Older adults’ engagement with social prescribing to active ageing centres in Singapore: a qualitative study from public primary care

Abstract

Background

Social prescribing (SP) has emerged internationally as a structured approach to link individuals with non-medical community resources to improve health and wellbeing. In Singapore, SP through referrals to Active Ageing Centres (AACs) is in the early stages of implementation within public primary care. Therefore, it is important to understand the perspectives of patients who are referred to such settings from primary care providers.

Objective

To explore older patients’ perspectives on social prescribing and factors influencing uptake and sustained engagement following referral to AACs from public primary care in Singapore.

Methods

We conducted a qualitative descriptive study using semi-structured, in-depth interviews with adults aged ≥ 60 years who had accepted, declined, or dropped out of SP referrals from a public primary care clinic. Purposive maximum-variation sampling sought to capture diversity in engagement status and ethnicity. We analysed the interviews using thematic analysis and performed interpretive mapping using the COM-B model.

Results

Of the sixteen participants approached, twelve participated: four were currently engaged in AAC activities, four had declined referral, and four had discontinued participation after initial attendance. Four interrelated themes described a familiar pathway of engagement: (1) referral legitimises participation; (2) access shapes entry; (3) social fit shapes belonging; and (4) perceived value turns attendance into habit.

Conclusions

Uptake and sustained engagement with SP via AAC referral appeared to be shaped less by individual capability than by physical opportunity, social opportunity and motivation. Primary care providers are well placed to facilitate engagement by leveraging trusted patient relationships, while centre managers and policymakers should address structural barriers such as zoning restrictions, language mismatch and age-identity mismatch. These findings offer actionable insights to optimise the reach and sustainability of social prescribing within Singapore’s primary care landscape.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12877-026-08067-5.

Background

Population ageing is a defining demographic trend in Singapore. By 2030, around one in four residents will be aged 65 years and above [1], with a greater proportion of older adults living alone. Alongside increased longevity, social isolation and loneliness have emerged as important public health concerns, as they are associated with increased likelihood of mortality [2]. These challenges highlight the need to address not only biomedical conditions, but also the social determinants of health that shape healthy ageing [3].

Social prescribing (SP) has emerged internationally as one approach to address such non-medical drivers of health. Broadly, social prescribing refers to the process by which healthcare professionals identify individuals with health-related social needs and connect them to non-clinical community resources in order to improve health and wellbeing [4]. These resources may include physical activity groups, social clubs, befriending programmes, educational activities, and volunteering opportunities. Internationally, SP has been associated with improvements in wellbeing and social connectedness and may reduce healthcare utilisation in some settings [5].

In Singapore, social prescribing remains at an early stage of development. Initial efforts were pioneered in community hospital settings, and momentum has increased with broader national attention to healthy ageing and ageing in place. Active Ageing Centres (AACs) have become key community nodes within this landscape, offering activities such as exercise, cognitive stimulation, social engagement, lifelong learning and volunteer opportunities [6]. At the study site, social prescribing involved referring suitable older adults from the public primary care clinic to AACs through a link worker-facilitated pathway. This formed the local service context for the present study.

Although social prescribing is expanding locally, little is known about how older adults themselves perceive this referral pathway. Existing local studies have largely focused on healthcare and community care workers’ knowledge or perception of SP, rather than patients’ lived experiences of accepting, declining, or discontinuing referral [7]. International qualitative studies suggest that engagement with social prescribing is shaped by factors such as accessibility, activity fit, digital literacy, and perceived benefits, but these findings may not fully translate to Singapore’s unique healthcare system and multicultural social context [8].

This study aimed to explore older patients’ perspectives on SP through referral to Active Ageing Centres from a public primary care setting in Singapore. Specifically, it sought to understand how older adults experienced this referral pathway, including reasons for accepting, declining, discontinuing or sustaining participation after referral. By examining the full engagement pathway among older adults who accepted, declined, or discontinued AAC referral, this study captures not only facilitators of participation, but also points at where social prescribing may break down after referral.

Methods

Study design and setting

We conducted a qualitative descriptive study using semi-structured, in-depth interviews to explore older adults’ perspectives on social prescribing through referral to Active Ageing Centres (AACs) from a public primary care setting in Singapore. We adopted a qualitative descriptive methodology because our aim was to generate practice-oriented accounts of older adults’ experiences of social prescribing. Data were analysed using reflexive thematic analysis, which provided a flexible analytic approach for identifying patterns of meaning and developing interpretive themes while remaining consistent with the study’s descriptive aims. The study was conducted in a public primary care clinic in the central-north region of Singapore, serving a catchment population of approximately 80,000 residents [9]. Public primary care clinics in Singapore provide subsidised care for acute conditions, chronic disease management, preventive health services, and selected child and women’s health services [10].

