key: cord-0691345-6zphb80a authors: Yamashita, Takashi; Punksungka, Wonmai; Van Vleet, Samuel; Helsinger, Abigail; Cummins, Phyllis title: Experiences of Older Adults During the 2020 COVID-19 Pandemic in the United States: An Initial Exploration of Nationally Representative Data at the Intersection of Gender and Race date: 2022-03-03 journal: J Appl Gerontol DOI: 10.1177/07334648211048258 sha: eca3cc2203264bdbe576f595981f00f4467093e2 doc_id: 691345 cord_uid: 6zphb80a Little is known about the overall experiences and feelings of diverse older populations during the 2020 COVID-19 pandemic. To provide the baseline information for future research and policy, this study analyzed the 2020 Health and Retirement Study COVID-19 project data (n = 1782). More than 70% of older adults reported the following activities: watching TV (98%), reading (90%), using a computer and the internet (83%), gardening (82%), walking (75%), baking and cooking (73%), and praying (73%). Volunteering and attending community groups, which are known to benefit well-being, were unpopular (less than 8%). During the pandemic, older adults were generally satisfied with their lives, but more than half of them were concerned about their own health, family’s health, and future prospects. Our study also showed the differences in the experiences and feelings by gender and race as well as the intersection of gender and race in the United States. Since early 2020, the COVID-19 pandemic (pandemic, hereafter) has impacted virtually all aspects of life among all members of society. However, older adults are known to face greater disadvantages (e.g., health, employment) resulting from events like the Great Recession and pandemics than their younger counterparts (Johnson & Butrica, 2012) . Also, given the gendered life course and diversity (e.g., race, ethnicity) among older populations, experiences during the pandemic are not homogenous (Moen, 2001) . Indeed, women and racial/ethnic minorities tended to face a greater risk of unemployment and discrimination during the pandemic (Gemelas et al., 2021) . Moreover, the theory of intersectionality suggests that these differences are exacerbated when we consider the intersection of gender and race due to unique advantages/disadvantages shaped by social and political systems (Crenshaw, 1989) . Individual characteristics (e.g., gender, race) are often examined separately in emerging COVID-19-related studies (Morrow-Howell et al., 2020) . For example, mental health by gender, race, and ethnic groups are reported separately (Bui et al., 2020) . This view is in alignment with the medical model, or disease-centered approach, which views an outcome of interest (e.g., mental health, diabetes) is the same regardless of the individual characteristics and social circumstances. While a specific area of inquiry is critical for systematic hypothesis testing and theory building, a bigger picture involving the overall experience, may be overlooked. This study takes a person-centered approach inspired by a gerontological paradigm-biopsychosocial model (Morgan, 2012) , the intersectionality framework (McCall, 2005) , and the whole-person wellness model, which depicts how multiple life domains jointly contribute to well-being (Montague & Frank, 2007) . Considering that more COVID-19 data are becoming available, timely national data exploration could provide baseline findings for future research, allow for initial/ immediate review of older adults' lived experiences, determine needs for emergency response through policy, and identify interventions to help older adults adapt to the ongoing pandemic and post-pandemic times. Taken together, the goal of this study was to provide a timely description of what diverse older populations did and felt during the pandemic and illustrate unique experiences at the intersections of gender and race. We employed the descriptive inter-categorical complexity analytic framework to document experiences at the intersections of gender and race (Bauer & Scheim, 2019; McCall, 2005) . We focused on the four cross-classified subgroups by gender (women and men) and race (White and Black). Nationally representative data were derived from the 2020 Health and Retirement Study (HRS) COVID-19 project. HRS is a biannual survey for U.S. adults aged 50 years and older. More detailed descriptions of HRS and the COVID-19 project have been published elsewhere (The Institute for Social Research, 2021). We analyzed the HRS COVID-19 project preliminary data (Early version 1.0), which provide information of older adults' experience during the pandemic. After excluding the missing values (n = 96 or about 4.9% of the total samples), our final sample consisted of 1782 older adults aged 50 years and older (see Tables 1 and 2 for more detailed breakdowns by gender and race). Considering the life expectancy at birth in the United States (about 79 years old), we considered the age of 50 as the typical transition to the second half of the adult life stage. Outcome Variables. In view of the whole-person wellness framework (Montague & Frank, 2007) , we selected the survey items which are relevant to all older adults rather than those relevant to only a sub-group of older adults, such as those with children. Specifically, we focused on the activities and subjective evaluation of life. To make a large set of measures more comparable, all variables of interests were dichotomized (see Supplementary Table S1 for the coding) and classified into one or more of seven wellness dimensions, including physical, social, emotional, vocational/financial, intellectual, spiritual, and environmental (see Supplementary Table S2 ) (Montague & Frank, 