Stressors associated with older adults’ depressive symptom during the pandemic: Does a caregiving role make a difference?

ABSTRACT

The present study examined COVID-19-related stressors associated with depressive symptom in the older population and whether a caregiving role made a difference in the results. Data from the COVID-19 Supplement to the National Health and Aging Trends Study (NHATS) were used to compare two groups of older adults: 248 older caregivers and 1,437 older adults without a caregiving role. A multiple linear regression analysis showed that a lack of time for self-care and healthcare delay were significant stressors in both groups. Missing social events and staying home were associated with depressive symptom only in the non-caregiver group.

KEYWORDS:

• older caregiver
• informal caregiving
• family caregiving
• stress
• psychological well-being
• COVID-19

The outbreak of COVID-19 and the ongoing pandemic have changed people’s lives in social, economic, and health contexts. . The most notable and radical changes people experienced were lockdown, quarantine, and social distancing. Due to the highly contagious nature of coronavirus, physical distancing had to be implemented as a public health measure to stop further community transmission (Mohapatra et al., 2020). Regardless of the health conditions, people are exposed to COVID-19-related stressors, which could lead to psychological problems such as depression and anxiety (Kujawa et al., 2020; Montano & Acebes, 2020). According to the literature, prolonged social isolation and changing personal care routines accounted for a lot of the stressors (Park et al., 2020; Van Jaarsveld, 2020).

Older adults as a vulnerable population during the pandemic

Although most people are exposed to the stressors during the pandemic, inequalities among individuals are often highlighted in times of crises. It was social distancing that played a crucial role in exacerbating difficulties for vulnerable populations in the case of COVID-19 (Kantamneni, 2020). Older adults are arguably one of the most marginalized populations because of COVID-19, based on the fact that those aged 65 or older constituted 79.7% of the total number of COVID-related death as of June 2021 in the United States (Centers for Disease Control and Prevention [CDC], 2021).

Not only are older adults more susceptible to the severity and fatality of coronavirus, but they can also be more vulnerable to psychosocial problems that result from other COVID-19-related stressors apart from the threats of the virus itself. A mental health survey in China showed that older adults aged 60 or above were experiencing higher levels of depression, anxiety, grief, and insomnia than other age groups during the pandemic (Qiu et al., 2020). In addition, when change in the levels of loneliness and social isolation were measured across groups of young, middle, and older adults from pre-COVID-19 to during COVID-19, older adults showed the largest increase in the two variables even though youngadults had shown the highest scores in pre-COVID-19 loneliness and social isolation (Teater, Chonody, & Hannan, 2021).

There also exist some studies that indicate no associations of age with COVID-19-related stress (Tull et al., 2020), and that report lower levels of stress and better psychosocial functioning in older adults compared to their younger counterparts (Losada-Baltar et al., 2020); however, there is a wide-spread consensus that social distancing during COVID-19 can be particularly challenging for older adults because of their lower ability and proficiency in employing digital tools (Gibson, Bardach, & Pope, 2020; Minahan, Falzarano, Yazdani, & Siedlecki, 2021; Seifert, Cotten, & Xie, 2021; Van Jaarsveld, 2020; Xie et al., 2020). Technology serves as an important gateway in promoting connectivity and social inclusion especially during pandemic as face-to-face interactions become limited. Since older adults are more likely to be inept at technology than the younger people, they are secluded from the advantages of virtual communication and online support which can work as effective coping mechanisms in the era of social distancing (Gabbiadini et al., 2020; Garfin, 2020).

Under the Double Burden Are Older Adults with a Caregiving Role

Older caregivers in the present study refer to the older adults who have a caregiving role but are not a professional caregiver. Their psychological well-being raises even more concerns since the amount of informal caregiving has grown because of a reduction in formal caregiving services amid the pandemic (Rodrigues, Simmons, Schmidt, & Steiber, 2021). Formal caregivers’ fear of contagion, as well as the need to protect patients, are the main causes of the reduction (Rodrigues et al., 2021). The rapid spread of COVID-19 also made early discharge of less critical inpatients inevitable due to the shortage of hospital facilities (Mirzaei, Raesi, Saghari, & Raei, 2020). With the need for informal caregiving to supplement public or paid services, family members experienced increased caregiving responsibilities (Stokes & Patterson, 2020).

