- Split View
-
Views
-
Cite
Cite
Yaeji Kim-Knauss, Frieder R Lang, Fiona S Rupprecht, Kristina Martin, Helene H Fung, COVID-19 Worries Predict Aging Preparation: Culture- and Domain-Specific Perspectives, The Journals of Gerontology: Series B, Volume 77, Issue 10, October 2022, Pages 1803–1813, https://doi.org/10.1093/geronb/gbac078
- Share Icon Share
Abstract
We investigated whether worrying about coronavirus disease 2019 (COVID-19) predicts people’s engagement in aging preparation. Furthermore, we expected that this association would have culture- (i.e., Hong Kong and Germany) and domain-specific (i.e., finances, housing, care needs, connectedness, and end-of-life) tendencies, as the culture and domains that are most severely hit by the pandemic differ.
A total of 360 and 1,294 adults (aged 18–98 years) living in Hong Kong and Germany, respectively, participated in a web-based study. We fitted our data to a multilevel model in order to take into account the interdependence of domains (i.e., Level 1) within the same individual (i.e., Level 2).
The results revealed that reporting higher COVID-19 worries were associated with pandemic-induced aging preparation, and this association was particularly apparent for Germans in comparison to those from Hong Kong. When domains were specified, this cultural difference appeared significantly stronger for the domains of care, connectedness, and end-of-life than finances and housing.
Findings imply that worrying about the COVID-19 pandemic predicts people to engage in aging preparation particularly in the culture and domains most affected by the pandemic. These results from those worried about the virus may be attributed to the increased self-relevance to the topic and hence motivation.
The coronavirus disease 2019 (COVID-19) has become a global pandemic in 2020 and fundamentally shifted our way of life. This pandemic not only threatens our health but also restricts our social relations as a result of the government-mandate social distancing restrictions. Moreover, it resulted in economic crises, such as job loss, which should be related to financial insecurities. Although COVID-19 jeopardizes the health, social relations, and financial situation and therefore disrupts individual well-being (Kivi et al., 2021; O’Connor et al., 2021), we submit that it may also provoke counteraction, such as preparatory or preventive behaviors, by leading us to envisage what it would be like to encounter such challenges in other life situations (i.e., proactive coping: Aspinwall & Taylor, 1997).
Given that physical, financial, and social challenges often occur as we age (Owen & Wu, 2007; Weeks, 1994), the experience with the pandemic might play a role as a situational cue and thereby motivate our engagement in aging preparation (i.e., pandemic-induced aging preparation). Those who are more worried about the virus may more likely be subject to a counteractive effect of the pandemic, as having more worries might increase people’s awareness of life’s challenges, in particular with regard to old age and the finitude of one’s life. We also consider the roles of culture (i.e., Hong Kong and Germany) and life domains (i.e., finances, housing, care needs, social connectedness, and end-of-life) in the association between COVID-19 worries and pandemic-induced aging preparation. We suppose that people in a culture that is more affected by the virus would be more willing to mitigate pandemic-related shocks than those in a less affected culture. As the domains that are most severely hit by the pandemic also differ by culture, we additionally investigate whether there are any domain-specific tendencies in the combined effects of culture and COVID-19 worries on aging preparation.
COVID-19 worries indicate the extent to which we are concerned about being infected and about our health deteriorating from the virus. Although the very nature of worry implies rather negative thoughts and feelings about the future, this affective response has been found to be positively related to subsequent engagement in preventive behaviors in nonpandemic contexts (McCaul & Mullens, 2003; Schmiege et al., 2009). Within the context of COVID-19, specifically, the motivational roles of worry and fear on people’s compliance with government-mandated restrictions have been well reported (Barber & Kim, 2021; Zhang et al., 2020). These results might be attributed to an increased self-relevance for those who are worried about the virus, which in turn encourages their engagement in safety-promoting behaviors (Harper et al., 2020).
The present study aims to expand the discussions about worry and explore its impact on aging preparation, which requires future-oriented thinking. Aging is related to challenges that occur in diverse life domains. Hence, individuals adjust their plans and engage in preparatory activities to cope with any anticipated losses and thereby achieve positive development (i.e., proactive coping). Such coping efforts undertaken in advance can be triggered by the recognition that physical, financial, and social challenges do occur (Aspinwall & Taylor, 1997). In light of terror management theory (Solomon et al., 1991), people may become aware of life’s finitude when experiencing severe illness, a natural disaster, or when only walking by a funeral home (Jonas et al., 2002). In a pathological sense, being infected with the coronavirus could make an individual sick and fragile and can even cost life if the individual develops severe symptoms. Having self-relevant worries about COVID-19 may, therefore, serve as a reminder of the finitude of life and thereby become an emotional burden that motivates proactive coping. One study found that those who perceived the virus as a health threat reported decreased future time perspective (Rupprecht et al., 2021). In this vein, it is plausible that worrying about COVID-19 might alter our stance toward the future and induce a higher intention to engage in aging preparation than before.
In addition, we postulate that the association between worries and pandemic-induced aging preparation might differ by culture. Our study particularly concerns the experience of adults from Germany and Hong Kong who might be differentially affected by this situation. For instance, although testing capabilities and reporting accuracy might have affected the numbers, the cumulative number of COVID-19 infections, as of September 2021, has been estimated as 50,927 cases and 1,120 deaths per million in Germany and 1,620 cases and 28 deaths per million in Hong Kong (Ritchie et al., 2020). Other indicators, such as the excess mortality during the pandemic that refers to the number of deaths from all causes exceeding the expected number of deaths under “normal” conditions, also showed a higher number in Germany (+3.4%) than in Hong Kong (+2.3%) on average until September 2021. Such differences in the numbers might engender different reactions to the pandemic. We suppose that, in Hong Kong, COVID-19 might be less likely to serve as a reminder about challenges in old age, and thus the association between worries and aging preparation might be weaker than in Germany. However, so far, only little is known about the role of culture on pandemic-induced behaviors and psychological reactions (Ruiz et al., 2021).