2007) . Five researchers with graduate-level training discussed and added the environmental dimension to the original six dimensions by Montague and Frank and then chose relevant wellness dimensions for each item (see Tables 1 and 3 ). Consideration of an additional environmental dimension (e.g., comfortable living space, safety) and multiple wellness dimensions was necessary as an activity may be linked to more than one wellness dimension. Grouping Variables. The respondents were cross-classified by gender (women and men) and race (White and Black). We considered four groups, including White women, White men, Black Women, and Black men, given the available data. Covariates. Per the analytic framework suggested by McCall (2005) , we adopted a descriptive intersectionality approach and included age (in years) and educational attainment (college [associate] degree or higher vs. less than college degree) as the only covariates in the analysis. Differences across age groups (e.g., 50 vs. 65 vs. 85 years old) and educational attainment, which is a widely accepted summary indicator of socioeconomic position and resource (e.g., financial, social) availability, are important even in a descriptive intersectionality study because the crude/unadjusted measures could result in misleading interpretations. While The sampling weights (CVWGTR) were applied; for all groups, the means were adjusted for age and educational attainment. more comprehensive sets of covariates should be considered in future analytic intersectionality studies (Bauer & Scheim, 2019) , age and education should still be the necessary adjustments for any baseline group comparisons. All analyses were conducted using the SAS software version 9.4 (Copyright © 2013, SAS Institute, Inc.), and the HRS COVID-19 module preliminary weights (CVWGTR) were applied. To examine the experiences across gender and race, survey-weighted age-education-adjusted proportions were computed, and multiple pairwise comparisons with Tukey-Kramer method were employed (Benjamini & Braun, 2002) . We used SAS PROC GLM with LSMEANS command (Cai, 2014) . Given the outcome measures were all dichotomous, the means are equivalent to the proportions. The least squares mean, which is estimated based on the linear combinations of all measures, can be considered the adjusted proportion in this study. Tukey-Kramer method corrects the inflated Type 1 error rates due to the multiple comparisons of unbalanced group sizes. The analytic approach is equivalent to the analysis of covariance or ANCOVA with post hoc tests for gender and race, as well as the linear model with gender, race, gender-race interaction term, age, and educational attainment as the predictors (DeMaris, 2004) . The statistical significance was determined based on whether the estimated 95% confidence intervals included 0. Tables 1-4 show the adjusted percentages of all outcome measures of interest. Regarding the activities, about 34% of older adults worked for pay during the pandemic in 2020. More than 70% of older adults reported the following activities: watching TV (98%), reading (90%), using a computer and the internet (83%), gardening (82%), walking 20 min or more (75%), baking and cooking (73%), and praying (73%). These activities are relevant to the physical, intellectual, spiritual, and vocational dimensions of wellness. The least popular activities were volunteering (7%) and attending community groups (5%). Regarding the subjective The sampling weights (CVWGTR) were applied; for all groups, the means were adjusted for age and educational attainment. evaluation of life, more than half of the older adults were concerned about their family's health (69%), own health (56%), and future prospect (53%) during the pandemic. Also, 59% felt stressed. At the same time, most older adults reported positive evaluations of life in general. For example, over 80% of respondents were satisfied with life and all life domains (e.g., place to live, leisure, financial situation). When compared across the intersections of gender and race, several statistically significant differences were observed. Regarding the activities, for example, Black women were less likely to work for pay than White men and Black men during the pandemic. Also, Whites were more likely to volunteer than Blacks, but White women were more likely to volunteer than White men and Black women. Moreover, White men were more likely to go to sport or social clubs than Black women. Regarding the subjective evaluation of life, Black men were more concerned about getting help if needed and their future prospects than White women. Finally, White women were more satisfied with their financial situation than Black men. The initial explorations of nationally representative data showed that during the pandemic, most older adults engaged in watching TV, reading, using a computer and the internet, gardening, walking, baking and cooking, and praying. Yet, despite the health and well-being benefits, volunteering and attending community groups were unpopular (Russell et al., 2019) . Also, whereas most older adults were satisfied with their lives, over half of them were also concerned about their own health, family's health, and future prospects. As shown in Tables 5 and 6, some of the differences at the intersections of gender and race would have been missed if gender and race were analyzed separately. A few limitations should be noted. Only communitydwelling older adults were included in the HRS data. Therefore, findings may be somewhat biased toward healthier, older populations. Also, we cannot address possible systematic activity patterns as well as underlying explanations of observed differences