Participants and sampling

Eligible participants were adults aged 60 years and above who had been offered social prescribing, through referral to an AAC from the study site and who fell into one of the three engagement groups: currently engaged, declined referral, or dropped out after initial participation. Participants with cognitive impairment affecting their ability to describe their experiences meaningfully were excluded.

We used purposive maximum-variation sampling to recruit participants across different social prescribing engagement statuses and ethnic groups, in order to capture a broad range of perspectives. Eligible participants were identified and approached face to face by study team members. Recruitment and data collection were conducted from 16 January 2025 to 18 July 2025.

Conceptual framework and interview guide

The Capability, Opportunity, Motivation-Behaviour (COM-B) model informed the design of the interview guide and the later interpretation of findings. COM-B was selected because it provides a useful framework for understanding behavioural determinants of uptake and sustained engagement in community-based programmes [11].

A semi-structured interview guide was developed specifically for this study and has not been previously published. It explored participants’ experiences of referral, including reasons for accepting, declining, or discontinuing participation. The guide was informed by COM-B as a sensitising framework while remaining open to emergent issues raised by participants. It was co-designed with two patient partners, pilot tested, and refined iteratively during the course of data collection. After each subsequent interview, the interviewer completed a debrief with the research team, and the guide was revised where appropriate to improve clarity and depth of exploration. The English-language semi-structured interview guide is provided as Additional file 1.

Data collection

We conducted the face-to-face English interviews as a four-member team comprising three primary care physicians (JHG, KPS, SYT) and one medical student (HDL). The interviewer team comprised three male researchers (JHG, KPS and HDL) and one female researcher (SYT). None of the interviewers had a prior clinical relationship with the participants. However, the physician interviewers worked within the same public primary care organisation as the study site. No non-participants were present during the interviews. At the start of each interview, interviewers introduced their role as researchers and emphasised that participation was voluntary and that participants’ responses would not affect their clinical care, referral, or access to AAC services. Interviews lasted between 29 and 65 min, were audio-recorded with consent, and transcribed verbatim.

Data collection and analysis proceeded iteratively. Sample adequacy was guided by the concept of information power [12]. The focused study aim, specificity of the referral-based sample, purposive inclusion of participants across different engagement statuses, depth of interviews, use of COM-B as a sensitising framework, and cross-case analysis together indicated that the dataset provided sufficiently rich and relevant information to address the study aim. No repeat interviews were required for clarification. These considerations indicated that the final dataset of 12 interviews provided sufficiently rich and relevant information to address the study aim. Participants were offered the opportunity to review their transcripts for factual accuracy and comment. Following each interview, the interviewer recorded brief field notes and participated in a team debrief to capture contextual observations, emerging issues, and reflections that informed iterative refinement of the interview guide.

Data analysis

Consistent with our qualitative descriptive methodology, data were analysed using reflexive thematic analysis, drawing on Braun and Clarke’s six-phase approach and subsequent conceptual guidance on reflexive thematic analysis [13, 14]. COM-B served as a sensitising framework during interview guide development while remaining open to issues raised by participants. Analysis proceeded iteratively alongside data collection and centred on patterns of meaning within participants’ accounts. Themes were generated inductively through reflexive thematic analysis and only subsequently interpreted through the Capability, Opportunity, Motivation–Behaviour (COM-B) model to illuminate the behavioural mechanisms shaping engagement at different points along the referral pathway.

We also familiarised ourselves with the transcripts through repeated reading and generated initial codes through close, line-by-line engagement with the data. Regular team meetings supported the development and refinement of codes and candidate themes, with discussion of alternative interpretations throughout the analytic process. Reflexive memos were maintained to document analytic decisions and to consider how the team’s clinical and research perspectives informed interpretation. NVivo version 14 was used to support data management.

Reflexivity and analytic rigour

The research team comprised three primary care physicians and one medical student. Our professional experience of caring for older adults and working with community referral pathways informed our contextual understanding of participants’ accounts, while also shaping the perspectives we brought to the analysis. Regular team meetings provided opportunities for collaborative reflexive dialogue, through which we examined assumptions, considered alternative interpretations, and attended to accounts that complicated emerging patterns. Reflexive memos documented these reflections and key interpretive decisions. An audit trail recorded refinements to the interview guide and the development of codes and themes. Rich contextual description and purposive sampling across engagement statuses were used to support assessment of the findings’ relevance to comparable settings [15].

Ethical considerations

Ethics approval was obtained from the National Healthcare Group Domain Specific Review Board (2024-4452), and the study was performed in accordance with the relevant guidelines and regulations laid down by it. Written informed consent was obtained from all participants before interviews were conducted. Sociodemographic data were also collected to characterise the sample. Participants received a voucher worth S$20 as a token of appreciation. This study is reported in accordance with the Consolidated Criteria for Reporting Qualitative Research (COREQ).

