Sleep disturbances impact the well-being of older adults, but there has been limited research on their longitudinal patterns and how they are affected by the quality of social relationships. This study explored longitudinal sleep disturbance patterns and examined the influence of social support and social strain on sleep disturbances in older adults using data from the Health and Retirement Study (HRS) spanning from 2010 to 2018. Group-based trajectory modeling identified distinct trajectory groups. Multinomial logistic regression and a fixed effects model investigated the relationship between sleep disturbances, social support, social strain, and sociodemographic factors. Three sleep disturbance trajectories were identified: high sleep disturbances, moderate sleep disturbances, and low sleep disturbances. Multinomial logistic regression showed that higher social support decreased the likelihood of classification in the high and moderate sleep disturbance groups. Conversely, higher social strain scores increased the likelihood of belonging to the high and moderate sleep disturbance groups. The fixed effects model confirmed the role of social support in predicting sleep disturbances over time, but social strain was not a significant predictor. These findings suggest the need for tailored interventions addressing social support to reduce sleep disturbances in this population.