STRENGTH AND LIMITATIONS OF THIS STUDY
-
This study focuses on a nationally representative sample of Medicare beneficiaries in the USA, and assesses the trends in forgone medical care and unmet needs among Medicare beneficiaries with a history of depression during the COVID-19 pandemic.
-
We use three rounds of survey data from the Medicare Current Beneficiary Survey COVID-19 Supplement Public Use Files from Summer 2020, Fall 2020 and Winter 2021 to address the above study aims.
-
The current study is limited in that the survey included information on whether the respondent had a history of depression, but no further information including the timing or severity of the diagnosis.
-
The cross-sectional data structure limited our ability to track individuals over time, and we could not include all the potential factors as covariates in our models.
-
We assessed time trends of outcomes for the period of Summer 2020 to Winter 2021 and further analyses are needed to elucidate the trends in the recent times.
Introduction
Depression is a common mental illness in older adults, with prevalence rates of 13%–22% in the USA,1 2 and is a leading cause of disability and a major contributor to suicide deaths according to the WHO.3 It is also well documented that depression is linked with significant loss of cognitive and social functioning, reduced productivity, and poor physical health and quality of life.4–8 Despite their vulnerabilities, older adults with depression have been more likely to report unmet healthcare needs compared with those without depression due to limited accessibility and availability, high costs of care, and concerns of stigma.9 10 Older adults with depression have also encountered greater risks of financial insecurity, poor nutritional status and social isolation.11 12
The COVID-19 pandemic that started in 2019 and was declared on 11 March 202013 has created significant challenges in daily lives such as reduced access to healthcare, food insecurity and housing stress, especially among vulnerable populations.14–16 The COVID-19 pandemic has also caused disruption of social connectedness as well as psychological distress (eg, stress, fear, anxiety and loneliness).17 18 These adverse impacts of COVID-19 may have disproportionately affected older adults with depression.
Previous studies related to COVID-19 and investigating the association of history of depression or depressive symptoms with forgone medical care as well as disruptions to basic needs, financial situation and mental health have focused on early period of the COVID-19 pandemic.19–21 For example, most relevant to our study, a study by Balasuriya et al19 analysed the ability to access healthcare, social and financial needs, mental health, source of COVID-19-related information, and protective measures during the COVID-19 pandemic among Medicare beneficiaries with and without a history of depression. However, they focused on the initial stage of the COVID-19 pandemic (Summer 2020). We expanded on this prior work by examining trends in these outcomes using more recent rounds of data collected during the COVID-19 pandemic.
Specifically, in this study, we aimed to investigate forgone medical care, basic needs disruption, financial and mental health disruption, disengagement of preventive behaviours guidelines, and perceived severity of COVID-19 among Medicare beneficiaries (individuals 65 years and above or individuals under 65 years with disability) with and without a self-reported history of depression throughout the period of Summer 2020 to Winter 2021. We hypothesised that Medicare beneficiaries with a history of depression were more likely to have forgone medical care, basic needs disruption, and financial and mental health disruption than Medicare beneficiaries without a history of depression throughout the COVID-19 pandemic. We also hypothesised that Medicare beneficiaries with a history of depression tended to be more disengaged in preventive behaviours and to perceive COVID-19 less severely compared with those without a history of depression during the pandemic.