Results

Sixteen patients were approached and twelve (75%) participated. Participants were aged 67 to 79 years and included Chinese, Malay, and Indian participants. Four were currently engaged in AAC activities, four had declined referral, and four had dropped out after initial participation. Participant characteristics are summarised in Table 1.

Table 1.

Participant characteristics

Characteristic	Participants, n
Gender
Male	5
Female	7
Age
65–70	2
70–75	5
75–80	5
Ethnicity
Chinese	7
Malay	4
Indian	1
Level of Engagement
Engaged	4
Declined Referral	4
Dropped out	4
Years of Education
0–6 years	2
6–10 years	7
10–12 years	2
> 12 years	1
Medical condition
Hypertension	11
Hyperlipidaemia	7
Type 2 Diabetes	5

Overview of themes

Four interrelated themes described how older adults experienced social prescribing through referral to active ageing Centres. Participants’ accounts suggested a staged journey: referral first legitimised participation, practical barriers shaped entry, social fit influenced whether participants felt they belonged and perceived value sustained ongoing engagement. The findings are summarised and presented in Fig. 1.

Fig. 1.

Data-derived engagement pathway following social prescribing referral to active ageing centres. The figure illustrates the four interrelated themes developed from participants’ accounts: referral legitimises participation, access shapes entry, social fit shapes belonging, and perceived value turns attendance into habit. The pathway is presented sequentially to aid interpretation; however, the stages could overlap, and barriers at one stage could remain unresolved despite progress at another. An illustrative within-participant example is presented following the four themes. The subsequent interpretation of these themes through COM-B is presented in Table 2.

Theme 1: Referral legitimises participation

For many participants, first exposure to SP occurred through an interaction with a healthcare provider. The referral was not experienced merely as administrative signposting to an Active Ageing Centre (AAC). Instead, the recommendation gained legitimacy because it came from a trusted professional who was perceived to understand the patient’s health, lifestyle, and social circumstances.

Subtheme 1.1: Personalised referral made AAC participation relevant

Participants described doctors as being well placed to identify unmet social or activity needs and to translate these into a concrete recommendation to attend an AAC. One participant explained that doctors could frame AAC participation in relation to the patient’s own medical and social context:

“You [Dr] can say … since you have this medical condition and you are living alone, it’s good that you join [Name of AAC] … you [Dr] can lead the path.” [Participant 03].

This suggests that healthcare provider endorsement helped patients make sense of AAC participation as relevant to their own health and wellbeing, rather than as a generic community activity.

Participants also felt that doctors could tailor recommendations because of their understanding of patients’ needs and functional status:

“Maybe [patients] lacking in exercise, then okay, have more exercise course, or you[doctor] know that … they [patients] need to develop some hobby … then you [doctor] can use at least your … position to … see the needs.” [Participant 08].

Subtheme 1.2: Professional endorsement gave permission to try

Referral also carried weight because it came from a trusted professional. Participants regarded doctors as credible sources of health-related guidance, and this professional authority strengthened their confidence in the potential value of attending an AAC.

“Of course, the doctor carries more value. Because … talking about medical, talking about … health things, all these, you are more qualified.” [Participant 04].

For some participants, this translated into willingness to act on the recommendation:

“I will respect, I will still take Doctor [advice] to go because I also need to respect his professional. Definitely, he refer me to be there, will be [of] benefit [to] me.” [Participant 09].

Taken together, these accounts suggest that primary care referral functioned as a legitimacy cue by creating awareness of AACs, establishing personal relevance, and giving participants confidence to take the first step.

Theme 2: Access shapes entry

Referral created motivation but older adults still had to cross practical thresholds before they could enter the AAC. Access operated at three levels: physical proximity, administrative eligibility and environmental welcome. Participants’ accounts showed that an AAC needed not only to be nearby, but also administratively accessible, visible, and inviting enough for referral to translate into attendance.

Subtheme 2.1: Accessibility and zoning restrictions as barriers to entry

Physical proximity lowered the effort required to attend. Centrally located AACs, especially those near public transport and within participants’ usual movement patterns were described as easier to try.

“Location is centralised area … is near the bus stop, is near the MRT.” [Participant 04].

This account illustrates how participation became frictionless when the AAC was already within the participants’ usual activity spaces.

However, geographic proximity alone was not always sufficient. Residential zoning rules assigned residents to specific AACs based on their block and several participants found themselves excluded from the centre nearest to them. The opacity of these rules generated confusion.

“I say I go to [Name of AAC], but they say I’m not under that zone, so I’m not sure.” [Participant 08].

This finding shows that access was not only a matter of physical distance, but also an administrative boundary. Zoning was therefore a strong local implementation bottleneck: patients could be referred, motivated, and living near an AAC, yet still be unable to attend if their residential block did not fall within the centre’s assigned catchment. This shifted access from an individual-level issue to a system-level constraint.