In response, researchers have discussed the effects of the pandemic on informal caregivers’ well-being. It was reported in Italy that social distancing was associated with an increase in caregivers’ anxiety (Rainero et al., 2020), and that being an informal caregiver during the pandemic was related to poorer psychological outcomes in Austria (Rodrigues et al., 2021). A study conducted in the United States reported that stress and pain among informal caregivers after the pandemic were statistically higher than before the pandemic (Sheth, Lorig, Stewart, Parodi, & Ritter, 2020). Although the pandemic has increased the acknowledgment of psychological strain among informal caregivers globally, older caregivers’ experiences were explored in very few studies as most studies focused on the well-being of informal caregivers who were not an older adult (Spatuzzi et al., 2021). In fact, a significant number of older adults are informal caregivers themselves (Wolff & Kasper, 2006), and some of them newly began to take caregiving responsibilities after the outbreak of COVID-19 (Freedman & Hu, 2020). It can be predicted that they are under the double burden which stems from their clinical vulnerability to coronavirus and thier role as a caregiver. That is, older caregivers are at risks of facing multiple stressors if they are under stress resulting from caregiving responsibilities and find the pandemic situations stressful as well. When individuals with a chronic stressor encounter a discrete new stressor, these two sources of stress produce synergetic effect on their psychological well-being since the newer stressor might change or magnify the chronic one (Pearlin, Menaghan, Lieberman, & Mullan, 1981).

Based on the current literature, it could be predicted that older adults were an at-risk-population amid the pandemic and that older adults with a caregiving role were possibly at higher risks due to the double burden. However, there was little evidence to date about COVID-19-related stressors significantly affecting psychological well-being among older adults including those with caregiving responsibilities. . With the hope to fill this knowledge gap, the authors examined stressors associated with depressive symptom, one of the manifestations of psychological distress (Masseé, 2000), among older adults with a caregiving role and those without the role. They believed that major stressors were the first things to identify in order to find resources which could be utilized to mitigate older adults’ stress during the pandemic. They also examined whether the caregiving responsibilities would make a difference in the stressors and depressive symptom. In this way, they aimed to seewhether older caregivers were in need of special attention or support as opposed to older adults without a caregiving role. As such, the aim of the present study was to answer two research questions in the following.

1. What are the stressors associated with older adults’ depressive symptom during the pandemic?

2. Do older adults experience significantly greater stress and depressive symptom when they have a caregiving role?

Out of six potential stressors put to test, four of them were related to social distancing: the number of missed events, time spent staying home, a lack of social contact, and a lack of social activities. The other two stressors were healthcare delay and a lack of time for self-care which were known to be common COVID-19-related stressors in previous studies (Czeisler et al., 2020; Stalnaker-Shofner, Lounsbury, Collagan, Keck, & Roberts, 2021). Depressive symptom during the pandemic was assessed to examine older adults’ psychological well-being. 2000 The authors expected to answer these questions so as to inform social work practitioners and policy makers in the field of gerontology as well as mental health. The answers will allow them to provide effective interventions for older adults’ psychological well-being during this long-running pandemic.

Methods

Data and sample

Information has been collected on a nationally representative sample of Medicare beneficiaries aged 65 or older in the United States since 2011 by the National Health and Aging Trends Study (NHATS) to guide efforts to maximize health and independent functioning in the older population (Freedman & Kasper, 2019). The present study used data from the COVID-19 Supplement to the NHATS which had collected survey responses from existing NHATS respondents via mail and had been released in 2020. A sample of 1,685 were analyzed for this study whose caregiving status information had been provided. Listwise deletion was applied in handling missing data for the following reasons. First, less than 5% of missing data were found on some of the independent variables, which did not significantly affect the effect size. Second, listwise deletion is known to be a more robust method than maximum likelihood or multiple imputation when the missingness of independent variables is not at random in a regression analysis (Allison, 2002). Applying listwise deletion was not expected to yield biased estimates in this study since cases with missing data on independent variables did not show significantly different values on the dependent variable from those without any missing data. The sample was divided into two groups of older adults aged 65 or above. One was the caregiver group consisting of 248 older adults with a caregiving role, and the other was the non-caregiver group consisting of 1,437 older adults without a caregiving role. The two groups were compared with respect to their stressors and depressive symptom during the pandemic. Sample characteristics of each group are presented in Table 1.

