Coping with being cooped up: Social distancing during COVID-19 among 60+ in the United States.

OBJECTIVES: This study examined the impact of sheltering in place and social distancing among adults aged 60 and older during the 2020 outbreak of COVID-19 in the United States.
METHODS: Using convenience sampling respondents were asked to complete a web-administered survey to explore impact of social distancing on loneliness, stress, and behavioral changes. The analytic sample consisted of 833 responses of persons aged 60 and older.
RESULTS: A large portion reported being stressed (36%), and/or being lonely (42.5%). Nearly 1/3 stated that their sense of loneliness increased during the time of social distancing. Respondents reported engaging in more solitary activity (and fewer in-person activities), using email and text messages more than usual, and spending more time on computers/tablet than usual. Approximately 2/3 reported using more social media than usual. These differed significantly by younger (age 60-70) and older (71+) respondents. Additionally, changes in physical activity, drinking, recreational drug use and sleeping pattern changes differed by age.
CONCLUSIONS: Social distancing has significant consequences on loneliness and health behaviors among adults in the United States, many of which differ by age group. Results have implications for continued shelter in place practices, but also for any older adult that may be homebound for other reasons.

The worldwide spread of COVID-19 has created a pandemic unlike any seen in a century, and it is disproportionately affecting older adults (1). Global data, including those from the United States, consistently show higher death rates among older adults and those with underlying conditions (2). To curb disease transmission and risk of infection, the US began recommending restrictions of social interactions for these most vulnerable populations. Specifically, on March 5, 2020, the Centers for Disease Control (CDC) warned that older adults were particularly vulnerable to the COVID-19 and encouraged older adults to take extra precautions (2). Recommended precautions included social distancing and sheltering in place. Social distancing is a public health practice to reduce disease transmission by preventing people with an illness from coming into close contact with non-infected people (3). Similarly, sheltering in place is aimed at reducing transmission rates by remaining at home except for essential travel. Adhering to these recommendations means many older adults in the US have been at home with minimal outside contact for an extended period of time.

Little is known about how this extended period of sheltering in place has impacted older adults. However, researchers are concerned that current public health safety measures pose a risk of loneliness, even among those that are living with other family members (1). Social interaction is integral to human well-being, and the impact of not having ones social needs met (aka loneliness) can be wide ranging. Loneliness has been connected to both morbidity and mortality (4, 5). Loneliness can be a particularly important concern for older adults, who may have additional risk factors, such as physical and cognitive impairment, death of loved ones, and decreases in economic resources (6, 7). Sheltering in place and social distancing may be adding additional risk factors for loneliness as older adults lose many potential outlets for social interaction, such as going to religious services, seeing friends and family, shopping, and group exercise classes.

While data on the impact of social distancing for COVID-19 is not yet available, research on prior quarantining events suggests that one unintended consequence of such measures is psychological distress. Most recently, research on the impact of mandated quarantine during the severe acute respiratory syndrome (SARS) outbreak in Canada showed a high level of psychological distress among respondents (8).

The purpose of this study was to explore how adults aged 60 and older in the US are coping with social distancing and sheltering in place during the COVID-19 pandemic.

Specifically, the survey hoped to answer two main questions: 1) What have been the effects of social distancing on adults aged 60 and older, and 2) What have been behavioral and social communication changes since social distancing started? The answers can shed light on challenges experienced by those social distancing, and can guide potential public health interventions strategies.

MATERIALS AND METHODS

Survey Instrument. We conducted a web-based survey in Qualtrics, which consisted of 28 questions. A web-based format was chosen due to budget and time constraints; it was important to reach those practicing social distancing during the event to avoid recall bias (8). The survey was open for two weeks, beginning on March 30th and concluding on April 12th, 2020. The sampling of the survey was a convenience sample, with links sent out on social media and gerontology listservs. Only persons aged 60 and older were asked to respond. Each respondent was asked to forward the survey on to other persons aged 60 and older. The study was approved by the University of Georgia Institutional Review Board (IRB# 00002133).

Analytic sample. 922 persons initiated the survey, but only 843 (91.4%) completed the survey. An additional 10 responses were not included in the analytic sample because they did not provide their age or because they were not in the United States. The final sample size was 833 respondents living in the US, aged 60 and older.

