFYs), of a NBSP intervention compared to active control (AC) as loneliness-reduction strategy among adults in Spain. Methods We developed a Markov state-transition model where participants moved between different health states to estimate LYs, QALYs, and LFYs. Loneliness, reduced loneliness, or relapsed loneliness, was combined with either being in good health or one of the following loneliness-related diseases: cardiovascular diseases, diabetes mellitus type2, depression, anxiety, and dementia. Death from other causes, loneliness-disease-related death and suicide were considered. A lifelong time horizon was applied to evaluate long-term health outcomes of NBSP and AC. The model was populated with data from the “Friends in Nature” (FiN) pragmatic randomized controlled trial conducted in Barcelona (Spain) within the EU Horizon 2020 RECETAS project supplemented with Spanish data. FiN, a nature-based, facilitator-led group-based social prescribing model was compared to AC. We followed a Spanish cohort with a mean age of 61 years, comprising 80% women, experiencing loneliness. Several scenario and sensitivity analyses were performed. Results NBSP resulted in 23.98 LYs, 15.15 QALYs, and 10.73 LFYs, versus 23.68 LYs, 14.76 QALYs, and 8.11 LFYs with AC. This corresponds to gains of 0.30 LYs, 0.39 QALYs and 2.62 LFYs for a Spanish cohort of lonely adults. Results remained robust in scenario analyses. Deterministic sensitivity analysis showed that all incremental outcomes were most sensitive to assumptions regarding the relative intervention effectiveness in reducing loneliness after six to twelve months, the relapse probability during this period, and the increased mortality risk associated with loneliness. Results of the PSA suggested NBSP as the optimal strategy under parameter uncertainty in terms of LYs, QALYs, and LFYs in 89% of all simulations. Conclusion Our analyses suggest that NBSP is effective over the long-term in reducing loneliness and improving health outcomes in Spain. Further research, including a cost-effectiveness analysis, is needed to better inform decision-makers and support the implementation of NBSP within healthcare systems across different countries. Clinical trial registration ClinicalTrials.gov , identifier NCT05488496.