Table 1. Sample characteristics

Variable	Caregivers (n = 248)		Non-Caregivers (n = 1,437)		X^2	p-value
	%	n	%	n
Age					25.821***	.000
65– 69	2.4	6	1.5	21
70– 74	32.7	81	25.3	363
75– 79	35.5	88	28.7	413
80– 84	15.7	39	23.3	335
85– 89	12.1	30	13.7	197
90 +	1.6	4	7.5	108
Gender					3.923*	.048
Male	40.3	100	47.1	677
Female	59.7	148	52.9	760
Race/Ethnicity					10.406*	.034
White	75.4	187	83.2	1,196
Black	16.5	41	11.1	159
Hispanic or Latino	3.2	8	3.1	44
Asian or Pacific Islanders or Hawaiian Natives	2.8	7	1.5	21
Other race	2.0	5	1.2	17

Differences between the caregiver and non-caregiver group were identified by Chi-squared test

***p < .001. **p < .01. *p < .05.

Measures

The dependent variable of this study was the participants’ depressive symptom. It was measured by one survey item asking how sad or depressed they felt during the pandemic. Responses to the item were assessed with a Likert scale (1 = Not at all; 2 = Some days; 3 = More than half of the days for some of the time; 4 = Nearly every day, during the day and at night), with higher scores indicating severer depressive symptom.

The remainders were six independent variables and consisted of potential stressors caused by the pandemic: the number of missed events, time spent staying home, a lack of social contact, a lack of social activities, healthcare delay, and a lack of time for self-care. The number of missed events was assessed with a composite of eight items asking whether the participants had missed any social events during the pandemic – either because they chose not to attend or because the events were canceled (1 = Yes and  0 = No). Events presented in the eight items included birthday parties, wedding, religious celebrations, vacation, and visits to hospitalized family. Higher scores indicate that the participants had more missed events.

Time spent staying home, a lack of social contact, and a lack of social activities were measured by the composite of several items. First, time spent staying home was assessed with three items asking the participants how often they have left home to their yard or patio just to be outside their home, to their immediate neighborhood, and to farther than their immediate neighborhood in a typical week. A five-point Likert rating was provided as the response option (1 = At least daily to 5 = Have not left home). Higher scores indicate longer time spent staying home , and lower scores indicate the opposite.

Similarly, a composite of four items was used to measure a lack of social contact. Each item was related to a means of social contact and was assessed by a five-point Likert rating. These items asked theparticipants to answer how often they have been in contact with family or friends in a typical week through virtual or in-person visits (1 = At least daily to 5 = Never. Higher scores are an indicator of shorter time for social contact. A lack of social activities also was measured by four items asking the participants whether they have done paid work, volunteering, and religious services, and whether they have attended clubs or other organized activities either online or in-person during the pandemic (1 = Yes and 0 = No). Higher scores indicate that the participants have engaged in fewer or no social activities.

Healthcare delay was assessed by one item asking if the participant had ever put off their plan to see a doctor or other healthcare provider during the pandemic (1 = Yes and 0 = No). Lastly, a lack of time for self-care was measured by one item asking how often the participants felt they could not get any time to themselves in a typical week during the pandemic. A five-point Likert rating was given as the response option (1 = Never to 5 = Everyday). The higher the scores were, the shorter time the participants had for self-care.

Analytic procedure

STATA 16.0 was used throughout the analytical procedures. First, the caregiver group was compared with the non-caregiver group with regard to their mean scores for the independent and dependent variables. For this purpose, an independent sample t-test was done, and then the homogeneity of variance of the two groups was tested. Secondly, Pearson correlations were examined among those variables in each group. Finally, six independent variables were included in the multiple linear regression analysis to examine if the variables would predict the participants’ depressive symptom during the pandemic and to identify any meaningful differences between the two groups. Demographic characteristics including age, gender, and race or ethnicity were treated as control variables.

