DOES A PERCEIVED CONNECTION TO A NEIGHBORHOOD REDUCE LONELINESS?

Abstract This study investigated whether perceived neighborhood quality was associated with chronic loneliness for adults 60 and older in the United States. Although loneliness can be episodic and overcome, chronic loneliness has been identified as a social determinant of health. Utilizing ecological systems theory we hypothesized that higher levels of neighborhood social cohesiveness would be associated with lower odds of chronic loneliness. We postulated that the networks available to people in the proximal area where they live could provide social opportunities for reducing loneliness. This idea was consistent with prior findings indicating the salience of neighborhoods for retirees, but inconsistent with research indicating the importance of a confidant in reducing loneliness. Data from the 2008 and 2012 Health and Retirement Study Psychosocial Surveys were used (n = 3530). Loneliness was measured using the 3-item scale developed by Hughes and colleagues in 2004. Findings from unadjusted logistic regression indicated that loneliness was inversely related to neighborhood cohesion as measured by an index of the trustworthiness, friendliness and helpfulness of neighbors and cleanliness, occupancy, lack of graffiti, and sense of belonging in the area (OR = .73, p < .001). When demographic and health-related factors were entered into the model the odds of being lonely were significantly lower for those with higher ratings of social cohesion (OR = .83, p < .001). These findings were consistent with the idea that neighborhoods are an important social place for older persons and interventions at the neighborhood level may be more effective than individualized treatment plans.

relationship was found between rural and UC groups and loneliness, indicated by a three-item modified version of the R-UCLA loneliness scale. Rural residents reported lower rates of loneliness (3.84 out of 9) than did UC (4.61) or UA (4.64) residents. However, a significantly higher percentage of rural residents (44%) reported at least one instance of reinstitutionalization at 24 months compared to UC (36%) or UA (30%) residents. Multivariate analyses seek to clarify these contradictory results. The findings of this study have the potential to further inform the literature regarding loneliness and connections between reinstitutionalization among older adults living in rural and urban environments.

COASTAL RETIREMENT: IT'S ALL FUN AND GAMES UNTIL THE HURRICANE HITS
Anne P. Glass, 1 Jenni Blair, 1 and Judith Nichols 1 , 1.

University of North Carolina Wilmington, Wilmington, North Carolina, United States
Many people dream of retiring to the beach. Twenty-four individuals who retired from out-of-state to a beach area in southern North Carolina had previously been interviewed regarding their retirement process and decision to move to this destination. Hurricane Florence brought major flooding and devastation to the area in September 2018. Shortly after, 10 participants agreed to complete a second interview and data collection about their hurricane-related experiences. This sample consisted of 8 women and 2 men, average age of 74.4 (range=68-88), and all were white. Nine evacuated, including one who went to a shelter. This project provided a unique opportunity to compare answers about stress levels and how they felt about their choice to move to the area, before and after the storm. Six and five rated their stress high/ very high just prior to, and during Florence, respectively, but stress levels returned to low/very low for 90%. Two stated the storm caused them to rethink their decision to move; one now says she feels ambiguous about her move and would probably not choose it again. Stress caused by uncertainty was a thread across all interviews. Anxiety and concern were experienced, but no one reported fear. Neighbors played an important role pre-, during, and post-storm. No participants had significant damages, although one had a break-in; all expressed gratitude. They reported some lessons learned to apply the next time. These findings will be of interest to planners and others. They also demonstrate the resilience of older adults in dealing with natural disasters. This study investigated whether perceived neighborhood quality was associated with chronic loneliness for adults 60 and older in the United States. Although loneliness can be episodic and overcome, chronic loneliness has been identified as a social determinant of health. Utilizing ecological systems theory we hypothesized that higher levels of neighborhood social cohesiveness would be associated with lower odds of chronic loneliness. We postulated that the networks available to people in the proximal area where they live could provide social opportunities for reducing loneliness. This idea was consistent with prior findings indicating the salience of neighborhoods for retirees, but inconsistent with research indicating the importance of a confidant in reducing loneliness. Data from the 2008 and 2012 Health and Retirement Study Psychosocial Surveys were used (n = 3530). Loneliness was measured using the 3-item scale developed by Hughes and colleagues in 2004. Findings from unadjusted logistic regression indicated that loneliness was inversely related to neighborhood cohesion as measured by an index of the trustworthiness, friendliness and helpfulness of neighbors and cleanliness, occupancy, lack of graffiti, and sense of belonging in the area (OR = .73, p < .001). When demographic and health-related factors were entered into the model the odds of being lonely were significantly lower for those with higher ratings of social cohesion (OR = .83, p < .001). These findings were consistent with the idea that neighborhoods are an important social place for older persons and interventions at the neighborhood level may be more effective than individualized treatment plans.

COUPLES LIVING WITH ADVANCED CANCER: RAMIFICATIONS OF SOCIAL ISOLATION AND BENEFITS OF AN EMBEDDED SOCIAL NETWORK
Victoria H. Raveis, 1 Sheindy Pretter, 2 Monique Carrero-Tagle, 1 Daniel G. Karus, 1 and Avani Shah 3 , 1. New York University,New York,New York,United States,2. Touro College,New York,New York,United States,3

. University of Alabama, School of Social Work, Tuscaloosa, Alabama, United States
Cancer remains a leading cause of death, especially among older adults. While spouses are commonly involved in the provision of emotional and practical assistance to their ill spouse, their caregiving is not without cost. Although knowledge of an impending death permits preparation for the loss, a long and protracted illness, or one marked by intense caregiving demands, can deplete the well spouse's personal resources, increasing the risk of morbid bereavement outcomes. Well spouses (n=138), aged 50 and older (mean age 63.6), 41% male, providing 8+ hours of caregiving to a spouse with advanced cancer and a life expectancy of 6 months or less were followed over the terminal illness period. Caregiving spouses' anticipatory grief, depression and anxiety were all significantly, inversely correlated with sufficiency of social support, specifically tangible, informational and emotional