Variables. The survey collected socio-demographic data, including age (top-coded at 85), sex (male, female, other), race (White, Black/African American, American Indian/Alaska Native, Asian, Native Hawaiian/Pacific Islander, Other), ethnicity (Hispanic or Latino as defined by the US Census (14)), employment status (part-time, full-time, not employed), and household size (continuous). Additionally, respondents were asked to estimate how many people they continue to be in contact with in person, but outside of the home (no one else, 1, 2-3, 4 or more). Respondents were also asked about the type of home they live in (apartment, house, nursing home, assisted living, other).

Loneliness was determined by asking respondents: “Since you started socially distancing, have you felt lonely?” (yes, most of the day/yes, some of the day/no, never). For analyses all respondents who answered most or some were coded as lonely. Respondents were then asked a follow up loneliness question: “Do you feel more, less, or the same level of loneliness compared to the time before you were socially distancing?”.

Stress was measured by self-report response to the prompt: “Stress means a situation in which a person feels tense, restless, nervous or anxious or is unable to sleep at night because their mind is troubled all the time. Do you feel this kind of stress these days?” (answer choices were a 5-point Likert scale). For analyses the five answer choices were collapsed into three groups (1=a great deal/a lot, 2=moderate amount, 3- A little/not at all). Relationships was determined by the question: “Do you feel like you have close relationships that bring you emotional security and well-being?” (answer choices were yes, definitely/yes, somewhat/no, not at all). Self-rated health was measured by a 5-point Likert scale ranging from poor to excellent.

Finally, respondents were asked to choose all that apply from a range of behavioral and communication changes since beginning social distancing. These included doing more or less: sleeping, engaging in physical activity, smoking, drinking alcohol, using recreational drugs, eating, solitary activities, engaging in in-person activities, making/receiving phone calls, using social media, emailing, ending/receiving text messages, and spending time on the computer/tablet/phone.

Analyses. Data are presented in descriptive format. Means were calculated to summarize continuous variables. For categorical variables, group proportions were calculated. To explore differences by age, the results were divided into younger (ages 60-70) and older (ages 71+). Because sample cell sizes were small, all statistical comparisons by age are conducted using Fisher’s Exact test. Stress, loneliness, and relationship responses were examined by whether respondents lived alone, using Pearson chi-square tests. A p-value of < 0.05 was considered to be significant for all analyses. All analyses were conducted using SPSS v21.

RESULTS

Description of the sample. The sample age range was 60 to 85+(top-coded), with 62.8% in the younger (aged 60-70) category (Table 1). A majority (80.5%) of the sample was female, White (96%), and not Hispanic (97.9%). Race/ethnicity did not differ by age of respondents, though the younger age group was significantly more likely to be female (84.3%) compared to the older age group (74.1%).

Table 1.

Description of the sample, by age category (n = 833)

	Total Sample	Aged 60-70 yes (n = 523)	Aged 71+ yes (n = 310)	p-value
Demographics
Race
White	96.0	94.8	98.0	0.082
Black or African American	1.9	2.9	0.3
American Indian or Alaska Native	0.2	0.2	0.3
Asian	0.2	0.4	0.0
Other	1.6	1.7	1.3
Hispanic (yes)	2.1	2.2	2.1	0.563
Female	80.5	84.3	74.1	0.000
Days since respondent began social distancing (mean, social distancing)	16.95 (5.8)	16.69 (5.4)	17.38 (6.2)	0.101
Employment Status
Not employed	67.7	56.5	86.7	0.000
Usually work from home, continue to work from home	30.3	30.5	28.6
Usually work outside the home, now working at home	59.7	61.6	48.6
Currently work outside the home and continue to work outside the home	4.6	3.9	8.6
Household
Household size				0.026
Living alone	21.3	19.0	25.3	0.083
Living with one other person	64.3	65.4	62.5
Living with two or more people	14.4	15.6	12.3
Type of home				0.132
House	85.6	87.5	82.6
Apartment	8.6	7.9	9.7
Other (e.g. Townhome, condo)	5.7	4.4	7.7
Social contact outside the home
How many people outside of household do you see regularly in person?				0.334
No one else	51.2	48.7	5.3
1 person	15.3	16.0	14.2
2-3 people	21.9	23.1	19.7
4+ people	11.6	12.1	10.7
Phone contact				0.246
Never	5.5	5.7	5.2
Once a day	37.5	36.4	39.5
2-3 times a day	38.6	37.5	40.5
4+ day	18.3	20.3	14.9
Video Call contact				0.000
Never	43.6	37.6	53.8
Once a day	43.9	46.7	39.1
2-3 times a day	10.4	12.8	6.4
4+ day	2.1	2.9	0.7
Do you feel like you have close relationships?				0.156
Yes, definitely	84.6	82.8	87.7
Yes, somewhat	14.2	15.9	11.3
Mental Health
Self-rated health				0.449
Excellent	24.3	22.2	27.1
Very Good	44.4	45.0	43.2
Good	25.7	27.0	23.5
Fair	4.7	4.8	4.5
Poor	1.2	1.0	1.6
Since social distancing do you feel lonely (yes, all or yes, some)	42.9	42.5	43.5	0.774
Compare to loneliness before social distancing				0.184
More lonely	30.9	33.1	27.0
Less lonely	3.5	3.5	3.6
About the same	65.6	63.4	69.4
Feel stressed				0.000
A little/not at all	63.1	56.5	74.2
A moderate amount	24.5	27.6	19.4
A lot/great deal	12.4	15.9	6.5

