NSAIDs indicates nonsteroidal anti-inflammatory drugs. Adjusted probabilities were derived from multivariate logistic regression models adjusting for age, gender, race/ethnicity, education, and multimorbidity (2 or more medical conditions, including hypertension, coronary artery disease, heart failure, arthritis, cancer, metastatic cancer, diabetes, emphysema, asthma, prior stroke, or dementia). P values represent an overall test of trend. Bars represent 95% CIs. Loneliness is measured using the 3-item UCLA Loneliness Scale (range: 0-6 points; no loneliness: 0 points, mild/moderate loneliness: 1-3 points, high loneliness: 4-6 points). Medication classes were defined using a proprietary drug database (Lexicon Plus; Cerner Multum). Polypharmacy was defined as ≥5 medications (not including dietary supplements or vitamins).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Kotwal AA, Steinman MA, Cenzer I, Smith AK. Use of High-risk Medications Among Lonely Older Adults: Results From a Nationally Representative Sample. JAMA Intern Med. 2021;181(11):1528–1530. doi:10.1001/jamainternmed.2021.3775
Loneliness, the emotional distress resulting from a discrepancy between one’s actual and desired level of social connection, is associated with physical and psychological symptoms in older adults, including pain, insomnia, depression, and anxiety.1 The relationship of loneliness to these symptoms is likely bidirectional; in some situations it acts as a risk factor for the symptoms and in others it is the consequence of symptoms. In either case, lonely older adults may be at risk for using high-risk medications commonly prescribed for physical or psychological symptoms.2 Our objective was therefore to investigate the relationship between loneliness and high-risk medication use.2 A better understanding of this relationship might inform strategies for addressing symptoms and deprescribing potentially risky medications through the use of nonpharmacologic, social interventions.
We used cross-sectional data from the National Social Life, Health, and Aging Project (NSHAP), waves 1 through 3 (2005, 2010, 2015), an in-home nationally representative survey of community-dwelling adults older than 65 years.3 We included 6336 participants who responded to the NSHAP leave-behind questionnaire (85% response rate among 7045 total participants), and excluded an additional 254 participants (4%) with incomplete responses to the loneliness scale and 65 participants (1%) with missing medications data, resulting in a final sample of 6017 participants. Loneliness was categorized as none, low/moderate, or high based on the 3-item UCLA Loneliness Scale.4
In-home medication logs were conducted by asking participants to physically bring to the interviewer all medications they take “on a regular schedule, like every day or every week.” A clinical pharmacist reviewed unique drug entries and facilitated matching to a drug database and coding by type.5 We examined medications commonly prescribed for physical and psychological symptoms associated with loneliness and on the American Geriatrics Society Beers Criteria of potentially inappropriate medications,2 including opioids, nonsteroidal anti-inflammatory drugs (NSAIDs), benzodiazepines, antidepressants, and sleep medications. We examined polypharmacy (≥5 medications), which may result from accumulated medication burden. We additionally assessed medications used to treat common chronic conditions (lipid-lowering agents, antihypertensives, and salicylates) where we would not expect an association with loneliness.
We first determined the unadjusted association of loneliness with medication use using χ2 tests. We then used multivariable logistic regression to determine the probability of medication use by loneliness level after adjusting for age, sex, race/ethnicity, education, and multimorbidity. We did not adjust for physical or psychological symptoms given their high correlation with loneliness and because they may lie on the causal pathway. Statistical analyses were conducted using Stata 16.1, used national survey weights accounting for the likelihood of survey participation, and accounted for clustered responses among individuals participating in multiple waves of NSHAP. All respondents provided written informed consent and the protocol was approved by the institutional review boards at the University of Chicago and National Opinion Research Center (NORC).