Subtheme 2.2: Visibility and first impressions influenced willingness to enter

Visibility shaped informational access. Participants described wanting to know what the AAC offered before entering and public-facing activity schedules helped residents understand whether the centre had activities that might suit them.

“I went there to have a look, but I could not see any, what you call that, activity or timetable that’s outside. I think there should be some kind of notice board outside to put up the courses … I pass by, I can … see already.” [Participant 08].

This account suggests that when information was concealed within the centre, opportunistic entry by nearby residents was more difficult, even when the AAC was physically close.

Subtheme 2.3: Space and comfort affected whether entry felt welcoming

Once participants arrived, the built environment and atmosphere influenced whether they felt comfortable enough to remain. Physical entry was therefore different from feeling welcomed.

“It doesn’t look like comfortable place. When you go in you must have that feeling that it’s very nice.” [Participant 04].

This account shows that environmental welcome mattered as a form of access. The centre needed to feel comfortable and inviting.

In summary, these accounts position access as the second stage of the engagement pathway. Referral could open the door, but actual entry required alignment between proximity, zoning eligibility, visible information, and a welcoming environment.

Theme 3: Social fit shapes belonging

After crossing the physical threshold of reaching the AAC, participants encountered a further threshold: whether they could integrate and belong socially. Social fit was not simply about whether members were friendly. It reflected the fit between participants’ self-image, expectations, language, cultural background and the centre’s existing social norms. Belonging depended on whether participants could recognise the AAC as a place “for people like me”, enter existing social networks, and communicate comfortably within its dominant language and cultural environment.

Subtheme 3.1: Identity fit shaped whether the AAC felt “for people like me”

Belonging required participants to see themselves as the intended audience for the AAC. When the centre was perceived as oriented towards an older demographic than participants identified with, the label of active ageing could paradoxically create distance rather than attraction.

“Active ageing centre? You don’t make them old … The name itself … I tell you is wrong. It’s old people … not for people who think young.” [Participant 04].

This account suggests that some older adults resisted being positioned as “old” or passive, even when they were chronologically eligible for AAC services. For these participants, the issue was not simply stigma attached to ageing, but a mismatch between their self-image and what they perceived the AAC to represent.

A similar mismatch appeared when participants expected intellectual challenge or skill development but perceived the activities as too basic.

“I will never learn anything because I think I’m better than what they are going to teach me.” [Participant 04].

In this sense, activity fit was also identity fit: activities perceived as too basic could signal that the AAC was designed for “other” older adults, rather than for participants who saw themselves as active, capable, or seeking more cognitively stimulating engagement.

Subtheme 3.2: Established groups made newcomer entry socially difficult

Longstanding members formed close-knit social networks that newcomers found difficult to enter, leaving some new participants at the periphery.

“Because [they] are so used to seeing each other and doing activity together, they are closer than other newcomers. So we know ourselves, we are new, don’t disturb until they call us to go in to join them.” [Participant 05].

This account shows that newcomers did not only lack established relationships. They also regulated their own behaviour in deference to existing groups, waiting for an invitation that might not come. Without active facilitation, the AAC could be busy yet still socially difficult for newcomers to enter. This made belonging contingent not only on attendance, but on whether someone actively drew newcomers into the social life of the centre.

Some participants described these social boundaries as allowing only superficial contact rather than deeper connection.

“They will clique You don’t clique. Just hello. Goodbye.” [Participant 08].

This finding points to the importance of intentional social bridging, such as staff introductions, newcomer orientation, or peer ambassadors, so that attendance can progress into belonging.

Subtheme 3.3: Language and ethnic representation shaped whether social participation was possible

In Singapore’s multicultural context, language and ethnic mix shaped whether participants experienced AACs as inclusive. Participants from minority groups described arriving at centres where the dominant language limited their ability to converse and connect.

“If we speak English and Tamil, you see very … few Indians there, majority all Chinese … so if you talk to them, you don’t know who to talk to … you can see all of them talking in Mandarin, which I don’t understand.” [Participant 04].

This account suggests that inclusion was shaped not only by formal access to the centre, but also by the everyday language of social interaction. Even when minority participants were physically present, the dominant language environment could make it difficult to initiate conversations and develop a sense of belonging.

Taken together, these accounts position social fit as the third stage of the engagement pathway represented in Fig. 1. Belonging required alignment between participants’ identity, language, and preferences and those of the wider AAC community. When this alignment was absent, participants could remain socially peripheral despite having entered the centre. When social fit was achieved, attendance could progress towards sustained engagement.

Theme 4: Perceived value turns attendance into habit

Sustained AAC engagement depended on whether participants experienced ongoing value from participation. Participants continued attending when activities produced visible functional gains, emotional rewards, meaningful relationships, and routine. Over time, these returns shifted attendance from a deliberate choice into a self-sustaining habit.

Subtheme 4.1: Personally meaningful gains legitimised repeated attendance

Participants of AAC verbalized health benefits that were translated into improved daily capabilities.