Given that the different experiences with the pandemic might regulate people’s behaviors, we further hypothesize that the role of culture should be subject to domain specificity (i.e., finances, housing, care needs, social connectedness, and end-of-life). Germany, compared with Hong Kong, has been more affected by the adverse outcomes of COVID-19 on the health of its residents. It implies that the topics of need for care or end-of-life might be of greater interest for those living under these circumstances than other life domains. Moreover, having experienced the Severe Acute Respiratory Syndrome epidemic in 2003, Hong Kong implemented a series of containment measures immediately after the official announcement of atypical pneumonia cases in Wuhan on December 31, 2019 (Cheng et al., 2020; Hung, 2003). Germany enacted nationwide stay-at-home regulations during each of the waves at a stricter level than Hong Kong (Reuters, 2022). Besides, as German residents are new to the pandemic situation and state-ordered restrictions on social gatherings, they might become more aware of social challenges than before. In fact, in an empirical study based in Luxembourg, whose infection rate (123,847 per million) and mortality rate (1,315 per million) are more comparable to Germany than Hong Kong (Ritchie et al., 2020), people’s perceptions of aging as physical and social losses were particularly affected during the COVID-19 pandemic (Kornadt et al., 2021). On the other hand, those from Hong Kong, where the economy was already reeling from the protests since 2019, might be susceptible to financial challenges. The Hong Kong economy contracted at a record high in 2020 and the unemployment rate rose to a 16-year high of 6.6% (International Monetary Fund, 2021a), while the German economy was less severely affected by the pandemic perhaps due to the German government’s expansive fiscal response to it such as tax cut and short-term work subsidies (International Monetary Fund, 2021b).
Taken together, we hypothesize that those who are more worried about COVID-19 will be more likely to deal with aging preparation because the life challenges associated with COVID-19 might remind them of the vulnerability in old age and life’s finitude. Specifically, this association is expected to be stronger for Germans than for Hong Kong Chinese as the number of infection cases and deaths from the virus are higher for the former culture, but this hypothesized cultural difference might show domain-specific tendencies. For instance, due to the higher prevalence and mortality rate of COVID-19, Germans may be more interested in health-relevant life domains such as end-of-life, while those in Hong Kong may be more encouraged to protect themselves from constraints in the financial domain.
Method
Participants
The data come from an online study of the “Ageing as Future (AAF)” project aimed at investigating future-related activities regarding age and aging in cross- cultural and domain-specific perspectives. The study has gone through many waves, and the latest wave was conducted in Germany (n = 1,294; aged 18−95) from November to December in 2020 and in Hong Kong (n = 360; aged 20−98) from February to April in 2021. Because only this latest wave included pandemic-related constructs, we used only data from this wave for analyses. In Hong Kong, the participants of the prior waves of the study were followed up, and new participants were recruited only to replace those who dropped out, taking into account age and gender distribution (nfollow-up = 273; nnew = 87). Although recruitment methods were comparable to Hong Kong, recruitment in Germany was expanded by including the sample from a newly added module focused on COVID-19 experiences in the existing AAF longitudinal study (nfollow-up = 479; nnew = 815; Kim-Knauss & Lang, 2021; Rupprecht et al., 2021). This accounts for the differences in sample size. All participants who provided consent received a link to access this study by email. Those who completed the questionnaire received a compensation of 20 euros in Germany and 200 Hong Kong dollars in Hong Kong, corresponding to approximately 20 and 25 U.S. dollars, respectively. Data were assessed in compliance with the ethical rules and data protection regulations provided by the Federal State of Bavaria in Germany and by the Survey and Behavioral Research Ethics Committee of the Chinese University of Hong Kong (SBRE-19-171).
The sample characteristics are presented in Table 1. As the sample was not inherently designed to be representative of the entire population of Germany and Hong Kong, we compared the sociodemographic characteristics of our sample with those of the general populations and statistically controlled their effects in the subsequent analyses (Destatis, 2021; Hong Kong Census and Statistics Department, 2021). The average age of the entire sample was around 57 years with no significant difference between the German and Hong Kong groups (n.s.). Compared with the general population, those aged 65 and older were overrepresented in both sample groups. Of the total sample, 1,004 (60.7%) were female and nine participants (0.5%) reported their gender as nonbinary. All the nonbinary individuals belong to the German sample, and the female proportion was significantly higher in Germany than in Hong Kong (p < .001). Compared with the general population, females less than 65 age group were overrepresented in Germany, while males in both less than 65 and more than 65 age groups were overrepresented in Hong Kong. A total of 1,155 participants (69.8%) reported that they were in a steady partnership. The participants from Hong Kong were more likely to have a partner than those from Germany (p < .001). Participants reported their subjective health on a six-point Likert scale from “(1) bad” to “(6) excellent.” The mean score of subjective health was 3.94, representing that participants perceived their health status as “fair,” and this score was significantly higher for Germans than Hong Kong Chinese (p < .001). Finally, we measured participants’ education level using the International Standard Classification of Education 2011 (ISCED 2011) on an eight-point scale ranging from “(1) primary education” to “(8) doctoral or equivalent level.” The median education level was “(6) bachelor or equivalent level” for the entire sample, “(6) bachelor or equivalent level” for the German sample, and “(3) upper secondary education” for the Hong Kong sample. Our German sample was more highly educated compared to the general population, while the Hong Kong sample was representative of the general population.