The mean age of participants (SD) was 73 (7.1) years; 3243 (54%) were women; 4556 (84%) were non-Hispanic White; 2388 (40%) were classified as low/moderately lonely; and 396 (7%) were classified as highly lonely. In unadjusted analyses, loneliness was significantly associated with self-reported pain, insomnia, depression, anxiety, multimorbidity, and medications of interest (Table). After adjustment, loneliness was significantly associated with use of NSAIDs (no loneliness, 14%; 95% CI, 11%-16%; low/moderate, 17%; 95% CI, 14%-20%; high, 22%; 95% CI, 16%-28%), benzodiazepines (no loneliness, 5%; 95% CI, 4%-6%; low/moderate, 7%; 95% CI, 6%-9%; high, 11%; 95% CI, 7%-15%), anxiolytics/sedatives (no loneliness, 9%; 95% CI, 7%-10%; low/moderate, 12%; 95% CI, 10%-14%; high, 20%; 95% CI, 15%-25%), antidepressants (no loneliness, 14%; 95% CI, 12%-16%; low/moderate, 19%; 95% CI, 16%-21%; high, 27%; 95% CI, 21%-33%), and polypharmacy (no loneliness, 41%; 95% CI, 38%-43%; low/moderate, 44%; 95% CI, 41%-47%; high, 50%; 95% CI, 44%-56%) (Figure), and there was a nonsignificant trend for opioid use (no loneliness: 7%; 95% CI, 5%-8%; low/moderate, 7%; 95% CI, 6%-9%; high, 10%; 95% CI, 6%-14%).
In this nationally representative cohort of older adults, loneliness was a powerful predictor of use of medications used to treat physical and psychological symptoms. Loneliness was associated with higher pain medication use, including use of opioids and NSAIDs, and more than twice the frequency of use of antidepressants, sleep medications, and benzodiazepines. These medications are associated with adverse consequences among older adults, including opioid dependence, gastrointestinal bleeds, falls, fractures, delirium or cognitive impairment, new functional disability, and death. In circumstances in which loneliness is a risk factor for the development of physical or psychological symptoms, medications may not treat the underlying social experience of loneliness. In circumstances in which loneliness is a consequence of symptoms such as pain or depression, loneliness may amplify the intensity of these symptoms. In both circumstances, clinicians should consider initiating social interventions for lonely older adults or “social prescribing” to local community-based support programs.6 Identifying and addressing loneliness may have the added benefit of allowing clinicians to reduce or avoid prescription of high-risk medications. A primary limitation of this study is the cross-sectional analysis, which limits our ability to draw causal conclusions on the directionality of our findings.
Accepted for Publication: May 28, 2021.
Published Online: July 26, 2021. doi:10.1001/jamainternmed.2021.3775
Corresponding Author: Ashwin A. Kotwal, MD, MS, Division of Geriatrics, Department of Medicine, University of California, San Francisco School of Medicine, 4150 Clement St, 181G, San Francisco, CA 94121 (firstname.lastname@example.org).
Author Contributions: Dr Kotwal had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Kotwal, Steinman.
Acquisition, analysis, or interpretation of data: Kotwal, Cenzer, Smith.
Drafting of the manuscript: Kotwal.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Kotwal, Cenzer.
Obtained funding: Smith.
Supervision: Steinman, Smith.
Conflict of Interest Disclosures: Dr Kotwal reported grants from the National Institute on Aging, the National Palliative Care Research Center, and the Hellman Family Foundation during the conduct of the study; and grants from Metta Fund and Humana Inc outside the submitted work. Dr Steinman reported grants from the National Institutes of Health Research Center, Network, and K24 during the conduct of the study; personal fees from UpToDate Royalties for authoring a chapter in UpToDate and personal fees from American Geriatrics Society Honoraria for serving as co-chair of the American Geriatrics Society Beers Criteria Guideline Panel outside the submitted work. No other disclosures were reported.
Funding/Support: The National Social Life, Health, and Aging Project is supported by the National Institute on Aging and the National Institutes of Health (R01AG021847; R01AG043538; R01AG033903; R01AG048511; R37AG030481). Dr Kotwal’s effort on this project was supported by grants from the National Institute on Aging (K23AG065438; R03AG064323), the NIA Claude D. Pepper Older Americans Independence Center (P30AG044281), the National Palliative Care Research Center Kornfield Scholar’s Award, and the Hellman Foundation Award for Early-Career Faculty. Dr Kotwal reports grants from Metta Fund and from Humana Inc outside the submitted work. Dr Smith was supported by R01AG057751 and K24AG068312 from the National Institute on Aging. Dr Steinman was supported by K24AG049057, P30AG044281, and R24AG064025 from the National Institute on Aging.
Role of the Funder/Sponsor: The National Institute on Aging and the National Institutes of Health had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Additional Contributions: We would like to thank the National Opinion Research Center (NORC), which was responsible for data collection.