Parkinson’s disease (PD) can interfere with social functioning, which can impact the overall health of those with PD (Henry et al., 2016). Social functioning can be impacted by several factors, such as age, disease, mobility limitations, and cognitive impairment. Older adults often have a smaller social network and reduced closeness to network members compared to young and middle-aged adults, due to factors such as retirement and bereavement (Cornwell et al., 2008), as well as chronic health problems, sensory impairment, and income instability (National Academies of Sciences, 2020). In particular, mild cognitive impairment is common among people with PD, and those who are aware of their own cognitive decline may avoid social activities, especially highly demanding cognitive activities (Ho et al., 2021; Anderson, 2019) or activities requiring digital literacy (Zhu et al., 2023) due to embarrassment and stigmatization (Zhu et al., 2023). People with PD may also have difficulty recognizing emotions, and such social impairment is associated with greater PD burden (Coundouris et al., 2019). PD also leads to decreased mobility, and those who have mobility challenges tend to report social withdrawal due to a lack of required physical support, stigmatization, and embarrassment (Ahn et al., 2022; Nilsson et al., 2015). Physical function and social support appear to be related, as adults aged 18–64 years with mobility disability who were unemployed and in fair or poor health reported greater loneliness, more isolation, and less satisfaction with social activity (Hall et al., 2022). Given that social support plays a protective role in overall health (Frick et al., 2012), and that people with PD have poorer social functioning (Henry et al., 2016), it is important to understand the impacts of reduced social support in this population.

Social support protects overall health in several ways. Forms of social support can include emotional, informational (e.g., advice leading to a solution to problems), instrumental (e.g., practical help), and companionship support (Ferlander, 2007; Thoits, 2011). A greater number and quality of social ties may increase a sense of belonging to others, self-esteem, and a sense of control over life situations (Thoits, 2011). Stress-buffering effects of social ties may be the mechanism of action explaining better health outcomes with greater social support (Thoits, 2011). Social support offered as facts or recommendations from others–such as where to find less expensive and more convenient goods and services–enables subsequent behavioral changes that make everyday vexing circumstances more efficient (Thoits, 2011). Coping assistance strategies are stress buffers because they dampen situational demands and the person’s emotional reactions to these demands, which can reduce the physical and psychological consequences of the stressor (Thoits, 2011). Social support also buffers the impact of adversity (Thoits, 2011). The social resource model argues that social relationships are embedded within social ties and framed as social resources or social capital, which can demonstrate benefits to physical health outcomes (Song et al., 2021). While social roles for people with PD may change throughout the course of the disease, social support can benefit people with PD by encouraging health-promoting peer social norms, such as adequate educational programs and exercise, and discouraging health-damaging behaviors such as smoking, excessive eating, and drug use (Perepezko et al., 2019). Social ties can also facilitate access to healthcare resources, such as medical referrals, life skills training, education, or support groups (Ferlander, 2007). Therefore, given the importance of social support for health outcomes, examining the impact of different levels of social support on response to interventions is important for those with PD.

Group exercise improves physical function and social connectedness in people with PD (Hackney and Earhart, 2010; McKee and Hackney, 2013). This improved connection may stem from increased socialization among participants and instructors during group exercise (Mays et al., 2021). Thus, clinical trials of group exercise often control for socialization in the design of the intervention when contrasting lifestyle intervention effectiveness with a comparison group (Capili and Anastasi, 2023; McKee and Hackney, 2013; Hackney and Earhart, 2010). However, people with PD who are socially isolated are less physically active (Terracciano et al., 2023) and exhibit poorer physical function than socially connected older adults (Ahn et al., 2022). Therefore, baseline levels of social connectedness may influence the responsiveness of physical function outcomes to intervention. A better understanding of the role of social support in physical function responsiveness to exercise interventions among people with PD could provide insight into more effective treatment strategies.

The purpose of this exploratory analysis of an ongoing randomized controlled trial (RCT) was to examine the influence of 12 weeks of group exercise (tango or walking) on physical function in people with PD who reported low perceived social support compared to those with high perceived social support. As a secondary aim, we explored whether the type of intervention (tango or walking) influenced the impact of social support on physical function outcomes. We hypothesized that people with PD with high social support would exhibit greater improvements in physical function than the low social support group following 12 weeks of either tango or walking exercise. We also hypothesized that we would observe no differences between the tango or walking interventions for the low compared to the high social support groups on physical function.

To our knowledge, this is the first study to examine whether baseline social support levels influence potential benefits in physical function following therapeutic intervention among people with PD. We found that both high and low social support groups similarly improved in physical function following group exercise (tango or walking). Secondary analyses examining the potential influence of intervention type revealed that people with PD who had low baseline social support and received a walking intervention demonstrated more improvement in the number of chair stands than those who received tango in the high or low social support groups or those in the walking program with high social support.