. | Total sample . | Hong Kong . | Germany . | . |
---|---|---|---|---|
. | (N =1,654) . | (n = 360) . | (n = 1,294) . | . |
N (%) | N (%) | N (%) | Δ Subsamples | |
Female | 1,004 (60.7) | 183 (50.8) | 821 (63.5) | 𝑥2(1) = 18.26*** |
Nonbinary | 9 (0.5) | 0 (0) | 9 (0.7) | 𝑥2(1) = 1.40 |
Steady partnership | 1,155 (69.8) | 291 (80.8) | 864 (66.8) | 𝑥2(1) = 25.78*** |
M (SD) | M (SD) | M (SD) | Δ Subsamples | |
Pandemic-induced aging preparation | ||||
Finances | 2.80 (1.84) | 4.37 (1.53) | 2.37 (1.68) | t(1,652) = 20.39*** |
Housing | 2.81 (1.96) | 4.28 (1.73) | 2.40 (1.82) | t(1,652) = 17.56*** |
Care | 2.82 (1.84) | 4.16 (1.60) | 2.44 (1.73) | t(1,652) = 16.89*** |
Connectedness | 3.31 (2.03) | 4.06 (1.62) | 3.11 (2.09) | t(1,652) = 8.00*** |
EoL | 3.10 (1.88) | 3.88 (1.62) | 2.88 (1.89) | t(1,652) = 9.13*** |
COVID-19 worries | 2.83 (0.96) | 3.00 (1.03) | 2.78 (0.94) | t(1,652) = 4.02*** |
Age | 57.46 (17.38) | 57.50 (17.34) | 57.45 (17.40) | t(1,652) = 0.05 |
Subjective health | 3.94 (1.06) | 3.78 (0.86) | 3.98 (1.10) | t(1,652) = −3.21** |
ISCED | 5.23 (1.99) | 3.78 (1.93) | 5.63 (1.82) | t(1,652) = −16.85*** |
. | Total sample . | Hong Kong . | Germany . | . |
---|---|---|---|---|
. | (N =1,654) . | (n = 360) . | (n = 1,294) . | . |
N (%) | N (%) | N (%) | Δ Subsamples | |
Female | 1,004 (60.7) | 183 (50.8) | 821 (63.5) | 𝑥2(1) = 18.26*** |
Nonbinary | 9 (0.5) | 0 (0) | 9 (0.7) | 𝑥2(1) = 1.40 |
Steady partnership | 1,155 (69.8) | 291 (80.8) | 864 (66.8) | 𝑥2(1) = 25.78*** |
M (SD) | M (SD) | M (SD) | Δ Subsamples | |
Pandemic-induced aging preparation | ||||
Finances | 2.80 (1.84) | 4.37 (1.53) | 2.37 (1.68) | t(1,652) = 20.39*** |
Housing | 2.81 (1.96) | 4.28 (1.73) | 2.40 (1.82) | t(1,652) = 17.56*** |
Care | 2.82 (1.84) | 4.16 (1.60) | 2.44 (1.73) | t(1,652) = 16.89*** |
Connectedness | 3.31 (2.03) | 4.06 (1.62) | 3.11 (2.09) | t(1,652) = 8.00*** |
EoL | 3.10 (1.88) | 3.88 (1.62) | 2.88 (1.89) | t(1,652) = 9.13*** |
COVID-19 worries | 2.83 (0.96) | 3.00 (1.03) | 2.78 (0.94) | t(1,652) = 4.02*** |
Age | 57.46 (17.38) | 57.50 (17.34) | 57.45 (17.40) | t(1,652) = 0.05 |
Subjective health | 3.94 (1.06) | 3.78 (0.86) | 3.98 (1.10) | t(1,652) = −3.21** |
ISCED | 5.23 (1.99) | 3.78 (1.93) | 5.63 (1.82) | t(1,652) = −16.85*** |
Notes: Finances, housing, care, connectedness, and EoL stand for pandemic-induced aging preparation for financial situation in old age, housing arrangements in old age, need for nursing care, social connectedness in old age, and end of life, respectively. ISCED = International Standard Classification of Education; SD = standard deviation.
**p < .01. ***p < .001.
. | Total sample . | Hong Kong . | Germany . | . |
---|---|---|---|---|
. | (N =1,654) . | (n = 360) . | (n = 1,294) . | . |
N (%) | N (%) | N (%) | Δ Subsamples | |
Female | 1,004 (60.7) | 183 (50.8) | 821 (63.5) | 𝑥2(1) = 18.26*** |
Nonbinary | 9 (0.5) | 0 (0) | 9 (0.7) | 𝑥2(1) = 1.40 |
Steady partnership | 1,155 (69.8) | 291 (80.8) | 864 (66.8) | 𝑥2(1) = 25.78*** |
M (SD) | M (SD) | M (SD) | Δ Subsamples | |
Pandemic-induced aging preparation | ||||
Finances | 2.80 (1.84) | 4.37 (1.53) | 2.37 (1.68) | t(1,652) = 20.39*** |
Housing | 2.81 (1.96) | 4.28 (1.73) | 2.40 (1.82) | t(1,652) = 17.56*** |
Care | 2.82 (1.84) | 4.16 (1.60) | 2.44 (1.73) | t(1,652) = 16.89*** |
Connectedness | 3.31 (2.03) | 4.06 (1.62) | 3.11 (2.09) | t(1,652) = 8.00*** |
EoL | 3.10 (1.88) | 3.88 (1.62) | 2.88 (1.89) | t(1,652) = 9.13*** |
COVID-19 worries | 2.83 (0.96) | 3.00 (1.03) | 2.78 (0.94) | t(1,652) = 4.02*** |
Age | 57.46 (17.38) | 57.50 (17.34) | 57.45 (17.40) | t(1,652) = 0.05 |
Subjective health | 3.94 (1.06) | 3.78 (0.86) | 3.98 (1.10) | t(1,652) = −3.21** |
ISCED | 5.23 (1.99) | 3.78 (1.93) | 5.63 (1.82) | t(1,652) = −16.85*** |
. | Total sample . | Hong Kong . | Germany . | . |
---|---|---|---|---|
. | (N =1,654) . | (n = 360) . | (n = 1,294) . | . |
N (%) | N (%) | N (%) | Δ Subsamples | |
Female | 1,004 (60.7) | 183 (50.8) | 821 (63.5) | 𝑥2(1) = 18.26*** |
Nonbinary | 9 (0.5) | 0 (0) | 9 (0.7) | 𝑥2(1) = 1.40 |
Steady partnership | 1,155 (69.8) | 291 (80.8) | 864 (66.8) | 𝑥2(1) = 25.78*** |
M (SD) | M (SD) | M (SD) | Δ Subsamples | |
Pandemic-induced aging preparation | ||||
Finances | 2.80 (1.84) | 4.37 (1.53) | 2.37 (1.68) | t(1,652) = 20.39*** |
Housing | 2.81 (1.96) | 4.28 (1.73) | 2.40 (1.82) | t(1,652) = 17.56*** |
Care | 2.82 (1.84) | 4.16 (1.60) | 2.44 (1.73) | t(1,652) = 16.89*** |
Connectedness | 3.31 (2.03) | 4.06 (1.62) | 3.11 (2.09) | t(1,652) = 8.00*** |
EoL | 3.10 (1.88) | 3.88 (1.62) | 2.88 (1.89) | t(1,652) = 9.13*** |
COVID-19 worries | 2.83 (0.96) | 3.00 (1.03) | 2.78 (0.94) | t(1,652) = 4.02*** |
Age | 57.46 (17.38) | 57.50 (17.34) | 57.45 (17.40) | t(1,652) = 0.05 |
Subjective health | 3.94 (1.06) | 3.78 (0.86) | 3.98 (1.10) | t(1,652) = −3.21** |
ISCED | 5.23 (1.99) | 3.78 (1.93) | 5.63 (1.82) | t(1,652) = −16.85*** |
Notes: Finances, housing, care, connectedness, and EoL stand for pandemic-induced aging preparation for financial situation in old age, housing arrangements in old age, need for nursing care, social connectedness in old age, and end of life, respectively. ISCED = International Standard Classification of Education; SD = standard deviation.
