eMethods. Supplementary methodological description
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.
Talcott WJ, Yu JB, Gross CP, Park HS. Social Connectedness Among Medicare Beneficiaries Following the Onset of the COVID-19 Pandemic. JAMA Intern Med. 2021;181(9):1245–1248. doi:10.1001/jamainternmed.2021.2348
Social distancing is an effective strategy to limit contagion and mortality from the COVID-19 pandemic. However, these measures may also decrease perceived social connectedness and conversely increase social isolation, states which are associated with psychologic and physiologic morbidity.1-5 Patients who are elderly or have disabilities are particularly encouraged to practice social distancing given their higher risk of severe COVID-19 infection, but they also have a higher baseline risk for reporting social isolation.5,6 We hypothesized that the pandemic and social distancing measures have negatively affected perceptions of social connectedness among these high-risk patients.
The Medicare Current Beneficiary Survey is an in-person, nationally representative survey of Medicare beneficiaries that is sponsored by the Centers for Medicare & Medicaid Services. We used the COVID-19 Summer 2020 Supplement data collected from June 10, 2020, to July 15, 2020. The institutional review board of Yale University approved the study and granted a waiver of informed consent because of the use of publicly available, deidentified data. The study cohort consisted of 9634 respondents who represented a weighted 50 851 437 community-dwelling beneficiaries who reported on changes in social connectivity during the pandemic (eMethods in the Supplement). The outcome variable, reporting decreased social connectedness, was defined as a beneficiary stating that that they felt “less socially connected to family and friends” since the onset of the pandemic.
We selected variables that we hypothesized were potentially associated with decreased social connectedness (eMethods in the Supplement). These variables included demographic and socioeconomic characteristics, adherence to 6 specific COVID-related social distancing precautions, ownership and use of technologies that allow for remote social engagement, and medical comorbidities and care receipt that may make beneficiaries more dependent on social contacts for health reasons. We used univariable and multivariable logistic regression to identify factors associated with reported decreased social connectedness since the start of the pandemic. Predicted probabilities (PPs) were estimated using population-averaged estimates, with 95% CIs calculated using the δ method. Statistical analyses were conducted using Stata, version 13.1 (StataCorp), and statistical significance was determined from contrast estimates using the χ2 statistic. The multivariable model included variables reaching a significance level of P < .10 on univariable analysis.
Since the start of the COVID-19 outbreak, a weighted 36.7% of enrollees reported feeling less socially connected with friends and family. Decreased social connection was reported by 41.2% of female, 31.2% of male, 24.6% of Black non-Hispanic, and 38.9% of White non-Hispanic beneficiaries (Table 1). On multivariable analysis (Table 2), demographic characteristics that were independently associated with feeling less socially connected included female (PP, 40.4%; 95% CI, 38.6%-42.2%) vs male sex (PP, 33.2%; 95% CI, 31.2%-35.1%; P < .001), White non-Hispanic (PP, 38.1%; 95% CI, 36.6%-39.7%) vs Black non-Hispanic race/ethnicity (PP, 30.4%; 95% CI, 25.9%-34.9%; P < .001), income of $25 000 or greater (PP, 38.5%; 95% CI, 36.8%-40.1%) vs less than $25 000 (PP, 34.2%; 95% CI, 31.4%-37.1%; P = .02), no Medicaid eligibility (PP, 38.1%; 95% CI, 36.5%-39.6%) vs full Medicaid dual eligibility (PP, 28.6%; 95% CI, 23.5%-33.7%; P = .002), history of non–skin cancer (PP, 40.8%; 95% CI, 38.0%-43.6%) vs none (PP, 36.3%; 95% CI, 34.9%-37.7%; P = .005), and depression (PP, 41.6%; 95% CI, 39.0%-44.3%) vs none (PP, 35.5%; 95% CI, 33.9%-37.1%; P < .001). Behaviors that were associated with feeling less socially connected included performing all 6 assessed anti–COVID-19 social distancing measures (PP, 38.8%; 95% CI, 37.4%-40.2%) vs 4 or fewer (PP, 30.5%; 95% CI, 26.2%-34.8%; P < .001), computer ownership (PP, 38.4%; 95% CI, 36.6%-40.2%) vs not (PP, 34.4%; 95% CI, 32.2%-36.6%; P = .01), and use of internet teleconferencing software (PP, 40.2%; 95% CI, 38.5%-42.0%) vs not (PP, 33.3%; 95% CI, 31.1%-35.5%; P < .001). Among other variables, metropolitan (PP, 37.6%; 95% CI, 36.1%-39.1%) vs nonmetropolitan residence (PP, 35.9%; 95% CI, 33.3%-38.5%; P = .23) and geographic region were not independently associated with reported social connectivity.
More than one-third of Medicare beneficiaries reported feeling less socially connected to friends and family since the start of the COVID-19 pandemic. Medicare beneficiaries who were women, had higher incomes, were not of Black non-Hispanic race/ethnicity, and had a history of cancer or depression were more likely to report a negative association of the pandemic with perceptions of social connectedness. The likelihood of reporting decreased social connection was widespread nationally and was associated with practicing more social distancing measures. The limitations of this study include an inability to assess the magnitude of the decreased connection reported and survey exclusion of the subset of Medicare-beneficiaries who are not community dwelling. The public health benefits and psychosocial costs of prolonged social distancing measures should be balanced carefully in this doubly vulnerable population.
Accepted for Publication: April 10, 2021.
Published Online: May 28, 2021. doi:10.1001/jamainternmed.2021.2348
Corresponding Author: Wesley John Talcott, MD, MBA, Smilow Cancer Hospital, PO Box 208040, New Haven, CT 06520-8040 (email@example.com).
Author Contributions: Dr Talcott 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: Talcott, Gross, Park.
Acquisition, analysis, or interpretation of data: Talcott, Yu.
Drafting of the manuscript: Talcott.
Critical revision of the manuscript for important intellectual content: Yu, Gross, Park.
Statistical analysis: Talcott.
Conflict of Interest Disclosures: Dr Yu reported personal fees from Boston Scientific and Galera Pharmaceuticals outside the submitted work. Dr Gross reported grants from the National Comprehensive Cancer Network (Pfizer/AstraZeneca), Johnson & Johnson, and Genentech, and travel/speaking fees from Flatiron outside the submitted work. No other disclosures were reported.