Note: totals may not sum to 100% due to rounding

Who is social distancing and what does their household look like? The average number of days spent practicing social distancing for this sample was 16.95 (range 2-65 days, SD 5.8). This did not differ significantly by age. For context, the survey opened 25 days after the official CDC recommendations were published for older adults and ended 39 days after the recommendations.

About 85% of survey respondents lived in a house, 8.6% lived in an apartment, and 5.7% lived in other locations, such as condos, or townhomes. A majority of respondents were living either with one other person (64.3%) or living alone (21.3%).

Approximately two-thirds (67.7%) of respondents were not employed, although this differed significantly by age. The older age group had higher rates (86.7%) or not working, compared to 56.5% of younger age group. Of those that were employed, 28-30% usually work from home and continue to work from home (across both age groups). Among the older respondents 48.6% usually worked outside the home, but were now working from home because of COVID-19, compared to 61.6% of the younger group.

What social contacts do people have? More than half (51.2%) of respondents saw no one else outside their household, 15.3% saw only one person, 21.9% saw 2-3 people, and 11.6% saw four or more people. Most (94.5%) respondents spoke with someone on the phone at least once a day, and 56.4% spoke to someone using video calling (e.g. Facetime/skype) at least once a day. Video calls were significantly more common among the younger respondents, with nearly 53.8% of older respondents saying the never used video calls, compared to 37.6% of younger respondents. Nearly all respondents (98.8%) reported having at least some close relationships that bring them emotional security and well-being.

How are people feeling? Respondents rated their health fairly highly, with 94.4% rating their health as good, very good, or excellent. Approximately 43% of respondents reported feeling lonely some or most days. For about two-thirds (65.6%) of the respondents their level of loneliness had not changed, though 30.9% stated that they are more lonely now compared to before they were socially distancing. About 3.5% felt less lonely.

Over one-third (36.9%) of the sample reported being moderately to a great deal stressed. This differed significantly by age. Older respondents reported higher rates of being only a little or not at all stressed out (74.2%) compared to younger respondents (56.5%).

Analyses stratified by living alone. Some health measured differed significantly by whether or not respondents lived alone (Table 2). Those living alone had much higher rates of current loneliness (59.3%) compared to those that lived with others (38.4%). Those living alone were also significantly more likely to report that their loneliness had increased since social distancing (42.4%) compared those living with others (27.9%). Those living alone were less likely to report having a close relationship that brings them emotional security and well-being, where 3.6% said they didn’t have this at all, and 20.4% said they had it somewhat. This was compared to 0.0% for not at all and 12.7% for somewhat among those living with others. Stress, having close relationships, and physical health did not differ significantly by living alone.

Table 2.

Physical and mental health by living alone status

	Live alone (n = 167)	Live with others (n = 623)	p-value
Since social distancing do you feel lonely (yes, all or yes, some)	59.3	38.4	0.000
Since social distancing how does loneliness now compare to before?
More lonely	42.4	27.9	0.000
Less lonely	0.0	4.1
About the same	57.6	68.1
Do you feel like you have close relationships?
Yes, definitely	76.0	86.7	0.000
Yes, somewhat	20.4	12.7
No Not at all	3.6	0.6

Changes in behaviors are presented below and in Table 3.

Table 3.

