Longitudinal Associations Between Loneliness and Prescription Medication Use

J Gerontol B Psychol Sci Soc Sci. 2023 Apr 1;78(4):730-735. doi: 10.1093/geronb/gbad010.


Objectives: Both loneliness and the use of psychotropic drugs are common in later life. Although loneliness has been found to be associated with psychotropic drug use, most studies have been cross-sectional, and we know less about their longitudinal associations.

Methods: Drawing on five waves of data from the Health and Retirement Study and two statistical approaches (fixed-effects and cross-lagged panel models), we examine longitudinal associations between loneliness and the use of prescription pain and depression/anxiety medications.

Results: Across 57,654 observations among 20,589 respondents, 22.8% reported regular use of pain prescription medications, 17.8% regular use of depression/anxiety prescription medication, and 15.6% feeling lonely in the past week. Loneliness and the use of depression/anxiety medications were associated according to both modeling approaches, net of covariates. In years when a respondent reported feeling lonely, the odds of regular use of depression/anxiety medications were 1.42 times higher (p < .001) than in years when they did not feel lonely. Regarding reciprocation, odds of regular depression/anxiety medication use in a given wave range from 1.3 to 1.5 times higher if loneliness was reported in the prior wave. Likewise, the odds of reporting loneliness in a given wave range from 1.5 to 1.8 times higher if regular depression/anxiety medication was reported in the prior wave.

Discussion: Prior loneliness predicts contemporaneous regular use of depression/anxiety prescription medications. Although this confirms the directional association found in prior studies, we found prior use of depression/anxiety medications is also associated with increased odds of loneliness, suggesting further research is needed to understand mechanisms that explain their associations and potential interventions.

Keywords: Health and Retirement Study; Loneliness; Longitudinal study; Medication use.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Cross-Sectional Studies
  • Depression* / drug therapy
  • Depression* / epidemiology
  • Humans
  • Loneliness*
  • Longitudinal Studies
  • Pain
  • Prescriptions