Sex-based differences in the association between loneliness and polypharmacy among older adults in Ontario, Canada.

BACKGROUND

Emerging evidence shows loneliness is associated with polypharmacy and high-risk medications in older adults. Despite notable sex-based differences in the prevalence in each of loneliness and polypharmacy, the role of sex in the relationship between loneliness and polypharmacy is unclear. We explored the relationship between loneliness and polypharmacy in older female and male respondents and described sex-related variations in prescribed medication subclasses.

METHODS

We performed a cross-sectional analysis of representative data from the Canadian Community Health Survey-Healthy Aging cycle (2008/2009) linked to health administrative databases in Ontario respondents aged 66 years and older. Loneliness was measured using the Three-Item Loneliness Scale, with respondents classified as not lonely, moderately lonely, or severely lonely. Polypharmacy was defined as five or more concurrently-prescribed medications. Sex-stratified multivariable logistic regression models with survey weights were used to assess the relationship between loneliness and polypharmacy. Among those with polypharmacy, we examined the distribution of prescribed medication subclasses and potentially inappropriate medications.

RESULTS

Of the 2348 individuals included in this study, 54.6% were female respondents. The prevalence of polypharmacy was highest in those with severe loneliness both in female (no loneliness, 32.4%; moderate loneliness, 36.5%; severe loneliness, 44.1%) and male respondents (32.5%, 32.2%, and 42.5%). Severe loneliness was significantly associated with greater adjusted odds of polypharmacy in female respondents (OR = 1.59; 95% CI: 1.01-2.50) but this association was attenuated after adjustment in male respondents (OR = 1.00; 95% CI: 0.56-1.80). Among those with polypharmacy, antidepressants were more commonly prescribed in female respondents with severe loneliness (38.7% [95% CI: 27.3-50.0]) compared to those who were moderately lonely (17.7% [95% CI: 9.3-26.2]).

CONCLUSIONS

Severe loneliness was independently associated with polypharmacy in older female but not male respondents. Clinicians should consider loneliness as an important risk factor in medication reviews and deprescribing efforts to minimize medication-related harms, particularly in older women.