Changes in behavior since COVID-19

	Total sample yes	Aged 60-70 yes (n = 523)	Aged 71+ yes (n = 310)	p-value
Sleep more than usual	27.1	26.0	29.0	0.192
Sleep less than usual	15.8	18.2	11.9	0.010
Engage in physical activity more than usual	24.6	29.3	16.8	0.000
Engage in physical activity less than usual	37.3	33.8	43.2	0.004
Smoke more than usual	1.3	1.7	0.6	0.159
Smoke less than usual	1.1	1.1	1.0	0.553
Drink more alcohol than usual	12.2	14.9	7.7	0.001
Drink less alcohol than usual	4.4	3.6	5.8	0.098
Use recreational drugs more than usual	1.4	1.3	1.6	0.482
Use recreational drugs less than usual	1.0	0.4	1.9	0.034
Eat more than usual	35.4	40.9	26.1	0.000
Eat less than usual	11.3	9.8	13.9	0.045
Engage in more solitary activities than usual (e.g. puzzles, reading)	56.4	53.3	61.6	0.012
Engage in fewer solitary activities than usual (e.g. puzzles, reading	1.9	1.3	2.9	0.094
Engage in more in person activities than usual (e.g. going to group exercise classes)	1.2	1.3	1.0	0.454
Engage in fewer in person activities than usual (e.g. going to group exercise classes)	48.4	46.7	51.3	0.111
Make/Receive more phone calls than usual	45.3	46.7	42.9	0.164
Make/Receive fewer phone calls than usual	4.6	4.0	5.5	0.208
More use of social media (e.g. Facebook) than usual	66.3	73.2	54.5	0.000
Fewer use of social media (e.g. Facebook) than usual	1.3	1.0	1.9	0.187
Email more than usual	40.2	37.9	44.2	0.042
Email less than usual	3.1	3.1	3.2	0.522
Send/Receive more text messages than usual	60.6	64.2	54.5	0.003
Send/Receive fewer text messages than usual	1.9	1.9	1.9	0.586
Spend more time on the computer/tablet/phone	67.2	67.5	66.8	0.444
Spend less time on the computer/tablet/phone	0.5	0.8	0.0	0.155

Note: Respondents were asked to choose all that apply

Health Behaviors. Sleep patterns were impacted for a little over 1/3 of the sample, with 27.1% reporting more sleep than usual, and 15.8% reporting less sleep. Younger respondents were significantly more likely to report sleeping less. Physical activity levels changed as well, in both directions. About one-quarter (24.6%) of respondents engaged in more physical activity, while 37.3% engaged in less physical activity than usual. Younger respondents were significantly more likely to increase their physical activity, whereas older respondents were significantly more likely to decrease their physical activity.

Alcohol intake increased for 12.2% of the sample, but decreased for 4.4% of the sample. Increase in alcohol consumption was significantly higher for the younger age group (14.9%) compared to younger respondents (7.7%). Smoking and recreational drug use both had smaller changes, with 1.3% smoking more than usual and 1.1% smoking less than usual, and 1.4% taking more recreational drugs and 1.0% taking fewer drugs. Less use of recreational drugs than usual was significantly more common among older adults, though sample sizes are small.

Since social distancing, over one-third (35.4%) of respondents report eating more than usual, and 11.1% ate less than usual. This differed significantly by age, where younger people were more likely to report eating more, and older respondents were more likely to report eating less.

A majority of respondents (56.4%) engaged in more solitary activities, and 1.2% engaged in fewer of those activities. A larger proportion of older respondents reported engaging in solitary activities (61.6%) compared to younger respondents (53.3%). While some (1.1%) have engaged in more in-person activities than usual, more (48.4%) report less in person activities.

Communication. Among respondents, 4.6% make/receive fewer phone call, 1.3% use social media less, 3.1% email less, 1.9% send/receive fewer text messages, and 0.5% spend less time on their computer/tablets/or phones. However, for a much larger group of respondents, each of these styles of communication increased. Among respondents, 45.3% make/receive more phone calls, 66.3% use social media more, 40.2% email more than usual, 60.6% send/receive more text messages, and 67.2% spend more time on their computer/tablet/or phones than usual.

Several of these communication styles different significantly by age. The younger respondents were significantly more likely to increase their social media use compared to older respondents (73.2 vs. 54.5%, respectively). Younger respondents also indicated an increased use of text messages (64.2%) compared to older respondents (54.5%) Alternatively, the older respondent group reported increased use of email (44.2% vs. 37.9%).

DISCUSSION

While social distancing is not a new phenomenon (13), it was never as sustained in the US or as prevalent as it has been since the COVID-19 outbreak. While the CDC provided guidance for older adults and adults with high-risk conditions, for most people the process of sheltering in place and social distancing is asking them to create a new normal with little to no precedent to lean on. One potential concern is the higher risk of loneliness as we decrease social contacts. This may be particularly concerning for older adults, who have more risk factors for loneliness. The goal of this survey was to see how people aged 60 and older are doing given the recommendations to practice social distancing.