Methods
Study design, setting and data source
This is a secondary analysis of data obtained from three rounds of repeated cross-sectional survey from the Medicare Current Beneficiary Survey (MCBS) COVID-19 Supplement Public Use Files (PUF) which were administered as a supplement to the main MCBS: the Summer 2020 (administered from June to July 2020), Fall 2020 (administered from October to November 2020), and Winter 2021 (administered from March to April 2021) Supplement PUFs. The MCBS, sponsored by the Centers for Medicare & Medicaid Services (CMS), is a nationally representative survey of Medicare beneficiaries that has been conducted for over 30 years to aid the CMS in assessing and evaluating the Medicare programme. The MCBS provides a variety of information about the beneficiaries including sociodemographic characteristics, health status and healthcare access, utilisation, and expenditure and it has been widely used in the literature.22
The MCBS COVID-19 Supplement PUFs were released based on rapid response surveys as a supplement to the main MCBS in response to the emergence of the COVID-19 pandemic in the USA to understand the impact of COVID-19 on the lives of Medicare beneficiaries. Survey topics included forgone medical care, access to telehealth, preventive measures, COVID-19 testing and vaccination, and well-being of the beneficiaries since the onset of the pandemic. The survey was administered three times (summer and fall 2020 and winter 2021) by telephone on existing MCBS sample members living in the community in a way that the sample represents a randomly selected cross-section of all Medicare beneficiaries with sampling weights.23 The eligibility criteria for the survey of COVID-19 Supplement PUFs are beneficiaries who were continuously enrolled in Medicare from the beginning of 2020 to the time of the interview, were still alive, and were living in the community. The survey was combined with the demographic and health status data collected from the MCBS respondents during most recent prior interviews. Additional information related to the survey is available on the CMS MCBS website.24
Study sample
The sample consisted of Medicare beneficiaries (individuals 65 years of age and above or individuals under age 65 with disability) residing in communities in the USA and have been continuously enrolled in Medicare from the beginning of 2020 and at the time of their interview. These individuals were included in the study based on their provision of valid data and samples with missing values in key variables were excluded from the analyses (see the online supplemental figure S1 for flow chart of sample exclusion procedure).
Supplemental material
Variables
The primary outcome variable was the forgone medical care because of the pandemic, defined as 1 if a respondent answered ‘yes’ to the question ‘since the onset of the pandemic, did [you/sample person (SP)] need medical care for something other than coronavirus, but not get it because of the coronavirus pandemic?’ and 0 if the respondent answered ‘no’.
Secondary outcome variables were disruption of basic needs, disruption of financial security and mental health, during the pandemic. Basic needs disruption was defined as 1 if a respondent answered ‘unable’ to any of the following questions: since the onset of the pandemic, ‘[have you/has SP] been able to pay rent/mortgage?’, ‘[have you/has SP] been able to get medication?’, ‘[have you/has SP] been able to get doctor appointment?’, ‘[have you/has SP] been able to get food wanted?’, or ‘[have you/has SP] been able to get household supplies, such as toilet paper?’; and 0 if the respondent answered ‘able’ or ‘have not needed’ to all the questions.
The binary outcome for financial and mental health disruption during the pandemic was defined as 1 if a respondents answered ‘less’ to the following questions: since the onset of the pandemic, ‘have you felt financially secure?’ or ‘have you felt socially connected to family and friends?’, or answered ‘more’ to the following questions: since the onset of the pandemic, ‘have you felt stressed?’ or ‘have you felt lonely or sad?’; and 0 if the respondent answered ‘more’ or ‘about the same’ to the questions: ‘have you felt financially secure?’ or ‘have you felt socially connected to family and friends?’, or answered ‘less’ to the following questions: ‘have you felt stressed?’ or ‘have you felt lonely or sad?’
Additional secondary outcomes included engagement of essential preventive behaviours and perceived severity of COVID-19. Engagement of essential preventive behaviours was defined as 1 if a respondent answered ‘yes’ to all the following three questions: ‘[have you/has SP] done any of the following in response to the outbreak of the new coronavirus? (a) washed [your/his/her] hands for 20 s with soap and water; (b) wore a facemask when out in public; and (c) kept a six-foot distance between [yourself/himself/herself] and people outside [your/his/her] household.’ Lastly, perceived severity of COVID-19 was measured with two binary variables, defined as 1 if a respondent answered ‘strongly agree’ or ‘agree’ to the statements (1) ‘coronavirus is more contagious than the influenza.’; (2) ‘coronavirus is more deadly than the influenza.’, respectively, and 0 if the respondent answered ‘neither agree nor disagree’, ‘disagree’, or ‘strongly disagree’. Perceived severity of COVID-19 was only available in two recent rounds (Fall 2020 and Winter 2021). Outcomes coded with ‘do not know’ and ‘refused’ were excluded from the analyses.
The key independent variable was a self-reported history of depression, which was defined as 1 if a respondent answered ‘yes’ to the following question: ‘has a doctor or other health professional ever told [you/SP] that [you/he/she] had depression?’, and 0 if the respondent answered ‘no’.
