Through social referrals, medical professionals, including general physicians and nurses, can divert patients with social difficulties, such as personal problems, toward nonclinical resources and community-based organizations. General practitioners and primary care services can benefit from this in relieving the extra workload by offering a different and versatile way to address patients’ problems when standard procedures are inappropriate (8). GPs occupy a central position in enforcing the comprehensive well-being of patients in a bid to promote the overall good of patients; social prescription is a new concept that has attracted attention (4). The purpose of a green park’s prescribing practice is to create a list of activities, such as being in touch with arts, nature, and exercise, which benefit mental and physical health (11). Research has shown that participants improve their health and increase social cohesion and community vitality, particularly those who deal with mental health problems (8). Social prescription, focused on well-being, is a holistic concept that involves physical, mental, relational, and emotional aspects in helping people develop and sustain good practices. Therefore, 17 countries have decided to design and apply dedicated social prescribing interventions to address psychological and social factors that could improve health and well-being within the framework of multiple health systems (24). With the help of local community organizations, China has started a community-led social prescribing scheme under a comprehensive evaluation team, which includes general practitioners (25). GPs carry out social prescribing activities effectively because of their close connection to patients. Knowing GPs’ education, thoughts, and viewpoints about their impact on social prescription is the prelude to effectively incorporating this practice into our medical practice. Existing studies (3–5, 7, 9) have highlighted the influence of social, cultural, and environmental circumstances on behavior patterns. Nevertheless, comprehensive research on how these factors impact general practitioners’ attitudes and knowledge in prescribing nonpharmacological interventions is lacking. They primarily work at the primary level (Tier 1) of the healthcare system of China, community health centers (CHCs) and stations within cities, and township health centers (THCs) within rural areas. Their core functions include preventive medicine, initial treatment/triaging of common diseases, management of chronic conditions, health prevention, and making referrals. Therefore, the current study aims to address this research gap by comprehensively understanding the influence of general practitioners’ perceptions (knowledge, attitudes, and perceived effectiveness) on arts, nature, physical activity, and social prescription for psychosocial health and well-being (see Figure 2).

Interest in the role of social prescription as a central part of modern healthcare, particularly in its influence on psychosocial health, is increasing. Healthcare providers’ positive opinions of collaboration help toward a complete collaboration process in which shared decision-making is encouraged in mental health care. This can be explained again by the instrumental cost–benefit reasoning behind the understanding that patients and GPs collaborate to reach better health (26). Educational interventions seem to result in better prescribing between GPs, and such educational input may contribute to the acceptability and effectiveness of social prescribing (27, 28). A previous study indicated that experts with insights into social support dynamics can devise effective social prescriptions for addressing particular clinical requirements of individuals and communities (29). This strongly indicates that encouraging enabling beliefs in social prescriptions drives adherence to such interventions and boosts their psychosocial benefits (30). At the community level, an evidence synthesis with a systematic scoping review demonstrated the relationships among health professionals’ feelings about social prescribing and patient outcomes, reduced loneliness, and enhanced community integration of vulnerable populations (31). The evidence suggests that a GP’s knowledge about social prescribing could influence the perceived effectiveness of SP for psychosocial health and well-being. The following hypotheses were proposed on the basis of the literature to achieve the study objective.

Art-based interventions in social prescriptions have made a significant difference in individuals’ mental and psychosocial health. A study by Jensen et al., indicated that arts-on-prescription could positively impact an essential improvement in mental health, specifically emotional well-being (32). Furthermore, GPs who are informed about social prescriptions are well placed to recommend creative interventions that encourage people to ‘feel’ their anxiety and depression as part of emotional regulation (8). Arts-based activities can play a therapeutic role and promote socialization, and they are integral in overcoming isolation and fostering a sense of community (8). Nature-based activities have been successful in people who are lonely and socially isolated, especially older people (31). GPs who are knowledgeable about nature-based interventions can advocate for prescribing, which may result in more positive patient outcomes (33, 34). These findings are supported by the literature, which confirms that GPs willing to prescribe PA are more likely to promote the health behaviors required for mental health promotion (35). It is also responsible for GPs to nurture an understanding of social prescription practices and how their understanding influences arts, nature, and exercise for psychosocial health and well-being. The following hypotheses were proposed on the basis of the literature to achieve the study objective.