“I also knew that I had improvement because I couldn’t walk down the staircase very well, so now at least I can. You know, they see … that radiance … if they can see the change, then it must be a good thing.” [Participant 08].

This account links AAC participation to a practical functional gain. Such improvement mattered because it made the value of participation visible to the participant. The AAC was not experienced as an abstract health-promoting setting but as a place that produced tangible changes in everyday functioning.

For some participants, attendance was understood as a form of preventive self-care aimed at maintaining health and reducing dependence on family members.

“I want to stay healthy. I don’t want to burden you all. If I get sick one day, the next day gone.” [Participant 10].

This account shows that the perceived value of attendance extended beyond immediate enjoyment. AAC participation was framed as a way of preserving independence and fulfilling a moral responsibility not to become a burden on family. This strengthened motivation for continued attendance because the AAC was positioned as part of the participant’s strategy for ageing well.

Emotional rewards of participating in AAC also facilitated sustained engagement with participants describing themselves as becoming happier and more socially expressive after attending AAC activities.

“I’m happier. I tell you …. I’m becoming chatty.” [Participant 05].

The value of attendance was therefore not confined to physical health or activity completion. It also lay in the experience of feeling more animated, connected, and socially present. These affective returns made attendance meaningful enough to repeat.

Subtheme 4.2: Social rewards deepened the emotional value of returning

Participants described relationships formed through repeated attendance as emotionally significant. These relationships did not appear to emerge from attendance alone, but from the regularity and structure of shared activities, which created opportunities for familiarity, trust, and attachment to develop.

“They are like family already. They have been [together] whole day. They have been spending there, you know, when they play Rummy-O.” [Participant 01].

This account shows how repeated participation transformed co-attendees from activity companions into a “family”. AAC had become more than a service location but rather a supportive community in which participants experienced belonging.

Intergenerational contact added another layer of emotional value as the presence of younger people made the centre feel lively and pleasant. Older participants described positive influence and enjoyment in the presence of younger people.

“Go there, see all the children’s running around, schoolchildren. To see them, it makes you happy.” [Participant 04].

This account suggests that intergenerational contact contributed to the atmosphere of the AAC. Rather than treating intergenerational interaction as a separate benefit, it can be understood as part of the broader emotional environment that made the AAC worth returning to.

Subtheme 4.3: Habit emerged when value became embedded in daily life

For some participants, sustained engagement reached a point where attendance no longer required active deliberation. The AAC had become incorporated into their daily rhythm. At this stage, participation appeared to shift from reflective motivation to a more automatic form of engagement, where attending was simply part of what the participant did.

“It’s like I cannot dream without [name of an AAC] now that you are asking this.” [Participant 05].

This quote illustrates the point at which AAC participation had moved beyond being an optional activity. The AAC had become embedded in the participant’s sense of self and everyday routine. This is important because habit represents a more durable form of engagement than initial motivation. While referral may prompt first attendance, and perceived benefits may encourage return visits, habit sustains participation when the AAC becomes part of the participant’s ordinary life.

Taken together, these accounts position perceived value as the final stage of the engagement pathway represented in Fig. 1. When referral, access and social fit had each been achieved, sustained engagement depended on participants’ ongoing experience of physical, emotional, social and intergenerational returns. Once these returns were realised, attendance became habitual, marking the transition from initial referral to ongoing participation.

Although Fig. 1 presents the themes sequentially for interpretive clarity, participants’ experiences did not always follow a fixed progression, and different stages could overlap. Participant 08, for example, reported tangible functional improvement from AAC participation, while also describing unresolved uncertainty regarding zoning eligibility and difficulty developing meaningful social connections within established groups. This within-participant account illustrates that perceived value could coexist with access and social-fit barriers, such that engagement could progress in some respects while remaining constrained in others.

Discussion

Principal findings

This study explored older adults’ experiences of social prescribing via referral to AACs from a public primary care setting in Singapore. Rather than describing engagement as a single decision to accept or reject referral, participants’ accounts suggested a staged, but overlapping engagement pathway rather than a strictly linear progression. Referral from a healthcare provider could legitimise participation and create an initial impetus to try. However, this initial motivation was insufficient on its own. Older adults then had to navigate practical access barriers, interpret the social environment of AACs, and decide whether continued participation felt personally meaningful and worthwhile. Overall, uptake and sustained engagement appear to be shaped less by individual capability alone, and more by the interaction between trusted referral, physical and social opportunity, and ongoing motivation.

Viewing the findings through this engagement pathway also helps locate potential points of breakdown. For participants who declined referral, engagement could be interrupted before first attendance when referral did not sufficiently resolve concerns about personal relevance, practical access, or anticipated fit. For those who discontinued after initial participation, barriers could arise after entry when a sense of belonging or ongoing value did not develop. These should not be understood as fixed group-specific trajectories as barriers could overlap across engagement statuses, and the small purposive subgroups did not support formal between-group comparisons.