Sociodemographic and clinical characteristics selected based on prior literature were included as covariates in all models.19 22 Sociodemographic characteristics include race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, or other/unknown), age group (<65, 65–74, or>74 years), sex (male or female), annual income (< US$25 000 or ≥ US$25 000), non-English language spoken at home (yes or no), Medicaid dual eligibility in 2019 (no, fully dual eligible, partially dual eligible, or Qualified Medicare Beneficiary eligible only), primary location of healthcare received (physician’s office/medical clinic, managed care plan centre, other clinic, urgent care/hospital/emergency room, Veterans Affairs facility, or other type of location), Metropolitan residence (yes or no), and geographical region (Northeast, Midwest, South, or West). Clinical characteristics included binary variables for whether the respondent had each of the set of comorbidities (immunodeficiency, myocardial infarction, congestive heart failure, stroke, cancer, dementia, asthma/chronic obstructive pulmonary disease (COPD), and diabetes).
Statistical analysis
We assessed differences in outcomes and sociodemographic and clinical characteristics by self-reported history of depression of beneficiaries, based on Wald 
tests. Weighted multivariable logistic regression models were used to estimate the association of history of depression with outcomes of interest. Models were adjusted for the set of sociodemographic and clinical characteristics described above and analyses were estimated with the MCBS sampling weights to make results nationally representative of the population of Medicare beneficiaries. Any missing values of outcome variables and covariates were excluded from the statistical analyses. All estimates were presented as adjusted ORs and 95% CIs. The weighted multivariable logistic regression was also used to estimate the predicted probabilities (PPs) of outcomes using population-averaged estimates for each round to investigate whether the association persisted over time during the study period. P values were two-tailed, and statistical significance of 0.05 was used. Data were analysed using Stata statistical software, V.16.0 (StataCorp).25 This study is based on the secondary analysis of existing publicly available survey data that do not require Institutional Review Board review. We followed the Strengthening the Reporting of Observational Studies in Epidemiology reporting guidelines.26
Patient and public involvement
Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Results
Our final study sample included 28 480 community-dwelling Medicare beneficiaries who were interviewed in Summer 2020, Fall 2020, or Winter 2021 and had complete information (table 1). Of these beneficiaries, 7629 (27%) had a history of depression, whereas 20 851 (73%) did not. Forgone medical care rate was higher in those who reported a history of depression than those who did not report it (14.7% vs 10.3%). Compared with those without a history of depression, those with a history of depression tended to be younger and female, to have annual income less than US$25 000, to speak non-English language at home, to be Medicare-Medicaid dual eligible, and to have other history of diseases such as any immunodeficiency, myocardial infarction, congestive heart failure, stroke, dementia, asthma/COPD and diabetes (table 1).
Descriptive characteristics of participants by self-reported history of depression
After adjusting for sociodemographic and clinical characteristics, the weighted multivariable logistic regression models estimated that Medicare beneficiaries with a history of depression had higher odds of foregone medical care (OR 1.29, 95% CI 1.16 to 1.42, p<0.001), of basic need disruption (OR: 1.49, 95% CI 1.33 to 1.67, p<0.001), and of financial and mental health disruption (OR: 1.75, 95% CI 1.60 to 1.91, p<0.001) during the COVID-19 pandemic when compared with those without a history of depression (table 2; see online supplemental table S1 for detailed regression results). The average PPs of self-reported forgone medical care showed that Medicare beneficiaries with a history of depression had consistently higher predicted self-reported forgone medical care (Summer 2020, PPs: 24% vs 19%; Fall 2020, PPs: 10% vs 7%; Winter 2021, PPs: 9% vs 5%) than those without a history of depression during the pandemic (figure 1). Similarly, we also found that beneficiaries with a history of depression had consistently higher predicted basic needs disruption (Summer 2020, PPs: 27% vs 19%; Fall 2020, PPs: 17% vs 8%; Winter 2021, PPs: 13% vs 6%) as well as financial and mental health disruption (Summer 2020, PPs: 71% vs 60%; Fall 2020, PPs: 71% vs 55%; Winter 2021, PPs: 71% vs 56%) potentially caused by COVID-19 than those without a history of depression throughout the pandemic (figure 2).