Social prescription is led by general practitioners (GPs), and it is assumed that perceived effectiveness will determine the implementation and embedding of working practices. Engagement with the arts, including arts interventions, has been identified as significantly impacting mental health and well-being. Participatory arts-based interventions, for example, have improved the resilience and perceptions of social connectedness among participants, particularly those in vulnerable subgroups (36). Arts on prescription have been shown to be powerful modalities that significantly improve one’s health and well-being. Consequently, many GPs endorse arts-based prescriptions in medical settings (32). The narrative of the social prescription context, especially of arts, nature, and physical activity, is that GPs can engage as they perceive that, through the domain, perceived improvement can be achieved. A change in perceived effectiveness can occur. Incorporating these factors into treatment paradigms benefits patient health outcomes and may help GPs feel that they are prescribing more effectively. Greater knowledge of local resources and how they can be incorporated into patient care might lead to increased use of social prescribing (37). On the basis of the extent of available evidence from the literature, they have different implications for mental and physical health and perceived efficacy by health workers. The following hypotheses were proposed on the basis of the literature to achieve the study objective.

Social prescribing, where GPs can prescribe to patients a nonclinical service/activity for psychosocial benefit, has the potential to be an innovative intervention. The beliefs of GPs toward the arts and related activities are likely to be a significant determinant of the capacity for social prescribing to impact patients’ psychosocial well-being. Engaging in the arts is known to have a beneficial effect on mental health. For example, integrated arts-based interventions in medical wards have been reported to have beneficial effects on symptoms of anxiety and depression, resulting in improved psychosocial outcomes (37). When GPs facilitate art-based prescriptions, the likelihood that patients engage in such interventions increases, benefiting patient well-being (34). The opinion of GPs regarding social prescriptions is a significant determinant of whether they are likely to be used and, thus, whether referrals will filter through and influence patient outcomes. A supportive stance by GPs with a special interest in nonmedical options establishes a new balanced patient–doctor relationship and is a prerequisite for successful social prescribing (34). GP active champions for these interventions might promote patient adherence and provide a way for models of care where mental health is considered as important as physical health (38). The extent to which arts and physical activity are conceptualized by GPs can significantly impact the precision and potential benefit of social prescriptions regarding psychosocial health and well-being. Through more optimism and the setting of limits in the interventions, the GP, in this manner, supports better health with a higher probability of a good quality of life for the patient. The following hypotheses were proposed on the basis of the literature to achieve the study objective.

The research was exploratory, and a cross-sectional survey design was adopted, aiming to collect data at a single point in time from a sample of general practitioners in China. The study participants gave their informed consent after fulfilling the eligibility requirements (participants must be 25 + years old, have a minimum of 3–5 years of general practice experience, possess knowledge of social prescription, and work in public settings involved in the social prescription for health and well-being). A total of 701 general practitioners were recruited via a simple random stratified sampling technique that fulfilled the inclusion criteria (Table 1).

The primary data were gathered via a self-administered questionnaire survey. It was constructed via a five-step sequential methodology, ensuring the reliability and validity of the instrument (39). After the items were created, a test group comprising 25 people evaluated the questionnaire. Pilot testing can help identify any problems related to the language used in the questions, the answer possibilities, or the clarity of the questions (40). In response to feedback from the pilot study, we revised the questionnaire to increase its user-friendliness. The final questionnaire survey comprised 35 items, each assigned a Likert scale score.

An online survey was conducted to collect the primary data. The online poll followed the CHERRIES guidelines (59). The data analysis for this inquiry used the Smart-PLS 3.2.9 software package (60). The data analysis approach used a biphasic methodology. First, the construct, reliability, and convergent validity were assessed via a measuring technique (see Table 2). A structural model was developed in the second stage to empirically analyze the proposed theories (61, 62).

Table 2 provides a complete summary of the statistical information for several items or variables, including their relative means, standard deviations, excess kurtosis, and skewness. The statistics below enhance the understanding of each item’s characteristics and data distribution.

The measurement model must fulfill the criteria for convergent and discriminant validity. Convergent validity is established when the loading is equal to or greater than 0.70, the average variance extracted (AVE) is 0.5, and the composite reliability (CR) achieves a minimum value of 0.7 (63). Table 3 and Figure 3 confirm the acceptable convergent validity of the research by demonstrating that the variance inflation factor (VIF), AVE and CR are more significant than the threshold values are. This indicates that there are no issues with the convergent validity of the study’s construct.

Discriminant validity (DV) was used to examine and define distinct notions. The DV also confirms all the measurements of the component differences. The discriminant validity of a factor was calculated via its average variance extracted (AVE). The discriminant validity results revealed that the square root of each construct and average variance extracted (AVE) had higher values than their correlations with other constructs (see Table 3). In structural equation modeling (SEM), the Heterotrait–Monotrait ratio of correlations (HTMT) measures discriminant validity across constructs (64). The HTMT ratio compares the correlations of items assessing two constructs. The HTMT ratios in the table indicate that the constructs under study have valid discriminant validity (see Table 4). Model fitness was assessed by calculating SRMR, d_ULS, d_G, chi-square, and NFI. The model fitness statistics show that the proposed model of the study is acceptable for structural equation modeling (see Table 4).