**p < .01. ***p < .001.
Measurements
Pandemic-induced aging preparation
Building on a domain-specific perspective on aging preparation (Kornadt & Rothermund, 2014; Lang & Rupprecht, 2021), we asked participants to what extent they dealt with the topic of aging preparation in five different domains (i.e., finances, housing, care needs, social connectedness, and end-of-life) as a result of the pandemic. Preparation encompasses diverse levels of engagement, such as awareness, deliberation, planning, information-seeking, and implementation (Kornadt & Rothermund, 2014; Sörensen & Pinquart, 2001), and concrete preparatory activities might not take place within a short period of time. We have therefore formulated questions to include a broad spectrum of preparatory activities (e.g., deliberation). The questions were structured in the same format, but their target domains were specific. The following five items were asked: “Due to the Coronavirus pandemic, I am dealing with the topic of … (financial preparation/living situation in old age/need for nursing care/loneliness in old age/my own dying and death) more intensively than before” on a scale anchored by “(1) Does not apply at all” and “(7) Applies very much.” The internal consistency of this scale was good (Cronbach’s α = 0.86).
COVID-19 worries
The construct of COVID-19 worries concerns whether people perceive this disease as a health threat for themselves. Referring to a study on dementia worry (Martin et al., 2021), we compiled five items that had been used in the worry literature (e.g., Werner, 2002) and adapted them to fit into the context of COVID-19. The following five items were asked: “I am worried about my health,” “The thought of being personally affected by COVID-19 is threatening,” “How concerned are you about getting sick with COVID-19?” “How much does the idea of getting sick with COVID-19 currently occupy your thoughts?” and “Are you afraid of COVID-19 related to yourself?” Respondents answered on a scale anchored by “(1) Not at all” and “(5) Very much.” The internal consistency of this scale was high (Cronbach’s α = 0.92).
Analytic procedures
Our data formed a two-level structure. As the domain-specific information is nested in individuals, we fitted our data to a multilevel model that simultaneously estimates both level variances in the outcome variable (i.e., domain- and individual-level), while taking into account the interdependence of data within the same individual (Lang & Rupprecht, 2021). Domain-level variables (Level 1) included pandemic-induced aging preparation and four dummy variables indicating the four domains (i.e., housing, care, social connectedness, and end-of-life) other than finances, which served as the reference category. Individual-level variables (Level 2) included culture with Germany serving as the reference category, COVID-19 worries, calendar age, gender, relationship status, subjective health, and education level.
First of all, an intercept-only model was fitted to estimate the extent to which the variance of the outcome variable was accounted for by Level 1 and Level 2. We then built hierarchical models where the marginal R2 increased as more predictors were added. Domain information was entered into Model 1, which was followed by Model 2 where the individual-level main predictor (i.e., worries) and covariates were included. The two-way interaction term between worries and Hong Kong (in comparison to Germany) was included in Model 3, and finally, a cross-level three-way interaction term was entered in Model 4 to investigate the differential effects of worries by culture by domain. Note that once the association is specified by domain, the original coefficient of worries by culture is no longer a domain-general average of the given association, but that of the reference domain, namely, financial preparation (i.e., Worries:HongKong = Worries:HongKong:Finances). The domain-specified coefficients, on the other hand, describe deviations from this reference domain coefficient instead of the total effects of the respective domains (e.g., Worries:HongKong:Housing = Worries:HongKong + Worries:HongKong:Housing). For Model 4, we tested the two-way interaction terms of worries by domain and culture by domain, but they neither changed the effects of the other parameters nor were they relevant to our research questions, so we did not report them. For calculating the interaction effects, the continuous variables were grand-mean centered. The models were fit by restricted maximum likelihood. We computed the two-level regression models and their marginal R2 values and plotted the interaction effects with R 4.1.0 (R Core Team, 2020) using the R packages lme4 (Bates et al., 2015), lmerTest (Kuznetsova et al., 2017), MuMIn (Barton, 2021), and ggplot2 (Wickham, 2016).
As the data were collected at different time points within- and between-culture, we conducted a supplementary model in which pandemic-related confounders were statistically controlled for (i.e., new cases and deaths per million, the government response stringency index, vaccination rates per hundred; derived from Lin et al., 2021; Richie et al., 2020). Rates reported on the date each participant had been surveyed were counted and statistically controlled for. This supplementary model showed, however, no change in the direction and significance of the main effects, and the included covariates per se did not appear significant (see Supplementary Material).
Results
Table 1 shows the mean score differences in the outcome variables and main predictors between German and Hong Kong residents. Overall, pandemic-induced aging preparation seemed to occur across domains (MFinances = 2.80; MHousing = 2.81; MCare = 2.82; MConnectedness = 3.31; and MEoL = 3.10). However, compared with Germans, those from Hong Kong engaged in more pandemic-induced aging preparation across the five domains (all p < .001). On average, the total sample reported a moderate level of worries about COVID-19 becoming a health threat to them (M = 2.83), while Hong Kong participants tended to be more worried about it than Germans (p < .001). A correlation matrix of the continuous variables is given in Table 2. Reporting higher levels of COVID-19 worries was associated with more engagement in pandemic-induced aging preparation across the five domains (all p < .001).
Variable . | . | 1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | 8 . | 9 . |
---|---|---|---|---|---|---|---|---|---|---|
1 | Finances | ― | 0.55 | 0.54 | 0.41 | 0.42 | 0.23 | −0.01 | −0.04 | −0.20 |
2 | Housing | ― | 0.68 | 0.52 | 0.51 | 0.23 | 0.19 | −0.14 | −0.23 | |
3 | Care | ― | 0.59 | 0.61 | 0.30 | 0.24 | −0.19 | −0.23 | ||
4 | Connectedness | ― | 0.60 | 0.25 | 0.12 | −0.13 | −0.17 | |||
5 | EoL | ― | 0.39 | 0.13 | −0.18 | −0.13 | ||||
6 | COVID-19 worries | ― | 0.01 | −0.26 | −0.04 | |||||
7 | Age | ― | −0.22 | −0.13 | ||||||
8 | Subjective health | ― | 0.10 | |||||||
9 | ISCED | ― |
Variable . | . | 1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | 8 . | 9 . |
---|---|---|---|---|---|---|---|---|---|---|
1 | Finances | ― | 0.55 | 0.54 | 0.41 | 0.42 | 0.23 | −0.01 | −0.04 | −0.20 |
2 | Housing | ― | 0.68 | 0.52 | 0.51 | 0.23 | 0.19 | −0.14 | −0.23 | |
3 | Care | ― | 0.59 | 0.61 | 0.30 | 0.24 | −0.19 | −0.23 | ||
4 | Connectedness | ― | 0.60 | 0.25 | 0.12 | −0.13 | −0.17 | |||
5 | EoL | ― | 0.39 | 0.13 | −0.18 | −0.13 | ||||
6 | COVID-19 worries | ― | 0.01 | −0.26 | −0.04 | |||||
7 | Age | ― | −0.22 | −0.13 | ||||||
8 | Subjective health | ― | 0.10 | |||||||
9 | ISCED | ― |
Notes: Coefficients printed in bold are significant (p < .05). Finances, housing, care, connectedness, and EoL stand for pandemic-induced aging preparation for financial situation in old age, housing arrangements in old age, need for nursing care, social connectedness in old age, and end of life, respectively. ISCED = International Standard Classification of Education. Only continuous variables are included in this correlation matrix.