Average predicted probabilities of self-reported forgone medical care because of COVID-19, % (Summer 2020, Fall 2020 and Winter 2021).
Average predicted probabilities of disruption to basic needs and financial and mental health caused by the pandemic, % (Summer 2020, Fall 2020, and Winter 2021).
Association of self-reported history of depression with foregone medical care and disruption to basic needs and financial and mental health caused by COVID-19
Engagement of essential preventive behaviours was not significantly different between beneficiaries with and without a history of depression in terms of washing hands properly (OR: 0.94, 95% CI 0.76 to 1.17, p=0.59), wearing a facemask in public (OR: 0.99, 95% CI 0.74 to 1.32, p=0.92), and keeping a six-foot distance (OR: 1.00, 95% CI 0.85 to 1.17, p=0.98) (table 3). Similarly, having high perception of COVID-19 severity was not significantly different between beneficiaries with and without a history of depression in terms of agreeing that COVID-19 is more contagious than the influenza (OR: 1.03, 95% CI 0.89 to 1.20, p=0.67) and that COVID-19 is more deadly than the influenza (OR: 1.01, 95% CI 0.88 to 1.15, p=0.92) (table 3).
Association of self-reported history of depression with engagement of essential preventive behaviours and perceived severity of COVID-19
Discussion
Our results indicated that the COVID-19 pandemic had affected Medicare beneficiaries with a history of depression in a disproportionate way throughout the pandemic, who had consistently experienced forgone medical care, basic needs disruption and financial and mental health disruption more often than those without a history of depression throughout the pandemic. However, engagement of essential preventive behaviours and perception of COVID-19 severity were not significantly different between Medicare beneficiaries with and without a history of depression.
Our findings of the disparity in access to healthcare during COVID-19 pandemic are consistent with findings of studies before the pandemic, which documented that older adults with mental illness greatly underutilised health services because of shortages of mental health providers, high cost of care, lack of perceived need for mental health service, patients’ own negative views about mental illness, and concerns of stigma and discrimination.27–32 During COVID-19 pandemic, people who are older or have disabilities may have faced additional barriers to accessing needed healthcare due to closed doctor’s office, difficulties in using telehealth, fear of COVID-19 infection in medical settings, and limited transportation.33–35 Although a previous study has identified challenges in accessing medical care among Medicare beneficiaries with depression in the early period of the pandemic,19 our findings provide an evidence of the persistent gap across all study periods in accessing medical care faced by Medicare beneficiaries with a history of depression even after the trends in forgone care decreased overall,36 implying continuing difficulties among these vulnerable populations with a history of depression.
Medical care delay is associated with increased morbidity and mortality risk due to worsened health condition especially among those with chronic mental conditions.33 Thus, our findings highlight the urgent need for efforts to reduce disparities in medical care access between those with and without a history of depression, especially during the time of public health crisis. Previous literature has found that both physician-related factors (eg, closed physician’s offices and availability of appointments) and patient-related factors (eg, fear of COVID-19 exposure) for forgone medical care were important during the pandemic.34 Thus, in the time of future public health emergencies, providing concise and accurate information about the situation including infection rate and quarantine is crucial to reduce uncertainty and public concerns and to minimise unnecessary cancellations of outpatient appointments for relatively urgent issues or group psychotherapy.37 To improve access to mental healthcare, enhancing management of mental health problems in primary care settings, especially by coordinated and integrated mental healthcare, to provide timely medical care access should be considered during and beyond future public health crisis.38 In addition, for older adults with underlying chronic conditions such as depression, use of telehealth for continued care and medication management may be a safe and viable option.39 In fact, many restrictions on telehealth use in outpatient settings were temporarily lifted by the US government on 17 March 2020, which made it possible that healthcare providers could switch to remote delivery of mental healthcare relatively quickly.40 Improving access to technology-enabling factors such as cell phones or computers and providing opportunities to learn technology (eg, by information technology staffs offering technical support for those with low digital literacy) can be an effective way to encourage telehealth use among Medicare beneficiaries with depression.35 41 Moreover, education and training healthcare providers on virtual mental healthcare delivery would also be needed to ensure adequate quality of care being delivered during future public health crisis.38