Results

Sample characteristics

Out of the six ranges of age, the most frequent range was 75–79, and the second frequent one was 70–74 among all participants. Those in their 70s were 68.2% in the caregiver group compared to 54% in the non-caregiver group. As for gender, females accounted more for the participants than males in both groups. Female participants were 59.7% and 52.9% in the caregiver and non-caregiver group respectively. When race or ethnicity was examined, White participants made up 75.4% of the caregiver group and 83.2% of the non-caregiver group. Black participants were 16.5% in caregiver group whereas they were 11.1% in the non-caregiver group. Asian, other Pacific Islanders, and Hawaiian Natives in the caregiver group (2.8%) were almost twice those in the non-caregiver group (1.5%).

Mean differences in major variables between caregiver and non-caregiver group

The authors confirmed that the homogeneity of variance was not violated in the independent sample t-test to compare the caregiver with non-caregiver group. The test results presented in Table 2 showed that the dependent variable, depressive symptom, did not significantly differ between the two groups although the caregiver group had slightly a higher mean score. The two groups showed no significant differences in healthcare delay and a lack of social activities either.

Table 2. Descriptive statistics in major variables

	All (n = 1,685)		Caregivers (n = 248)		Non-Caregivers (n = 1,437)			t
Variable	M	SD	M	SD	M	SD	Cohen’s d
Number of missed events	3.28	2.5	3.75	2.59	3.20	2.51	.2	.66**
Time spent staying home	4.25	2.94	3.94	2.61	4.31	3.00	−.1	7.14*
A lack of social contact	8.52	3.08	7.85	2.96	8.63	3.09	−.3	.24***
A lack of social activities	3.25	.92	3.19	.87	3.27	.93	−.1	1.29
Healthcare delay	.40	.49	.44	.50	.39	.49	.1	6.03
A lack of time for self-care	.50	.74	.77	.96	.46	.69	.4	51.88***
Depressive symptom	.96	.79	.97	.74	.96	.79	.0	1.96

Differences between the caregiver and non-caregiver group were identified by an independent sample t-test. ***p<.001. **p< .01. *p< .05.

In contrast, statistically significant differences were found in the mean scores of four independent variables. In two variables, the number of missed events and a lack of time for self-care, did the caregiver group show higher scores than the non-caregiver group. On the contrary, mean scores in time spent staying home and a lack of social contact were higher in the non-caregiver group than they were in the caregiver group. Cohen’s d indicated that the group differences in time spent staying home and the number of missed events had trivial to small effect sizes. On the othe hand, the group differences in a lack of social contact and alack of time for self-care showed larger effect sizes.

Correlations among major variables

Pearson correlation coefficients in each group are described in Table 3. There were commonalities between the two groups. For instance, a positive correlation was found between time spent staying home and a lack of social contact as well as between time spent staying home and a lack of social activities. In addition, depressive symptom was positively correlated with healthcare delay and a lack of time for self-care.

Table 3. Intercorrelations between major variables

Variable	1	2	3	4	5	6	7
1. Number of missed events	–	−.027	−.205***	−.180***	.139***	.073**	.132***
2. Time spent staying home	−.119	–	.276***	.266***	.054*	−.031	.112***
3. A lack of social contact	−.189**	.295***	–	.218***	−.067*	−.024	−.042
4. A lack of social activities	−.160*	.229***	.082	–	−.013	−.033	.027
5. Healthcare delay	.131*	.066	−.107	−.048	–	.073**	.192***
6. A lack of time for self-care	.002	.176**	.101	−.034	.069	–	.243***
7. Depressive symptom	.034	.020	−.034	.047	.280***	.292***	–

Pearson correlation coefficients are shown for caregivers (n = 248) below the diagonal and for non-caregivers (n = 1,437) above the diagonal.

***p < .001. **p < .01. *p < .05.

Several differences were found between the two groups as well. A positive correlation was found only in the non-caregiver group between a lack of time for self-care and the number of missed events as well as between a lack of time for self-care and healthcare delay. The non-caregiver group also demonstrated a positive correlation between depressive symptom and the number of missed events as well as depressive symptom and time spent staying home. In addition, a lack of social contact and a lack of social activities were positively correlated,and time spent staying home and healthcare delay were also positively correlated in this group. On the contrary, a positive correlation between time spent staying home and a lack of time for self-care was found only in the caregiver group. Out of all coefficients, the highest was found between time spent staying home and a lack of social contact in the caregiver group.