Variable . | . | 1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | 8 . | 9 . |
---|---|---|---|---|---|---|---|---|---|---|
1 | Finances | ― | 0.55 | 0.54 | 0.41 | 0.42 | 0.23 | −0.01 | −0.04 | −0.20 |
2 | Housing | ― | 0.68 | 0.52 | 0.51 | 0.23 | 0.19 | −0.14 | −0.23 | |
3 | Care | ― | 0.59 | 0.61 | 0.30 | 0.24 | −0.19 | −0.23 | ||
4 | Connectedness | ― | 0.60 | 0.25 | 0.12 | −0.13 | −0.17 | |||
5 | EoL | ― | 0.39 | 0.13 | −0.18 | −0.13 | ||||
6 | COVID-19 worries | ― | 0.01 | −0.26 | −0.04 | |||||
7 | Age | ― | −0.22 | −0.13 | ||||||
8 | Subjective health | ― | 0.10 | |||||||
9 | ISCED | ― |
Variable . | . | 1 . | 2 . | 3 . | 4 . | 5 . | 6 . | 7 . | 8 . | 9 . |
---|---|---|---|---|---|---|---|---|---|---|
1 | Finances | ― | 0.55 | 0.54 | 0.41 | 0.42 | 0.23 | −0.01 | −0.04 | −0.20 |
2 | Housing | ― | 0.68 | 0.52 | 0.51 | 0.23 | 0.19 | −0.14 | −0.23 | |
3 | Care | ― | 0.59 | 0.61 | 0.30 | 0.24 | −0.19 | −0.23 | ||
4 | Connectedness | ― | 0.60 | 0.25 | 0.12 | −0.13 | −0.17 | |||
5 | EoL | ― | 0.39 | 0.13 | −0.18 | −0.13 | ||||
6 | COVID-19 worries | ― | 0.01 | −0.26 | −0.04 | |||||
7 | Age | ― | −0.22 | −0.13 | ||||||
8 | Subjective health | ― | 0.10 | |||||||
9 | ISCED | ― |
Notes: Coefficients printed in bold are significant (p < .05). Finances, housing, care, connectedness, and EoL stand for pandemic-induced aging preparation for financial situation in old age, housing arrangements in old age, need for nursing care, social connectedness in old age, and end of life, respectively. ISCED = International Standard Classification of Education. Only continuous variables are included in this correlation matrix.
The intercept-only model revealed that about 47% of variance was attributed to Level 1, whereas 53% of variance was attributed to Level 2. Hence, 53% of variance is likely due to interindividual differences, whereas the remaining 47% of variance could potentially be explained by domain-specific indicators. Table 3 presents the results of the two-level regression models. Model 1 estimated the effects of the four domains where financial preparation served as the reference domain. Participants reported equivalent levels of pandemic-induced aging preparation for the domains of finances, housing, and care (Housing = 0.00, n.s.; Care = 0.01, n.s.). In contrast, they reported more intensive aging preparation in the domains of connectedness and end-of-life (Connectedness = 0.51, p < .001; EoL = 0.29, p < .001). Model 2 depicts the regression coefficients of the individual-level variables. Averaged across domains, having more COVID-19 worries increased the likelihood of pandemic-induced aging preparation ( = 0.48, p < .001). Living in Hong Kong (rather than Germany) was associated with more pandemic-induced aging preparation when domains were not specified ( = 1.33, p < .001). Next, in Model 3, we specified the effects of COVID-19 worries by culture. The positive association between COVID-19 worries and pandemic-induced aging preparation was particularly apparent for Germans in comparison to those from Hong Kong (Worries:HongKong = −0.16, p < .05; see Figure 1). Yet, once the domains were specified in Model 4, this interaction between culture and COVID-19 worries became nonsignificant for the domains of finances and housing (Worries:HongKong = Worries:HongKong:Finances = 0.02, n.s.; Worries:HongKong:Housing = −0.14, n.s.) but significant and thus stronger for care, connectedness, and end-of-life (Worries:HongKong:Care = −0.21, p < .05; Worries:HongKong:Connectedness = −0.22, p < .05; Worries:HongKong:EoL = −0.31, p < .01; see Figure 2). Adding the three-way interaction term significantly increased the model fit at p < .05 when compared with the model where all possible two-way interaction terms were entered. As to the effects of the covariates, older age, lower education, and no steady relationship were significantly associated with more pandemic-induced aging preparation.