Our study further showed that Medicare beneficiaries with a history of depression have experienced more social and economic disruptions than their counterpart throughout the three rounds of surveys. Although unmet basic needs and perception of financial risk were prevalent among older adults with depression even before the pandemic,42 43 the COVID-19 pandemic tended to worsen them,6 possibly due to greater challenges in daily lives and financial circumstances caused by uncertainty and lockdown during the pandemic. In addition, mental health disruptions such as social disconnectedness, stress, anxiety and loneliness were consistently more prevalent among older adults who had a history of depression throughout the study period. Our findings raise concerns for this vulnerable population as poor mental health can greatly reinforce the symptoms of depression. Higher mental health disruption among Medicare beneficiaries with depression during the pandemic may also negatively affect their mental health even beyond the pandemic. Therefore, it is imperative to better support Medicare beneficiaries with a history of depression to alleviate their financial difficulty and mental health disruptions and enhance social support for them during and beyond the pandemic. Several states made rapid policy changes in response to the COVID-19 pandemic such as providing better supports for community-based organisations, issuing waivers to relieve restrictions for getting needed medications in some limited cases, and organising basic housing and food support for the most vulnerable population with mental illness; these state efforts were very helpful and can guide future state-level and federal-level efforts in similar public health crisis.44 45
Our study demonstrated that engagement of essential preventive behaviours (washing hands properly, wearing a facemask in public, and keeping a six-foot distance) and having high perception of COVID-19 severity were not significantly different between Medicare beneficiaries with and without a history of depression. These findings suggest that depression is not a risk factor that may reduce the likelihood that Medicare beneficiaries follow public health measures to mitigate COVID-19 infection or possibly other infections in future public health emergencies.
Limitations
We note several limitations of this study. First, survey responses were self-reported which may subject to recall bias. Second, the survey included information on whether the respondent had a history of depression, but no further information including the timing or severity of the diagnosis and current status of depressive symptoms. Third, our sample included only Medicare beneficiaries who had stable source of health insurance, so it is not possible to generalise our results to other populations. Fourth, the cross-sectional data structure of the MCBS limited our ability to track individuals over time, and we could not include all the potential factors as covariates in our models. Therefore, our results should be interpreted as association, not causality. Lastly, we assessed time trends of outcomes for the period of Summer 2020 to Winter 2021 which did not include more recent period of the pandemic. Further analyses are needed to elucidate the trends in recent times. Despite these limitations, this study offers important evidence of persistent disparities in healthcare access and social and financial outcomes between Medicare beneficiaries with and without a history of depression.
Conclusions
In this repeated cross-sectional study, we found that Medicare beneficiaries with a history of depression consistently experienced more difficulties in access to medical care and meeting their basic needs caused by the pandemic, including ability to pay rent or mortgage and access to medication, doctor appointment, food and household supplies, compared with those without a history of depression throughout the pandemic (June 2020–April 2021). In addition, those with a history of depression persistently experienced more financial and mental health disruption caused by COVID-19, but we did not find evidence for a significant association of a history of depression with engagement of essential preventive behaviours or perceived severity of COVID-19. COVID-19 pandemic may exacerbate existing barriers to healthcare access and financial and social needs, possibly leading to persistent unmet needs among Medicare beneficiaries with depression beyond the pandemic. This study emphasises the greater challenges faced by Medicare beneficiaries with a history of depression during the pandemic and informs healthcare policy experts about the needs of healthcare, financial and administrative support to these populations. Efforts are needed to address the patient, physician and other factors that underlie these unmet needs for Medicare beneficiaries with a history of depression.
Data availability statement
Data are available in a public, open access repository. The Medicare Current Beneficiary Survey is maintained by the Centers for Medicare and the datasets generated and/or analysed during the current study are publicly available.
Ethics statements
Patient consent for publication
Ethics approval
This study is based on the secondary analysis of existing survey data that does not require Institutional Review Board review.
This post was originally published on https://bmjopen.bmj.com