Could your neighborhood or social connections impact your long-term health behaviors? While public health initiatives often treat health-risk behaviors as purely individual choices, a growing body of research suggests that our surroundings play a powerful and complex role. A new study published in the American Journal of Preventive Medicine suggests that risky behaviors among older adults—specifically substance misuse, tobacco use, and binge drinking—are deeply linked with a web of individual, interpersonal, neighborhood, and community-level environmental factors. Understanding these layered dynamics can help identify potential multilevel invention targets to make aging communities healthier.

What were the researchers studying and why?

Substance use among older adults is a serious and growing public health concern. Nationally, approximately 12% of older adults have used illicit drugs in the past year, 12% currently smoke cigarettes, and 13% have engaged in binge alcohol use within the past month. Older adults are uniquely vulnerable to the harmful effects of substance use due to age-related biological changes, chronic pain, medication interactions, and major life transitions (e.g., retirement or bereavement). Historical multilevel research has primarily focused on adolescents and young adults, leaving a substantial gap in our understanding of how social environmental ecosystems impact aging populations. Additionally, many studies examine each social factor individually—failing to capture how community conditions shape interpersonal relationships—which in turn shape individual experiences and behavior. This study addresses that gap by considering multiple dimensions of the social environment simultaneously and testing the hypothesis that older adults living in more supportive and cohesive social environments are less likely to engage in risky substance use.

How did the researchers conduct the study?

The researchers conducted a secondary data analysis to simultaneously evaluate the social environment across four distinct levels: Individual; Interpersonal; Neighborhood; and Community. They used a 2021 survey sample of almost 2,780 participants 50 years and older residing across seven mid-Atlantic states, oversampling rural residents and minoritized racial and ethnic subgroups. Generalized Structural Equation Modeling was used to examine both direct and indirect pathways connecting multilevel factors to substance use behaviors, while controlling for age, sex, race/ethnicity, marital status, income, education, and urban/rural status.

Using a socioecological framework, the team defined and measured key variables using established, validated scales linked to geo-coded administrative data to capture dimensions of the built and social environment:

• Risky Behavior Outcomes: Substance misuse of non-medical prescription or illicit drugs, tobacco use, and binge drinking measures were adapted from national surveillance instruments. 
• Individual Level: Loneliness (UCLA Loneliness Scale), resilience (Brief Resilience Scale), and perceived daily discrimination (Everyday Discrimination Scale).
• Interpersonal Level: Emotional social support (companionship and feeling loved).
• Neighborhood Level: Social cohesion broken into three subscales: attraction to neighborhood, acts of neighboring (interacting with residents), and sense of belonging.
• Community Level: Census-tract and county-level data assessing social environmental factors, including crime index, area deprivation index, and racial residential segregation (White/Black dissimilarity index).

What Did the Study Results Show?

The findings revealed a complex web of direct and indirect relationships. At the individual level, higher loneliness and greater perceived discrimination were each associated with an increased likelihood of tobacco use and binge drinking.  Also, lower resilience was linked to greater substance misuse. Perceived discrimination showed the most consistent relationship across all three outcomes.

Neighborhood social cohesion showed a striking paradox. Greater acts of neighboring and sense of belonging were directly and positively associated with tobacco use (incident rate ratio (IRR) = 1.2, p<0.05; and 1.3, p<0.01) and binge drinking (IRR = 1.2, p<0.01). Yet these same neighborhood factors were indirectly and negatively associated with all three risk behaviors through their effects on loneliness, resilience, discrimination, and emotional support. Together, these indirect pathways led consistently to lower likelihood of substance use with reductions ranging from 3% (p<0.05) to 9% (p<0.01).

At the community level, racial residential segregation was indirectly associated with increasing substance misuse by 12% (p<0.01), tobacco use by 4% (p<0.05), and binge drinking by 7% (p<0.05), mediated through greater perceived discrimination and lower neighborhood cohesion.

What is the potential impact of these findings?

This study makes a compelling case that substance use among older adults cannot be fully understood or effectively addressed by solely focusing on individual-level risk factors. The social contexts in which people age—from the cohesion of their neighborhoods to residential segregation—powerfully shape the risk of harmful substance use.

The direct link between neighborly acts and risky habits suggests that neighborhood social events may inadvertently encourage drinking or smoking; however, the underlying protective benefits of neighborhood social cohesion show that supportive local environments may shield individuals from the psychological distress that can drive substance use. The study underscores that racial residential segregation can directly damage physical health by fostering environments of discrimination, which increase the use of harmful substances.

Future research should transition from cross-sectional data to longitudinal designs and causal analyses. Investigators should also explore how these multilevel frameworks function across more racially and ethnically diverse cohorts. Ultimately, this research highlights a critical shift. To successfully curb substance risks in aging populations, we must look beyond individual clinical interventions and invest in macro-level strategies. These may include improving neighborhood walkability; increasing green space access; and building age-friendly, integrated communities that support thriving, healthy social lives.

Citation

Zhang X, Linn BK, Moss JL. A. (2026). A multilevel analysis of social environmental factors related to substance use, tobacco use, and binge drinking behaviors among older adults. American Journal of Preventive Medicine. Jan;70(1). doi: 10.1016/j.amepre.2025.108115.xs