The survey results showed that some respondents are adjusting well and not feeling lonely or stressed. However, a significant portion of adults reported experiencing at least some distress. Over a third (36%) of respondents reported feeling moderate to a great deal of stress during this time. Those over age 70 were less likely to report stress, however. This stress and age pattern follows previous data showing decreases in stress by age (9). It is not clear from this cross-sectional data if this age pattern is a cohort effect (reflecting history and life experiences born around the same time) or a difference in individuals’ circumstances.

Additionally, a large portion of respondents (43%) noted being lonely. Loneliness rates did not differ significantly by age. A recent review suggests that 25-29% of Americans aged 70 years and older reporting being lonely (10), suggesting that this sample had relatively high rates. Rates of loneliness were significantly higher among those living alone (59.4%). However, even those living with others had relatively high rates of loneliness (41.7%). This supports previous research that simply living with others does not erase the risk of loneliness (11).

Perhaps most notable is that many respondents (31%) stated that their sense of loneliness increased during the time of social distancing. This suggests that social distancing practices are having an impact on a portion of those aged 60 and older. Respondents who lived alone were more likely to report an increase in loneliness. This indicates that those living alone are a particularly vulnerable group to increased loneliness during periods of quarantine.

Not surprisingly, a large portion of respondents reported engaging in more solitary activity (and fewer in-person activities). Respondents also reported using email and text messages more than usual and spending more time on computers/tablet than usual. Two-thirds of respondents reported using more social media than usual.

Some of these behaviors differed significantly by age. For the older age group, adults they were more likely to engage in less physical activity, and to eat less. While participating in less physical activity is probably not a healthy behavior, it is unclear whether eating less is healthy or not. This likely depends on the individual’s weight and health, as well as in the nutritional changes made with eating less. Unlike their younger counterparts (aged 60-70), older respondents were not drinking more alcohol (in fact, they were more likely to report lower alcohol intake, though this difference was only marginally significant). More older adults reported less recreational drug use (though this was a very small response rate). Younger respondents were more likely to report sleeping less.

While there is no one clear overall pattern by age, the data suggest that those in the younger age category may be engaging in less healthy behaviors during COVID-19. In this sample, younger people are more stressed, drinking more alcohol, sleeping less, and eating more. They are, however, engaging in more physical activity, which for many is assumed to be a healthy behavior.

Communication also differed by age. While both age groups reported using social media more than usual, those in the younger age group reported this much more often than those aged 70 and older (73% vs. 55%, respectively). The younger group also used texts more often. Those in the older age category reported more email use instead. This suggest that intervention and public health communication strategies may need to differ by age category.

These results have implications for continued shelter in place practices, but also for how to intervene with any population of older adults that may be homebound for any number of reasons (e.g. health issues, transportation). The high level of engagement with social media suggests that for all groups social media can play an important role in interventions. This is especially the case for those aged 60-70. Doing media-based interventions through platforms such as Facebook live may be an effective way to reach this population. Interventions could include social activities, live exercise activities, or social media public health messaging campaigns. On the other hand, those in the older age category were more likely to use the phone and email. It may be better to target interventions through phone or email for this population. Examples include connecting people to hotline/warmline services such as the Institute on Aging’s Friendship line, which provides free telephone support for older adults that are lonely (12).

While interventions should certainly be targeted to those living alone, it is important not to ignore those that are living with others. These data show that those living with others are still vulnerable to loneliness.

This study had some limitations. The survey was a convenience sample of a relatively small number of adults compared to the total number of adults that are actively practicing social distancing in the US. In fact, since the survey was online, a specific subset of adults were likely not reached with this survey. This includes, but is not limited to, rural elders with limited internet access. It is also possible that a self-selection effect may have occurred (8), where those who either felt the greatest or the least effects of social distancing chose to complete the survey. Therefore, the results are not generalizable and should be interpreted with caution. Additionally, the survey was fielded in the first 2-3 weeks of the social distancing, which likely only captures the immediate impact of social distancing. Future research should explore the impact of a longer duration of social distancing.

In conclusion, this survey provides a first glimpse into the lives of adults aged 60 and older living through the unprecedented public health situation caused by the COVID-19 pandemic. The results can increase awareness of potential stress, loneliness and health behavior changes for public health officials, as well as point towards the communication methods that may be most effective for interventions.

Disclaimer.

Authors hold sole responsibility for the views expressed in the manuscript, which may not necessarily reflect the opinion or policy of the RPSP/PAJPH and/or PAHO.