Stressors associated with depressive symptom

The data did not produce extreme numbers in skewness or kurtosis, meaning they were normally distributed (Hopkins & Weeks, 1990). Therefore, the authors proceeded with a multiple linear regression analysis separately with the two groups of sample. The analysis demonstrated that the COVID-19-related stressors explained a significant extent of depressive symptom in both groups, though those stressors had a greater explanatory power for the caregiver group (18%) than for the non-caregiver group (12%). Analysis outcoems of each group are described in Tables 4 and 5 respectively. In both tables, Model 1 displayed control variables (i.e., age, gender, and race/ethnicity), and Model 2 displayedindependent variables in addition to the control variables. . As for demographic characteristics , being Black (β = −.20, p < .01) was a statistically significant predictor of lower depressive symptom in the caregiver group. On the other hand, being female (β = .13, p < .001) was a statistically significant predictor of higher depressive symptom in the non-caregiver group.

Table 4. Multiple linear regression coefficients among caregivers (n = 248)

Variable	Model 1				Model 2
	B	β	SE	p-value	B	β	SE	p-value
Constant	.85		.15	.000	.44		.27	.099
Age	.01	.01	.04	.885	.01	.01	.04	.853
Female ^a	.28	.18**	.09	.003	.16	.10	.09	.097
Black ^b	−.44	−.22***	.12	.000	−.39	−.20**	.12	.002
Hispanic or Latino ^b	.17	.04	.26	.507	.28	.07	.25	.253
Asian or Pacific Islanders or Hawaiian Natives ^b	.11	.03	.28	.692	−.07	−.02	.27	.799
Other race ^b	−.18	−.04	.33	.577	−.16	−.03	.31	.596
Number of missed events					.00	.01	.02	.865
Time spent staying home					−.01	−.04	.02	.583
A lack of social contact					−.01	−.03	.02	.636
A lack of social activities					.08	.09	.05	.141
Healthcare delay					.35	.23***	.09	.000
A lack of time for self-care					.21	.28***	.05	.000
R^2	.09				.22
Adjusted R^2	.06				.18

^aReference group = Male.

^bReference group = White.

B = Unstandardized Beta, β = Standardized Coefficients Beta

***p < .001. **p < .01. *p < .05.

Table 5. Multiple linear regression coefficients among non-caregivers (n = 1,437)

	Model 1				Model 2
Variable	B	β	SE	p-value	B	β	SE	p-value
Constant	.76		.06	.000	.40		.11	.000
Age	.01	.02	.02	.466	.01	.02	.02	.392
Female ^a	.30	.19***	.04	.000	.20	.13***	.04	.000
Black ^b	−.04	−.02	.07	.515	−.04	−.02	.07	.557
Hispanic or Latino ^b	.17	.04	.12	.159	.07	.01	.12	.568
Asian or Pacific Islanders or Hawaiian Natives ^b	.13	.02	.17	.436	−.02	−.01	.17	.883
Other race ^b	−.18	−.03	.19	.340	−.15	−.02	.18	.406
Number of missed events					.03	.09***	.01	.000
Time spent staying home					.02	.08*	.01	.010
A lack of social contact					−.01	−.03	.01	.238
A lack of social activities					.03	.03	.02	.216
Healthcare delay					.23	.14***	.04	.000
A lack of time for self-care					.26	.22***	.03	.000
R^2	.04				.13
Adjusted R^2	.03				.12

^aReference group = Male

^bReference group = White.

B = Unstandardized Beta, β = Standardized Coefficients Beta

***p < .001. **p < .01. *p < .05.

Among the six independent variables, two of them were found to be statistically significant stressors associated with depressive symptom in both groups, which were healthcare delay (caregivers: β = .23, p < .001; non-caregivers: β = .14, p < .001) and a lack of time for self-care (caregivers: β = .28, p < .001; non-caregivers: β = .22, p < .001). In addition to these stressors, the non-caregiver group had two more significant stressors associated with depressive symptom, which were the number of missed events (β = .09, p < .001) and time spent staying home (β = .08, p < .05).