. | Model 1 . | . | Model 2 . | . | Model 3 . | . | Model 4 . | . |
---|---|---|---|---|---|---|---|---|
. | Est. . | SE . | Est. . | SE . | Est. . | SE . | Est. . | SE . |
Intercept | 2.80*** | 0.05 | 2.98*** | 0.18 | 2.96*** | 0.18 | 2.84*** | 0.18 |
Housing | 0.00 | 0.04 | 0.00 | 0.04 | 0.00 | 0.04 | 0.03 | 0.05 |
Care | 0.01 | 0.04 | 0.01 | 0.04 | 0.01 | 0.04 | 0.08 | 0.05 |
Connectedness | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.75*** | 0.05 |
EoL | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.53*** | 0.05 |
Age | 0.01*** | 0.00 | 0.01*** | 0.00 | 0.01*** | 0.00 | ||
Female | 0.10 | 0.07 | 0.10 | 0.07 | 0.10 | 0.07 | ||
Nonbinary | −0.44 | 0.44 | −0.44 | 0.44 | −0.44 | 0.44 | ||
Subjective health | −0.03 | 0.03 | −0.03 | 0.03 | −0.03 | 0.03 | ||
ISCED | −0.05** | 0.02 | −0.05** | 0.02 | −0.05** | 0.02 | ||
Steady relationship | −0.18** | 0.07 | −0.19** | 0.07 | −0.19** | 0.07 | ||
Hong Kong | 1.33*** | 0.08 | 1.36*** | 0.09 | 1.86*** | 0.11 | ||
Worries | 0.48*** | 0.03 | 0.52*** | 0.04 | 0.34*** | 0.05 | ||
Worries × Hong Kong | −0.16* | 0.08 | 0.02 | 0.10 | ||||
Worries × Hong Kong × Housing | −0.14 | 0.11 | ||||||
Worries × Hong Kong × Care | −0.21* | 0.11 | ||||||
Worries × Hong Kong × Connectedness | −0.22* | 0.11 | ||||||
Worries × Hong Kong × EoL | −0.31** | 0.11 | ||||||
Intercept | 1.99 | 1.30 | 1.29 | 1.30 | ||||
Domain | 1.67 | 1.67 | 1.67 | 1.61 | ||||
R2Marginal | 0.01 | 0.20 | 0.20 | 0.21 | ||||
ΔF | ― | 591.74*** | 4.54* | 259.63*** |
. | Model 1 . | . | Model 2 . | . | Model 3 . | . | Model 4 . | . |
---|---|---|---|---|---|---|---|---|
. | Est. . | SE . | Est. . | SE . | Est. . | SE . | Est. . | SE . |
Intercept | 2.80*** | 0.05 | 2.98*** | 0.18 | 2.96*** | 0.18 | 2.84*** | 0.18 |
Housing | 0.00 | 0.04 | 0.00 | 0.04 | 0.00 | 0.04 | 0.03 | 0.05 |
Care | 0.01 | 0.04 | 0.01 | 0.04 | 0.01 | 0.04 | 0.08 | 0.05 |
Connectedness | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.75*** | 0.05 |
EoL | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.53*** | 0.05 |
Age | 0.01*** | 0.00 | 0.01*** | 0.00 | 0.01*** | 0.00 | ||
Female | 0.10 | 0.07 | 0.10 | 0.07 | 0.10 | 0.07 | ||
Nonbinary | −0.44 | 0.44 | −0.44 | 0.44 | −0.44 | 0.44 | ||
Subjective health | −0.03 | 0.03 | −0.03 | 0.03 | −0.03 | 0.03 | ||
ISCED | −0.05** | 0.02 | −0.05** | 0.02 | −0.05** | 0.02 | ||
Steady relationship | −0.18** | 0.07 | −0.19** | 0.07 | −0.19** | 0.07 | ||
Hong Kong | 1.33*** | 0.08 | 1.36*** | 0.09 | 1.86*** | 0.11 | ||
Worries | 0.48*** | 0.03 | 0.52*** | 0.04 | 0.34*** | 0.05 | ||
Worries × Hong Kong | −0.16* | 0.08 | 0.02 | 0.10 | ||||
Worries × Hong Kong × Housing | −0.14 | 0.11 | ||||||
Worries × Hong Kong × Care | −0.21* | 0.11 | ||||||
Worries × Hong Kong × Connectedness | −0.22* | 0.11 | ||||||
Worries × Hong Kong × EoL | −0.31** | 0.11 | ||||||
Intercept | 1.99 | 1.30 | 1.29 | 1.30 | ||||
Domain | 1.67 | 1.67 | 1.67 | 1.61 | ||||
R2Marginal | 0.01 | 0.20 | 0.20 | 0.21 | ||||
ΔF | ― | 591.74*** | 4.54* | 259.63*** |
Notes:
*p < .05,
**p < .01,
***p < .001. Housing, care, connectedness, and EoL stand for pandemic-induced aging preparation for housing arrangements in old age, need for nursing care, social connectedness in old age, and end of life, respectively. Germany and financial preparation served as the reference categories for culture and domain, respectively. ISCED = the International Standard Classification of Education; SE = standard error.
. | Model 1 . | . | Model 2 . | . | Model 3 . | . | Model 4 . | . |
---|---|---|---|---|---|---|---|---|
. | Est. . | SE . | Est. . | SE . | Est. . | SE . | Est. . | SE . |
Intercept | 2.80*** | 0.05 | 2.98*** | 0.18 | 2.96*** | 0.18 | 2.84*** | 0.18 |
Housing | 0.00 | 0.04 | 0.00 | 0.04 | 0.00 | 0.04 | 0.03 | 0.05 |
Care | 0.01 | 0.04 | 0.01 | 0.04 | 0.01 | 0.04 | 0.08 | 0.05 |
Connectedness | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.75*** | 0.05 |
EoL | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.53*** | 0.05 |
Age | 0.01*** | 0.00 | 0.01*** | 0.00 | 0.01*** | 0.00 | ||
Female | 0.10 | 0.07 | 0.10 | 0.07 | 0.10 | 0.07 | ||
Nonbinary | −0.44 | 0.44 | −0.44 | 0.44 | −0.44 | 0.44 | ||
Subjective health | −0.03 | 0.03 | −0.03 | 0.03 | −0.03 | 0.03 | ||
ISCED | −0.05** | 0.02 | −0.05** | 0.02 | −0.05** | 0.02 | ||
Steady relationship | −0.18** | 0.07 | −0.19** | 0.07 | −0.19** | 0.07 | ||
Hong Kong | 1.33*** | 0.08 | 1.36*** | 0.09 | 1.86*** | 0.11 | ||
Worries | 0.48*** | 0.03 | 0.52*** | 0.04 | 0.34*** | 0.05 | ||
Worries × Hong Kong | −0.16* | 0.08 | 0.02 | 0.10 | ||||
Worries × Hong Kong × Housing | −0.14 | 0.11 | ||||||
Worries × Hong Kong × Care | −0.21* | 0.11 | ||||||
Worries × Hong Kong × Connectedness | −0.22* | 0.11 | ||||||
Worries × Hong Kong × EoL | −0.31** | 0.11 | ||||||
Intercept | 1.99 | 1.30 | 1.29 | 1.30 | ||||
Domain | 1.67 | 1.67 | 1.67 | 1.61 | ||||
R2Marginal | 0.01 | 0.20 | 0.20 | 0.21 | ||||
ΔF | ― | 591.74*** | 4.54* | 259.63*** |
. | Model 1 . | . | Model 2 . | . | Model 3 . | . | Model 4 . | . |
---|---|---|---|---|---|---|---|---|
. | Est. . | SE . | Est. . | SE . | Est. . | SE . | Est. . | SE . |
Intercept | 2.80*** | 0.05 | 2.98*** | 0.18 | 2.96*** | 0.18 | 2.84*** | 0.18 |
Housing | 0.00 | 0.04 | 0.00 | 0.04 | 0.00 | 0.04 | 0.03 | 0.05 |
Care | 0.01 | 0.04 | 0.01 | 0.04 | 0.01 | 0.04 | 0.08 | 0.05 |
Connectedness | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.51*** | 0.04 | 0.75*** | 0.05 |
EoL | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.29*** | 0.04 | 0.53*** | 0.05 |
Age | 0.01*** | 0.00 | 0.01*** | 0.00 | 0.01*** | 0.00 | ||
Female | 0.10 | 0.07 | 0.10 | 0.07 | 0.10 | 0.07 | ||
Nonbinary | −0.44 | 0.44 | −0.44 | 0.44 | −0.44 | 0.44 | ||
Subjective health | −0.03 | 0.03 | −0.03 | 0.03 | −0.03 | 0.03 | ||