These findings are particularly important in Singapore, where SP is being developed within a rapidly ageing, high-density, multicultural society and where AACs are positioned as key community nodes for healthy ageing and ageing in place. Public primary care clinics, with their longitudinal doctor-patient relationships, are well placed to legitimise and bridge patients into community-based support. In this sense, our findings provide empirical support for Roland et al.’s argument that social prescribing can reshape the clinical consultation by enabling physicians to respond to social, not only biomedical, needs through connection to community resources [16]. However, this bridge is incomplete unless AACs are also accessible, visible, socially inclusive and perceived as meaningful by diverse older adults.

Comparison with existing literature and contribution of this study

Consistent with current literature on the physical benefits of social group activities, our study participant verbalised improvement in health after joining AAC activities [17].

Our findings also corroborate existing social prescribing literature showing that engagement is shaped by trusted referral, accessibility, perceived need, perceived benefit, stigma, and the availability of supportive community resources. A recent UK qualitative systematic review found that patients’ responses to social prescribing varied according to readiness, support, temporal responsiveness and the degree of alignment between the referral and patients’ needs and expectations [18]. This resonates with our finding that social prescribing should not be understood as a single referral event, but as a staged engagement pathway in which referral legitimacy, accessible entry, social fit and perceived value each shape whether older adults accept, discontinue or sustain participation. Prior qualitative work has similarly found that trust in general practitioners, perceived need and benefits, navigator support, and service accessibility influenced engagement [19], while stigma and mismatched expectations could hinder participation [20]. Importantly, Pescheny et al. also highlighted the need to include patients who refuse or disengage from social prescribing. Our study addresses this gap by including older adults who were engaged, declined referral, or dropped out after initial participation.

This study extends existing literature in three ways. First, it shows how healthcare provider endorsement may operate in a primary care context. Referral did not merely signpost patients to a community resource; it legitimised AAC participation, framed it as personally relevant to health, and for some participants provided permission to try an unfamiliar community setting. Second, it adds nuance to the concept of access by showing that access was not limited to distance or transport, but also included zoning eligibility, programme visibility, environmental welcome, and ease of first entry. Third, it shows that sustained engagement depended not only on perceived benefits, but on whether AAC participation became socially rewarding, personally meaningful, and embedded into everyday routine. Our finding that sustained engagement depended on perceived value is consistent with qualitative syntheses showing that social prescribing can improve perceived wellbeing, reduce loneliness, and increase social connection, while also producing drawbacks or difficulties for some participants [21, 22]. Similar benefits have been reported in link worker social prescribing studies, including improved self-confidence, reduced social isolation, and positive changes in health-related behaviours [23]. Our study adds that perceived value is not simply an outcome of participation, but a mechanism for retention. Participants were more likely to remain engaged when AAC participation became enjoyable, meaningful, socially rewarding, and embedded into everyday routine.

Finally, our findings add local nuance to the equity dimension of social prescribing. Existing literature has raised concerns that social prescribing may not adequately reach ethnic minority communities when cultural, religious or language barriers are not addressed [24]. In our multicultural setting, language mismatch, limited perceived ethnic representation, established social cliques, perceived educational mismatch, and age-identity mismatch shaped whether AACs were experienced as welcoming spaces. These findings suggest that social prescribing should not be conceptualised as a single referral event, but as a pathway requiring trusted initiation, accessible entry, social belonging and sustained perceived value.

Interpretation through COM-B

To complement the data-derived engagement pathway, we interpreted the four themes through the Capability, Opportunity, Motivation–Behaviour (COM-B) model (Table 2). Interpreting the findings through COM-B highlights the relative importance of opportunity and motivation. Opportunity and motivation were particularly salient in participants’ accounts, whereas physical and cognitive capability appeared to be less prominent. Many were community ambulant, able to select activities according to their abilities, and did not describe difficulty understanding or participating in activities as the main reason for non-engagement. This is important because older adults’ non-participation in community programmes may easily be misread as frailty-related inability. Our findings suggest that, in this setting, the more important question is not simply whether older adults are capable of attending AACs, but whether the surrounding system makes engagement accessible, socially acceptable, and meaningful.

Table 2.