All hypotheses demonstrated statistically significant relationships, as indicated by p-values of < 0.001 (see Table 5 and Figure 4).

The findings of H1 (t = 10.418, β = 0.405, p < 0.001) indicate a positive relationship, implying that higher levels of knowledge among general practitioners are associated with greater perceptions of the effectiveness of social prescriptions in enhancing psychosocial outcomes.

The findings of H2 (t = 2.772, β = 0.107, p < 0.001) indicate that more favorable attitudes toward social prescriptions correlate with heightened perceptions of their effectiveness in psychosocial well-being.

The findings for H3 (t = 0.172, β = −0.010, p < 0.001) suggest a very small inverse relationship. This implies that greater knowledge may slightly diminish the emphasis on arts-based social prescriptions, possibly because practitioners prioritize other modalities or perceive art as less central to psychosocial benefits.

H4 (t = 79.05, β = 0.002, p < 0.001) suggests that knowledge is positively linked to nature-based social prescriptions. This means that knowledge enhancement directly influences the promotion of nature-oriented interventions for psychosocial health.

The findings of H5 (t = 1.276, β = 0.049, p < 0.001) indicated that increased knowledge enhances physical activity as a social prescription. This finding implies that knowledge-building efforts encourage practitioners to integrate physical activity to improve psychosocial well-being.

The findings of H6 (t = 0.655, β = 0.000, p < 0.001) indicate a statistically significant but minimal effect of the perceived effectiveness of social prescription on psychosocial health and well-being by the general practitioner and arts on social prescription.

The findings of H7 (t = 0.540, β = 1.024, p < 0.001) point a strong positive association. This implies that a higher perceived effectiveness substantially bolsters nature-based social prescriptions. This relationship highlights the potential of efficacy-focused interventions to amplify the use of nature-oriented approaches in psychosocial care.

The findings for H8 (t = 0.767, β = 0.120, p < 0.001) indicate a small positive effect. This means that stronger perceptions of effectiveness are linked to greater integration of physical activity as a social prescription for psychosocial benefits.

Findings for H9 (t = 0.201, β = 0.000, p < 0.001). This finding implies the need to explore other factors that drive the adoption of art interventions in psychosocial health contexts.

The findings for H10 (t = 1.409, β = 0.110, p < 0.001) reflect a small positive association. This suggests that positive attitudes modestly enhance nature-based social prescriptions, pointing to attitude modification as a strategy to increase the utilization of psychosocial well-being.

The findings of H11 (t = 0.395, β = 0.013, p < 0.001) showed a significant relationship between general practitioner attitudes and physical activity as social prescriptions. This implies that attitudes have a direct impact on physical activity as social prescriptions.

These findings highlight the multifaceted roles of knowledge, attitudes, and perceived effectiveness in shaping social prescription practices. Overall, the results affirm the theoretical model, with GPK and GPSE generally exhibiting stronger effects than GPA.

This study is an essential contribution to the existing literature on social prescription from the perspective of primary care physicians, but it has methodological limitations. Given that the investigation was cross-sectional, it is only possible to conclude an association at one point in time, without any possibility of inferring causality. Consequently, the temporal dynamics cannot be determined, and the danger of reverse causation cannot be excluded. Reliance on self-reported questionnaire data brings the potential for bias due to response bias, specifically the social desirability effect and recall bias, which can affect the fidelity of measurements directed toward the knowledge, attitude, and perceived efficacy. The analytic sample included 701 general practitioners working in the public sector of China who were recruited according to pre-specified inclusion criteria (aged 25 years or older, had 3–5 years of clinical experience, and had previous experience of social prescription). Accordingly, the external validity of the findings is potentially limited when applying them to other settings, such as the private sector, different urban and rural contexts, and foreign jurisdictions, where cultural, systemic, and policy variables may vary and, in turn, impact physicians’ perceptions. The use of an online CHERRIES-based survey runs the risk of selection bias in favor of those respondents with enhanced technological competencies. Moreover, although the structural equation model has been empirically validated and found reliable, adjustments for numerous potential confounders, such as demographic and extrinsic factors known to influence prescribing behavior, were absent. This omission may cover up subtler nuances in the relations. To overcome these limitations, future studies will be longer, employ mixed-methods research designs, and recruit more participants from diverse ethnic backgrounds. Implementing more rigorous and comparable techniques will increase the robustness and generalizability of the study’s findings and participants’ responses.