Discussion

The results provided significant findings about the research questions. As far as the participants’ demographic features were concerned, the ratio of female to male was 60 to 40 in the caregiver group, which aligned with previous research findings. For instance, it was reported that almost two-thirds of family caregivers were female (Feinberg, Reinhard, Houser, & Choula, 2011). Additionally, the literature shows that race or ethnicity was associated with caregiver burden (Chiao et al., 2015). Black caregivers spent more time in caregiving per week than White caregivers (Skarupski, McCann, Bienias, & Evans, 2009), but impressively, they tended to perceive lower burden than White caregivers, which would be attributed to their religious coping strategies (Sun, Kosberg, Leeper, Kaufman, & Burgio, 2010). Black caregivers in this study also were less likely to have depressive symptom.

With regard to the first research question to identify significant stressors associated with older adults’ depressive symptom during the pandemic, two stressors were highlighted in both groups. First , higher scores in healthcare delay were a significant predictor of greater depressive symptom. It is well-known that healthcare delay has been a widespread stressor as well as a threat to well-being in the general population since the COVID-19 outbreak. It was reported that 41% of American adults experienced delayed or avoided medical care out of concerns about COVID-19, and 12% of the delayed medical care was urgent or emergency (Czeisler et al., 2020). Changes in the healthcare operation were another cause for healthcare delay in addition to the fear of contagionas medical resources and capacities were highly affected by the pandemic situation (Leite, Lindsay, & Kumar, 2020). The deleterious consequences of healthcare delay can include a delayed diagnosis of fatal illnesses and accompanying deterioration of health conditions (Papautsky et al., 2021). Maintaining health is particularly important for older adults because many of them are with comorbidities (Vitaliano, Zhang, & Scanlon, 2003).

Therefore, a provision of state and national support systems to counteract this crisis is needed. Utilizing telehealth can be one way to provide healthcare services in a contact-free manner. The telehealth market has been growing rapidly in recent years, but COVID-19 has certainly enhanced the progress. Medicare primary care visits through telehealth have surged from 0.1% in February 2020 to 43.5% in April 2020 (Bosworth et al., 2020). As older adults generally are less familiar with utilizing new technology, supports to help them better utilize the technology are recommended.

Second, the shorter time the participants had for self-care, the more depressed they were. This is in accordance with the existing literature because poor self-care was associated with psychological distress among various populations (Brouwer et al., 2021; Slonim et al., 2015). It was also found that the caregiver group showed significantly higher scores in the lack of time for self-care than non-caregiver group. Self-care includes activities independently carried out to promote and maintain personal well-being (Orem, Taylor, & Renpenning, 1995). Older caregivers are likely to have less time for self-care than older adults without a caregiving role since their time and energy should be used for their care recipients.

On the other hand, two stressors significantly associated with depressive symptom only in the non-caregiver group were the number of missed events and time spent staying home. Both of these stressors were related to social distancing policy, which was implemented to encourage people to distance from others to slow the spread of coronavirus. Therefore, the prolonged implementation of it could have led to social isolation as well as constraints on daily activities. Social isolation was known to be a predictor ofolder adults’ morbidities and mortalities (Holt-Lunstad, Robles, & Sbarra, 2017), and perceived social isolation was related to their depressive symptoms (Taylor, Taylor, Nguyen, & Chatters, 2018). In fact, 42% of older respondents in a study considered reduced social interaction as the biggest challenge caused by the pandemic (Heid, Cartwright, Wilson-Genderson, & Pruchno, 2021). Since the lack of social contact particularly impacted depressive symptom of the older adults without a aregiving role in the present study, opportunities to engage them in social contact with minimal risks of contracting coronavirus should be developed.

The second research question was whether older caregiversexperienced greater stress and depressive symptom than older adults without a caregiving role. The results indicated that there was no significant difference in the level of depressive symptom between the caregiver and non-caregiver group. Also, there were only two significant stressors associated with depressive symptom in the caregiver group, compared to four significant stressors in the non-caregiver group. Thus, the answer to this question would be that caregiving responsibilities did not lead to greater stress or depressive symptom in the older population.