ISCED | −0.05** | 0.02 | −0.05** | 0.02 | −0.05** | 0.02 | ||
Steady relationship | −0.18** | 0.07 | −0.19** | 0.07 | −0.19** | 0.07 | ||
Hong Kong | 1.33*** | 0.08 | 1.36*** | 0.09 | 1.86*** | 0.11 | ||
Worries | 0.48*** | 0.03 | 0.52*** | 0.04 | 0.34*** | 0.05 | ||
Worries × Hong Kong | −0.16* | 0.08 | 0.02 | 0.10 | ||||
Worries × Hong Kong × Housing | −0.14 | 0.11 | ||||||
Worries × Hong Kong × Care | −0.21* | 0.11 | ||||||
Worries × Hong Kong × Connectedness | −0.22* | 0.11 | ||||||
Worries × Hong Kong × EoL | −0.31** | 0.11 | ||||||
Intercept | 1.99 | 1.30 | 1.29 | 1.30 | ||||
Domain | 1.67 | 1.67 | 1.67 | 1.61 | ||||
R2Marginal | 0.01 | 0.20 | 0.20 | 0.21 | ||||
ΔF | ― | 591.74*** | 4.54* | 259.63*** |
Notes:
*p < .05,
**p < .01,
***p < .001. Housing, care, connectedness, and EoL stand for pandemic-induced aging preparation for housing arrangements in old age, need for nursing care, social connectedness in old age, and end of life, respectively. Germany and financial preparation served as the reference categories for culture and domain, respectively. ISCED = the International Standard Classification of Education; SE = standard error.
Discussion
In our research, we investigated whether worrying about COVID-19 predicted people’s engagement in aging preparation and whether these associations would have culture- (i.e., Hong Kong and Germany) and domain-specific (i.e., finances, housing, care needs, connectedness, and end-of-life) tendencies. Findings suggested that being worried about COVID-19 was associated with a higher likelihood of engagement in aging preparation across domains. Although people who lived in Hong Kong generally reported a higher level of worries and a higher level of pandemic-induced aging preparation, the positive association between worries and aging preparation was stronger among Germans than Hong Kong Chinese. This cultural difference was, however, only significant for the domains of care, connectedness, and end-of-life. In the following, we will discuss the role of COVID-19 worries in predicting aging preparation and its cultural- and domain-specific tendencies.
Consistent with our hypothesis, worrying about COVID-19 becoming a health threat to the self was associated with more engagement in overall aging preparation. This finding is in line with prior literature on worries as a determinant of behaviors (e.g., Barber & Kim, 2021). It implies that having COVID-19 worries may alter one’s perception toward vulnerability even after controlling for the effects of current physical conditions and calendar age. Unlike other coping efforts made to minimize extant threats or stressors, proactive coping takes place before stressors occur, thus requiring future-oriented thinking (Aspinwall, 2005). We speculate that being worried about the virus functions as a “recently primed cue” and evokes the contemplation of the future time when one becomes especially worried about the situation. Given that the perception of life’s finitude is heightened for those who perceived COVID-19 as a health threat (Rupprecht et al., 2021), people seem to cope with this emotional burden by engaging themselves in aging preparation.
We found that the association between COVID-19 worries and pandemic-induced aging preparation appeared stronger in Germany than Hong Kong. Because the COVID-19 mortality rate is about 40 times higher in Germany compared with that in Hong Kong (Ritchie et al., 2020), situational cues related to COVID-19 are perhaps more widespread in Germany, such as media coverage on the spread of the virus (Bendau et al., 2021). The existence of such cues might urge people to deal with expected life challenges in the future more intensively than before. A rather unexpected finding to address is the greater COVID-19 worries reported by participants from Hong Kong than those from Germany. One explanation for this difference might be the higher level of the neuroticism of East Asians than Western Europeans reported in a few studies (e.g., Schmitt et al., 2007). While individual differences likely exist, Hong Kong Chinese as a group may have higher neuroticism than Germans and thus report more worries about COVID-19.
As hypothesized, the different courses and outcomes of the pandemic in each culture might influence individuals to selectively focus on certain domains of aging preparation. Although one should be aware of within-culture heterogeneity, our findings suggested that in general the pandemic might have induced the German participants to more intensively prepare for future care needs, social connectedness in old age, and their own end of life than for financial situations and housing arrangements in old age, relative to the Hong Kong participants. We suppose that such findings stem from the entangled roles of each culture’s past experience with the epidemic (hence, government-mandated restrictions) and areas mostly hit by the COVID-19 pandemic. Perhaps, Germans are more likely to come in touch with deaths by COVID-19 in their extended networks than individuals from Hong Kong. Knowing someone who died or was close to death because of the virus should induce certain coping strategies. In addition, considering that the level of social engagement (e.g., going to a church, joining a sports club) has been reported to be high in Germany (Hajek et al., 2017), the unprecedented restrictions on social gatherings may trigger preparation for anticipated future losses in social contacts, more so for Germans than Hong Kong Chinese. The results could also be attributed to the fact that the domains of finance and housing might be more urgently needed by Hong Kong Chinese, leaving them fewer resources to attend to the other three domains at the same level as German participants did.