Contrary to our hypothesis, we found that participants with either high or low social support exhibited largely similar physical function improvements. Across high and low social support groups, participants showed a trend toward improvement in the left tandem stance test. Additionally, both groups showed a clinically meaningful improvement in fast gait speed (>0.1 m/s; Palombaro et al., 2006) following the 12-week intervention. This finding is encouraging, as the intervention likely not only challenged physical function but also fostered social integration, social cohesion, and a sense of belonging (Hawkins et al., 2016) among participants and staff. This social cohesion may have been particularly important for those with low social support, as it could have provided them with the social resources they needed, greater resilience, and a greater propensity for improved physical function outcomes, which aligns with the social resource model (Song et al., 2021). However, it is possible that the low social support group did not socially connect during the intervention, and the improvements in physical function could be solely due to exercise. Nevertheless, our findings highlight the importance of therapeutic intervention in improving physical function regardless of social support in people with PD.

Our exploratory analysis of an ongoing RCT only found an improvement in one outcome measure (chair stand test) for participants in the low social support group receiving the walking intervention compared to the high social support group receiving the walking intervention or those with high or low social support receiving the tango intervention. While the dose, volume, frequency, intensity, and adherence of exercise were similar in the walking and tango interventions and led by the same intervention staff, differences in the level of social engagement or the specificity of training may explain these findings. For example, the topics of discussion during the tango program were likely centered on problem-solving when learning new tango moves, which may have increased cognitive demands for people with PD, especially those with concomitant mild cognitive impairment, as they have shown to have reduced attention capacity (Saunders and Summers, 2011). In contrast, the walking program likely involved socially engaging, participant-driven conversations. These differences in conversational content and social dynamics may explain the walking program’s superior improvement in physical function. We did not record the conversations, so this explanation remains speculative. Social engagement is a major motivator for older adults to engage in physical activity (Spiteri et al., 2019). Alternatively, the walking program may have more closely mimicked the biomechanical demands of the chair stand test, leading to greater improvements. Walking involves repetitive activation of lower body muscles (e.g., quadriceps, hamstrings, and glutes) in a rhythmic, weight-bearing manner, similar to the sit-to-stand motion, which requires concentric and eccentric contractions of these muscle groups. In contrast, tango dancing, while engaging lower body muscles, emphasizes complex, multidirectional movements (e.g., pivots, steps, and turns) that may not directly translate to the repetitive, linear motion of the chair stand test. The principle of training specificity suggests that exercises closely aligned with test movements yield greater improvements (Fleck and Kraemer, 2014). Nevertheless, the improvements in the number of chair stands were only observed in the walking group with low social support, and not high social support; thus, the specificity of training may not fully account for the different results. In sum, social connectedness and/or the type of exercise may have contributed to greater improvements in the number of chair stands in the participants with PD with low baseline social support in the walking program, compared to those with high baseline social support in the walking program or participants in the tango program.

Researchers have questioned why some participants respond to interventions while others do not. Prospective cohort data from people with PD document a robust predictive relationship between greater social support and better health outcomes (Terracciano et al., 2023; Ahn et al., 2022). Prospective longitudinal studies have demonstrated that both structural aspects (e.g., large social network size and cohesion) and functional aspects (e.g., high perceived emotional support) of social support are associated with better physical function among older adults with chronic kidney disease (Slaven et al., 2021). Social support has a robust relationship with improved health through projected security of surviving future health crises, reinforced positive health habits (e.g., healthy eating, exercise), and improved health through social ties (e.g., dampened stress; Ross and Mirowsky, 2002). Nonetheless, our preliminary findings provide evidence that low baseline social support may contribute to greater responsiveness to a walking intervention among people with PD.

Current research demonstrates that people with PD with mobility limitations are at risk of low social connectedness and tend to have poorer health outcomes (Ahn et al., 2022). However, our findings highlight that people with PD who have low social support exhibit similar improvements in physical function following an intervention as those with high social support. Our findings emphasize the importance of therapeutic intervention for people with PD who have varying levels of social support.

Since the present study is the first exploratory analysis examining the role of social support in exercise intervention outcomes for people with PD, larger and longitudinal trials are needed. We defined the level of social support based on the total MSPSS score, but it is unclear whether other aspects of social support not captured by this measure, such as the existence or quantity of social relationships, their formal structure (e.g., density or reciprocity), or the actual content of these relationships (e.g., positive or negative), drive differences in responsiveness of physical function measures to exercise. Future research could consider using scales that measure social networks, such as the Lubben Social Network Scale (Lubben, 1988). The average MSPSS score in the low social support group was 4.6/7 points, and all participants had the means to participate in this study. Possibly, those with very low social support may not have the necessary resources, such as time outside of work, to participate in an intervention. The interventions may have directly improved social network and physical function, or it is possible that the intervention had a positive effect on physiological, cognitive, or psychological function as well as symptoms of PD, which may also explain improvements in the high and low social support groups. Social support and lower physical function may also be correlated, and sometimes those who have the most room for improvement have the biggest gains after physical rehabilitation. Future work could consider examining the proportion of participants in trials who are recruited and retained who have low compared to high baseline social support. Future studies should also examine more objective measures of social inequalities that may influence social support, such as low socioeconomic status or other social determinants of health.

This exploratory analysis of an ongoing RCT found that people with PD who have high or low baseline social support similarly improved in physical function following a 12-week tango or walking intervention. These results may have implications for clinical care management and lifestyle interventions in people with PD. Given the broader relationship between social network and future health outcomes, clinicians should still monitor people with PD who have low social support, even if they respond similarly to intervention.