Interpretive mapping of the older adults’ engagement pathway to COM-B components

Data-derived engagement theme	Key processes identified in participants’ accounts	COM-B interpretation
Referral legitimises participation	A trusted healthcare professional created awareness of AACs, framed participation as relevant to the individual’s circumstances, and gave some participants confidence to try an unfamiliar service.	Reflective motivation: professional endorsement enhanced perceived relevance and anticipated benefit. Social opportunity: the referral relationship provided interpersonal encouragement and support for initial engagement.
Access shapes entry	Proximity, transport links, zoning eligibility, programme visibility, and the physical environment influenced whether a referral could translate into first attendance.	Physical opportunity: environmental and organisational conditions either enabled or constrained attendance.
Social fit shapes belonging	Identity fit, language, perceived cultural representation, established social groups, and the suitability of activities influenced whether participants felt able to integrate into the AAC community.	Social opportunity: everyday language, group norms, and opportunities for social inclusion shaped belonging. Reflective motivation: perceived fit with one’s identity, aspirations, and interests influenced willingness to continue.
Perceived value turns attendance into habit	Functional gains, emotional rewards, meaningful relationships, and enjoyable routines encouraged repeated attendance and, for some, made AAC participation part of everyday life.	Reflective motivation: anticipated and experienced benefits supported continued participation. Automatic motivation: enjoyment, emotional reward, and routine contributed to habit formation. Social opportunity: repeated interactions and relationships reinforced return attendance.

This table presents an interpretive mapping of the inductively developed themes to COM-B components. The components are not mutually exclusive, and individual themes may relate to more than one COM-B domain. Capability did not emerge as a dominant constraint within this sample; participants more commonly described structural, relational, and motivational influences on engagement

Physical opportunity was shaped by location, zoning rules, programme visibility, first impressions of the AAC, and the physical environment. Social opportunity was shaped by language, ethnicity, existing cliques, age identity, perceived educational mismatch and whether participants felt they could belong. Reflective motivation was shaped by whether older adults believed AAC participation would improve health, wellbeing or social life, while automatic motivation developed when participation became habitual, enjoyable and emotionally rewarding. These findings reframe social prescribing as more than a referral transaction. For referral to translate into sustained engagement, the community setting must provide a “receiving environment” that is easy to enter, socially inclusive, and responsive to the heterogeneity of older adults.

Implications for primary care, AACs and policy

For primary care providers, effective social prescribing requires more than identifying eligible patients and issuing a referral. Clinicians and link workers should personalise the referral by linking AAC participation to the patient’s own goals, such as staying active, reducing loneliness, maintaining independence, or developing meaningful routines. They should also anticipate common barriers by discussing location, language preference, activity interests and concerns about “not fitting in” before referral. Follow-up after the first visit, or within the first few weeks, may be particularly important because this appears to be the period when patients either begin to form a routine or disengage. Finally, care teams could consider using peer ambassadors or patient stories to normalise AAC participation and reduce stigma, especially among older adults who do not identify with the label of “active ageing” or who perceive AACs as a space for someone older or less independent than themselves.

At the programme and policy level, the findings highlight the need to design AACs as inclusive receiving environments for social prescribing. Zoning policies may need to be reviewed or applied flexibly, particularly when older adults wish to attend activities with friends or at centres that better match their interests, language, or accessibility needs. AACs could improve visibility and approachability by displaying activity schedules clearly, welcoming entry points, and conducting structured orientation for newcomers. To improve equity, programme planners should consider multilingual facilitation, culturally diverse activities, and deliberate strategies to include minority groups and socially peripheral newcomers.

The finding that the term “Active Ageing Centre” did not resonate with some participants merits attention. Chronological eligibility for an age-targeted service does not necessarily align with subjective age identity, aspirations, or desired level of independence. Previous research suggests that some older adults distance themselves from activities or services perceived as being “for old people” when these are associated with decline, dependency, or a social group with which they do not identify [25, 26]. In the present study, this mismatch appeared less as a rejection of activity or social connection itself than as a concern that the centre’s name, activities, and perceived clientele did not reflect how some participants saw themselves.

AACs could therefore consider messaging and programme designs that foreground interests, skills, contribution, wellbeing and social connection, rather than age eligibility alone. Wellness-oriented, skills-based and intergenerational offerings may broaden appeal among older adults who do not identify with age-labelled services. Such approaches should be co-designed and evaluated with diverse older adults, however, to ensure that they do not unintentionally privilege a narrow ideal of “successful” or highly independent ageing.

Implications for research

Future research should examine the full social prescribing pathway rather than referral rates alone, including first attendance, early drop-out, sustained engagement and reasons for disengagement. Our findings suggest that different mechanisms may operate at different stages: healthcare provider endorsement may influence initial uptake, access and visibility may determine entry, social fit may influence belonging, and perceived value may sustain long-term engagement.

Future work should also examine the provider and implementation side of social prescribing in Singapore’s primary care context. Since healthcare provider endorsement appeared to legitimise participation, further studies should explore clinicians’ and link workers’ attitudes, confidence, consultation practices, training needs and perceived barriers when discussing social prescribing with older adults. Implementation studies could also evaluate interventions targeting the specific barriers identified in this study, such as flexible zoning, clearer activity visibility, newcomer orientation, multilingual facilitation, peer ambassador models, intergenerational programmes, and wellness-oriented rebranding.