We should note that the pandemic-induced aging preparation does not necessarily refer to people’s engagement in concrete preparatory activities. Given that the engagement in concrete activities hardly takes place within a short period of time, we had chosen to examine general engagement in aging preparation while specifying the target domains (Kornadt et al., 2020). In fact, a supplementary analysis revealed that this construct correlated significantly with information-seeking but not with the implementation of concrete preparatory activities (see Author Note 1; Kim-Knauss & Lang, 2021). These results endorse using a multifaceted construct of aging preparation and imply that aging preparation could involve a certain phase or course of action, such as deliberation, planning, and action implementation (Achtziger & Gollwitzer, 2018). Besides, it has been reported that thinking about the topic without making concrete plans was not as functional as making concrete plans (Pinquart & Sörensen, 2002). Hence, future studies drawing on longitudinal data should attend to whether pandemic-induced aging preparation indeed develops into engagement in concrete preparatory activities and its consequences on the self.
It is undeniable that the COVID-19 pandemic is threatening humanity and has been changing the way of living. Yet, current findings showed that it could also lead people to take action in hopes of better states in the future when faced with similar life challenges. Furthermore, we found that the reported level of pandemic-induced aging preparation was higher in Hong Kong than in Germany. This result is especially noteworthy given the reported lower levels of aging preparation in nonpandemic times in Hong Kong than in Germany (Kornadt et al., 2018). Although this difference might be attributed to a sort of ceiling effect in Germany or culture-oriented response styles (Johnson et al., 2005), the result may indicate that people living in Hong Kong have become more aware of the need for aging preparation during this global crisis. The same also applies to the higher level of pandemic-induced aging preparation found among people with a lower education level, while people with lower SES have been consistently reported as a less prepared group (Preston et al., 2019). Efforts should be made to ensure that those with limited financial, health, and time resources are able to engage in concrete aging preparation when they wish to do so.
Limitations
There are a few caveats that ought to be considered when interpreting the findings of this research. First, the measure we used in the present study, as well as the cross-sectional data set, lack information on actual changes in engagement in aging preparation in the pre and postpandemic periods. Data sets spanning a longer time and including more explicit and concrete measures of aging preparation will allow the discussion on how pandemic-induced aging preparation develops over time.
Second, in relation to the different times of data collection between- and within-culture (i.e., Germany: November−December 2020, Hong Kong: February−April 2021), the time-in-pandemic may have been confounded with cultural and individual differences. For instance, the survey time in Hong Kong coincides with a milestone of the pandemic when vaccines become globally more available for the general public. However, considering that the vaccination campaign began relatively later in Hong Kong than in Germany (see Author Note 2), we supposed its affects to be rather negligible. The supplementary model also showed nonsignificant affects of pandemic-related confounding factors. Nevertheless, time-specific experiences of the pandemic may have differently shaped its threat, worries, and preparations related to it. One should also be wary of other nonpandemic but potentially influential events that occurred in the period proximal to the studies (e.g., 2019 protests in Hong Kong), which might have influenced the cultural differences observed in this study as well.
Third, the within-culture heterogeneity should be considered. While we drew our hypotheses based on culture-level experiences of the pandemic, direct experiences of COVID-19 regarding the self or close others, psychological appraisals of the given situation, and thus their effects on one’s thoughts and actions may still vary by individual (Beck et al., 2021; Bendau et al., 2021). Furthermore, while the general directions of the measures were the same, the stringency of measures (e.g., lockdown) differed between states in Germany, implying that the pandemic’s threat would vary across regions (Armbruster & Klotzbücher, 2020). Also, the ethnic group information provided by the Hong Kong group suggests that the sample was rather racially homogeneous, and likewise, the German participants were rarely from immigrant families. Therefore what the current study demonstrated might not be applicable to those who share different characteristics, such as place of residence, access to resources, or ethnicity (Armbruster & Klotzbücher, 2020; Sörensen et al. 2014). Future studies should attempt to replicate our findings after taking more diverse contextual factors or socioeconomic status into account.
Fourth, the smaller sample size in Hong Kong as compared to Germany may have also yielded less reliable estimates for coefficients and thus inflated some of the observed cultural differences; though we may add that sample compositions did not differ much concerning age and unequal sample sizes should not have caused biased statistical inferences in linear mixed models (Pinheiro, 2014). In a supplementary model where the sample type―whether the participant belongs to the follow-up sample or to the newly recruited sample―was controlled for, we found no change in the model (see Supplementary Material).
Conclusion
To conclude, our findings suggest that having self-relevant worries about COVID-19 encourages people to engage in overall aging preparation. Yet, COVID-19 worries could have a particularly profound impact on those from cultures that are more severely affected by the virus. According to our results, Germans who worry more about the pandemic may feel a greater urge to invest in their future than those from Hong Kong. In addition, the different courses and outcomes of the pandemic in each culture might lead people to focus selectively on certain domains of aging preparation. The stronger association between pandemic-induced worries and aging preparation in Germany than in Hong Kong might therefore be more evident in the domains of care, connectedness, and end-of-life than finances and housing in old age. We submit that based on our research, public health efforts may serve to promote and support proactive aging preparation at a time when people might be particularly motivated to engage in this process and to target diverse dimensions of action, such as information-seeking.
Acknowledgments
Data and analytical methods for this study will be available from the first author upon request. Data from this research project will be published in the PsychArchives of the Leibniz Institute for Psychological Information and Documentation (ZPID) in 2022. This study is not previously registered.
Author Notes
1. Prior studies have discussed diverse facets of aging preparation. For instance, Carr and Khodyakov (2007) distinguished formal (e.g., having a living will) from informal (e.g., discussing with relatives) end-of-life planning, and Kim-Knauss and Lang (2021) distinguished implementation (e.g., having created an advance directive) from information-seeking (e.g., talking with relatives about care preferences). The correlation between the measure we used in the present study with the information-seeking scale was significant ( = 0.11, p < .05) but not with the implementation scale ( = 0.02, n.s.).
2. In Hong Kong, the reported vaccination rates (at least one dose) for the earliest, first quartile, median, third quartile, and latest survey dates were 0.0%, 0.0%, 0.5%, 2.2%, and 13.5%, respectively, while 0.3% was reported on the latest survey date in Germany and 0.0% on the others (Ritchie et al., 2020).
Funding
This work was supported by the Volkswagen Foundation (grant number Az 93 273 to F. R. Lang, Az 93 273-2 to F. R. Lang, F. S. Rupprecht, and H. H. Fung) and the Hong Kong Research Grants Council General Research Fund (grant number #14604220 to H. H. Fung and F. R. Lang).
Conflict of Interest
None declared.