When previous studies that had compared psychological well-being of caregivers with that of non-caregivers were reviewed, the findings seemed mixed. For instance, the caregiving role was known to be associated with psychological symptoms and poor health outcomes such as depression, anxiety, overwhelmed or abandoned feelings, and irritability in a number of studies (Chiao et al., 2015; Given, Given, & Kozachik, 2001; Judge et al., 2011; Pinquart & Sörensen, 2003; Vitaliano et al., 2003). However, Sibalija, Savundranayagam, Orange, and Kloseck (2020) reported that no group differences in depressive symptoms were observed between caregivers and non-caregivers, which aligned with the results of the present study.

This finding highlights the fact that albeit the well-known caregiver burden (Vitaliano et al., 2003), positive effects of the caregiving role also prevail. Older caregivers’ resilience can be explained by several theories. Jones, Winslow, Lee, Burns, and Zhang (2011) have developed the Caregiver Empowerment Model (CEM) to explain the caregiving role in yielding positive outcomes among caregivers. Similar to the role enhancement perspective that emphasized productive roles reinforcing the well-being of older adults (Rozario, Morrow-Howell, & Hinterlong, 2004), the CEM posited that a caregiving role would offer personal rewards and a sense of fulfillment, improving caregivers’ psychological well-being (Jones et al., 2011; Pierce, Steiner, Govoni, Thompson, & Friedemann, 2007). These theories were based on a belief that older adults experienced immense changes in work and social roles in the transition from midlife, so they were prone to feel psychological distress stemming from role loss. In this sense, involvement in productive activities such as caregiving could be particularly beneficial for them

In addition to the CEM, the role of social participation should be considered. Because human beings are inherently social, the extents of social support and social participation are associated with depression (Sibalija et al., 2020). For this reason, social distancing measures ironically gave rise to a concern about its effects on psychological well-being although they were implemented for public health purposes. It can be expected that older caregivers were able to consistently interact with their care recipients and thus experienced less stress from social distancing than older adults without a caregiving role. This difference possibly mitigated the negative effects of social distancing on older caregivers’ depressive symptom. In support of this argument, the results of this study showed the non-caregiver group had a higher mean score in the lack of social contact compared to the caregiver group, indicating older adults had more social contact when they had a caregiving role.

In sum, the present study is meaningful in that it examined major challenges faced by older caregivers and their psychological well-being. When older adults were generally seen as care recipients rather than caregivers (Spatuzzi et al., 2021), the present study shed light on older adults who provided informal caregiving. Although older caregivers in this study did not demonstrate more vulnerabilities than older adults without a caregiving role, it is crucial to be attentive to older caregivers’ needs as their number will increase with the population aging, especially in times of public health crisis such as COVID-19 pandemic. Considering 89% of scientists expecting COVID-19 to become endemic according to a survey conducted by Nature, and many scientists fearing another pandemic to arise in the future (Phillips, 2021), the results of the present study will help develop precautionary measures to support older caregivers in crisis situations to come.

Limitations

The present study is not without limitations. First, a cross-sectional dataset might not capture the full impacts of the stressors on depressive symptom since the pandemic is ongoing. Once more data are accumulated, future studies are recommended to investigate long-term changes in older adults’ psychological well-being, using a longitudinal dataset.

Second, the heterogeneity of older caregivers was not considered adequately in the present study. For example, older caregivers’ situations might vary, depending on the number of hours spent caregiving per week, the type of illnesses their care recipients have, and other responsibilities they might have. Information regarding the diverse surroundings of older caregivers should be collected and considered as critical factors in future research.

Conclusion

Despite its limitations, the present study is significant because it contributes to understanding older adults’ stressors associated with depressive symptom during the COVID-19 pandemic. Its findings highlight the importance of providing older adults with resources to address delayed healthcare and little time for self-care. It also supports existing perspectives that focus on the positive aspects of a caregiving role. Continued attention from service providers and policy makers is needed tomental health among vulnerable populations, including older adults, who are experiencing multiple stressors in this era of public health crisis.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This study was not funded by a specific grant.