Future research should deliberately include older adults who are more comfortable communicating in languages other than English. Bilingual or multilingual interviewers, translated and culturally adapted study materials, and recruitment through community partners serving diverse linguistic groups could help examine how language concordance, peer composition, culturally meaningful activities and communication practices shape access and belonging within AACs.

Finally, future longitudinal and mixed methods studies should examine equity and outcomes. Such work could explore whether minority groups, less socially connected older adults, or those who do not identify with “active ageing” branding are more likely to decline or disengage. Health services research should evaluate whether sustained AAC engagement translates into improvements in loneliness, wellbeing, physical function, healthcare utilisation and cost-effectiveness. Existing reviews have noted a wide range of outcomes used in social prescribing research, reflecting both the promise and heterogeneity of the field [27]. A Singapore-specific outcomes framework would help clarify what successful social prescribing should mean for patients, primary care teams, AACs and the wider health system.

Strengths and limitations

This study has several strengths. First, it deliberately recruited participants across three engagement statuses, including those who were currently engaged, had declined referral, or had dropped out after initial participation. This enabled analysis of both facilitators and failure points in the social prescribing pathway, rather than focusing only on successful participation. Second, purposive maximum-variation sampling across engagement status and ethnicity allowed exploration of diverse patient perspectives within Singapore’s multicultural primary care context. Third, the use of COM-B provided a structured interpretive lens to examine how capability, opportunity and motivation shaped uptake and sustained engagement.

There are also limitations. The study was conducted in a single public primary care setting, which may limit transferability to other neighbourhoods, healthcare settings or AAC models in Singapore. Interviews were conducted in English, which may have under-represented older adults who are more comfortable in Mandarin, Malay, Tamil, or dialects. This is particularly important because language emerged as a potential barrier to AAC belonging. Although Malay participants were represented in the sample, only one Indian participant was interviewed. Consequently, the account of a Mandarin-dominant social environment should be interpreted as an important equity signal requiring further investigation, rather than as a generalisable account of Indian or Tamil-speaking older adults’ experiences across AACs. Although interviewers had no prior clinical relationship with participants, the involvement of primary care physicians from the same public primary care organisation may also have influenced the interview encounter. Participants may have perceived clinician interviewers as authority figures and may therefore have been less willing to express criticism of the referral pathway, AACs, or healthcare services. We considered this possibility during interpretation through reflexive memos and team discussions; nevertheless, social desirability cannot be excluded. Finally, this study focused on older adults’ perspectives and did not include caregivers, primary care providers, link workers or AAC staff. Incorporating these perspectives in future studies would provide a fuller understanding of how referral, onboarding, and sustained engagement are negotiated across the social prescribing pathway.

Conclusion

Social prescribing through referrals to AACs should be understood not as a single referral event, but as a staged engagement pathway. While trusted primary care providers can legitimise participation, sustained engagement depends on accessible entry, social belonging and perceived value. Strengthening social prescribing in Singapore will require attention not only to referral workflows, but also to the inclusiveness, visibility and relevance of AACs for diverse older adults.

Biographies

Deputy Clinic Head, Associate Consultant

MBBS, MMed(FM)

Clinician Researcher, Senior Consultant, Family Medicine Specialist

MBBS, MMed(FM), MCI, FCFP

House Officer

MBBS

Deputy Clinic Head, Consultant, Family Medicine Specialist

MBBS, MMed(FM), FCFP

Authors’ contributions

J.H.G, K.P.S, and S.Y.T conceptualised the study and designed the study and interview guide. J.H.G, K.P.S, S.Y.T and H.D.L.M conducted and coded the interviews. J.H.G, K.P.S, S.Y.T performed the thematic analysis. JHG drafted the initial manuscript. K.P.S and S.Y.T provided methodological and discussion input. K.P.S provided overall supervision and substantial critical revision of the manuscript, including in response to the editor’s and reviewers’ comments. All authors contributed to interpretation of the findings, reviewed and approved the final manuscript, and agree to be accountable for the accuracy and integrity of the work.

Funding

This research received no external funding.

Data availability

The datasets generated and/or analysed during the current study are not publicly available because the qualitative interview transcripts contain potentially identifiable participant information and were collected under ethics and consent conditions that do not permit public sharing. Anonymised excerpts supporting the findings are included in this article. Requests for access to additional anonymised data may be directed to the corresponding authors and will be considered subject to institutional and ethics approval.

Declarations

Ethics approval and consent to participate

Ethical approval was obtained from the NHG Health Domain Specific Review Board (DSRB) Ethics and Compliance Online System (ECOS) Reference: 2024–4452. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki and relevant institutional guidelines. Participation was voluntary, and written informed consent was obtained from all participants before interviews were conducted.

Consent for publication

Not applicable. This manuscript does not contain identifiable individual participant data.

Competing interests

The authors declare no competing interests.