at the intersections of gender and race, for example, due to functional limitations/disability, caregiving responsibilities, access to certain activity spaces The sampling weights (CVWGTR) were applied; for all groups, the means were adjusted for age and educational attainment. All least squares means/percentages and multiple comparisons were adjusted for age and educational attainment (college or higher vs. less than college) and by the survey weights. All least squares means/percentages and multiple comparisons were adjusted for age and educational attainment (college or higher vs. less than college) and by the survey weights. (e.g., walkable parks and neighborhoods) and community/ societal-level policy (e.g., social distancing) in place in our exploratory analysis. This study made two contributions. First, the explorations of the nationally-representative data, which reflect older adults' activities and subjective assessment of life during the pandemic, are useful to identify important areas of future research and policy discussions for well-being in later life. Second, our initial findings of differential experiences at the intersections of gender and race in later life could inform future policies targeting specific sub-groups (McCall, 2005) and benefit the development of an analytical intersectionality study with specific hypotheses. In view of whole-person wellness (Montague & Frank, 2007) , four preliminary implications are worth noting. First, most older adults engaged in physical (e.g., walking, gardening), intellectual (e.g., reading), and spiritual (e.g., praying) activities, which benefit physical and mental health (Zimmer et al., 2016) . Watching TV, which is one of the relatively passive leisure activities, from a wellness standpoint, was the most popular activity. Compared to the pre-pandemic period (89%; see Krantz-Kent, 2018), our findings (98%) showed a potential increase in TV watching, although the purpose (e.g., leisure vs. public health information seeking), changes in the amount of screen time, and contexts (e.g., lockdown policy in place), need to be further studied. Second, future research needs to identify changes more clearly in activity participation and the reasons for reduced and missed wellness opportunities (e.g., volunteering). For instance, while our findings (7%) indicated that the volunteer participation rate might have decreased, compared to the prepandemic period statistic (25%; see Grimm, 2018) , qualitative changes in volunteering (e.g., formal, informal, virtual, COVID-19-related) are yet to be investigated (Sun et al., 2021) . Third, using only one life satisfaction assessment item may fail to capture overall subjective well-being as well as concerns about specific life domains. One could be simultaneously satisfied with life and concerned about a specific domain of life (Diener et al., 2013) . Finally, the preliminary findings on the intersection of gender and race should be verified with more comprehensive sets of covariates. In conclusion, while most older adults engaged in physically, intellectually, spiritually, and vocationally beneficial activities during the pandemic, only a few participated in volunteering and community groups despite known wellbeing benefits. During the pandemic, older adults were generally satisfied with their lives, but more than half of them were concerned about their own health, family's health, and future prospect. Our study also showed the differences in the activities and subjective evaluation of their lives by gender and race as well as the intersection of gender and race in the United States. Methods for analytic intercategorical intersectionality in quantitative research: Discrimination as a mediator of health inequalities John W. Tukey's contributions to multiple comparisons. The Annals of Statistics Race and ethnic group disparities in emotional distress among older adults during the COVID-19 pandemic Making comparisons fair: How LS-means unify the analysis of linear models Demarginalizing the intersection of race and sex: A black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics Regression with social data: Modeling continuous and limited response variables Theory and validity of life satisfaction scales Inequities in employment by race, ethnicity, and sector during COVID-19 Where are America's volunteers? A look at America's widespread decline in volunteering in cities and states Age disparities in unemployment and reemployment during the great recession and recovery. Unemployment and Recovery Project Brief Television, capturing America's attention at prime time and beyond The complexity of intersectionality The gendered life course Creating whole-person wellness Paradigms in the gerontology classroom: Connections and challenges to learning Recovering from the COVID-19 pandemic: A focus on older adults Volunteering and wellbeing among ageing adults: A longitudinal analysis. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations Older adults' attitudes toward virtual volunteering during the COVID-19 pandemic Spirituality, religiosity, aging and health in global perspective: A review. SSM -Population Health The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was in part supported by the Sloan Research Network on Aging and Work at Boston College, small grant program for "unequal prospects for continued work in later life amid a global pandemic: possible solutions." The opinions expressed are those of the authors and do not represent views of the Sloan Research Network on Aging and Work at Boston College. Takashi Yamashita  https://orcid.org/0000-0003-2325-126X Supplemental material for this article is available online.