ORIGINAL RESEARCH

published: 08 March 2021

doi: 10.3389/fpsyt.2021.635832

Stressful Life Events and Resilience

During the COVID-19 Lockdown
Measures in Italy: Association With
Mental Health Outcomes and Age
Rodolfo Rossi 1*, Tommaso B. Jannini 1 , Valentina Socci 2 , Francesca Pacitti 2 and
Giorgio Di Lorenzo 1,3
1
Department of Systems Medicine, University of Rome Tor Vergata, Rome, Italy, 2 Department of Biotechnological and
Applied Clinical Sciences, University of L’Aquila, L’Aquila, Italy, 3 Istituto di Ricovero e Cura a Carattere Scientifico (IRCCS)
Fondazione Santa Lucia, Rome, Italy

Background: The COVID-19 pandemic, due to its disproportionated higher morbidity

and mortality rates in the older age, has been considered to be a “geropandemic.” Several
studies, however, have found that older age is associated with lower psychological
distress in relation to the COVID-19 outbreak and related lockdown measures.
Edited by:
Gianfranco Spalletta,
Aim: To explore the role of Resilience as a mediator between stressful COVID-19 related
Santa Lucia Foundation (IRCCS), Italy life events and depressive and, anxiety symptoms and perceived stress, and to ascertain
Reviewed by: the role of age as a moderator of the mediator’s effect.
Debanjan Banerjee,
National Institute of Mental Health and Methods: An on-line survey was spread through social networks during the first
Neurosciences (NIMHANS), India lockdown in Italy. Depressive and anxiety symptoms and perceived stress were measured
Sanjeev Kumar,
University of Toronto, Canada
using the Italian version of the Patient Health Questionnaire-9 (PHQ-9), the Generalized
*Correspondence:
Anxiety Disorder Questionnaire-7 (GAD-7) and the Perceived Stress Scale (PSS).
Rodolfo Rossi Resilience was measured using the Italian version of the Resilience Scale for Adults (RSA).
rudy86.rossi@gmail.com
Stressful COVID-19 related life events were explored using a checklist of events derived
from the International Adjustment Disorder Questionnaire (IADQ). After a preliminary
Specialty section:
This article was submitted to panel of linear regressions, mediation was tested using Structural Equation Modeling
Aging Psychiatry, and inspecting the bootstrapped indirect effects. Afterwards, age was introduced as a
a section of the journal
Frontiers in Psychiatry mediator of the indirect effect in a moderated mediation analysis.
Received: 30 November 2020 Results: Twenty one thousand three hundred and thirty four subjects completed the
Accepted: 10 February 2021
questionnaire, 17,178 (80.52%) were female, 748 (3.5%) were >60 years old. In the
Published: 08 March 2021
whole sample, the presence of any stressful event was associated with depressive and
Citation:
Rossi R, Jannini TB, Socci V, Pacitti F anxiety symptoms and perceived stress. Resilience mediated the effects of stressful
and Lorenzo GD (2021) Stressful Life COVID-19-related events on depressive and anxiety symptoms and perceived stress.
Events and Resilience During the
COVID-19 Lockdown Measures in
The moderated mediation analysis revealed that age moderated the mediation effect of
Italy: Association With Mental Health Resilience between the presence of a stressful event and the selected outcomes.
Outcomes and Age.
Front. Psychiatry 12:635832. Conclusion: Taken together, our results show that age moderates the mediating
doi: 10.3389/fpsyt.2021.635832 effect of Resilience in the relationship between COVID-19-related stressful events and

Frontiers in Psychiatry | www.frontiersin.org 1 March 2021 | Volume 12 | Article 635832

Rossi et al. COVID-19, Stress and Resilience

depressive and anxiety symptoms and perceived stress. Older adults’ Resilience was
less influenced by stressful events, and this could be one of the reasons accounting for
the better mental health outcomes observed in the older age.

Keywords: old age, resilience, COVID-19, mental health, stress

INTRODUCTION population (2). In his study, the mitigating effect of high

resilience with lower anxiety was stronger in older age. Such
The COVID-19 pandemic, due to its disproportionated higher finding was associated with lower COVID-19-related worries in
morbidity and mortality rates in the older age, has been the older age. In another study on a Turkish sample, older age
considered to be a “geropandemic” (1). Several studies, however, was associated with both higher resilience and lower depression
have found that older age is associated with lower psychological rates (11).
distress in relation to the COVID-19 outbreak and lockdown The exact pathways by which resilience interacts with age
measures. Younger individuals, especially women, report higher in affecting mental health remains unclear. The present study
levels of depressive, anxious and stress-related symptoms aims to address the role of resilience in older adults with respect
compared to older age (2–4). The evidence that older age is to mental health outcomes during the COVID-19 pandemic.
associated with better psychological outcomes is in apparent Our hypothesis is that the role of Resilience in mitigating the
contrast with the increased physical vulnerability of the elderly impact of COVID-related stressful life events may be different at
to the COVID-19. Despite morbidity and mortality is highly different ages. To test this hypothesis, it was firstly tested whether
correlated with age (5), a recent study found that older adults resilience would mediate the relationship between COVID-19-
show slightly less COVID-19-related worries compared to related stressful events and depressive and anxiety symptoms
younger participants (2). On the other hand, several studies have and perceived stress (mediation). Secondly, we tested whether
highlighted issues such as isolation and loneliness. the indirect effect of stressful events on depressive and anxiety
Resilience could be one of the putative psychological factors symptoms and perceived stress via resilience would be moderated
that could account for a better adaptation to the COVID-19 by age. Such model is referred to as Moderated Mediation, i.e., a
pandemic in the elderly. In the geriatric literature, resilience model in which a mediator has a different effect at different levels
is associated with successful aging (6), lower mortality, lower of a moderating variable (12).
depressive symptoms, increased quality of life and better
lifestyle behaviors.
Resilience is considered a protective mechanism operating in
the face of negative stressors (7), and it is constantly associated METHODS
with better psychological well-being and lower mental illness.
It has been suggested that older adults may express higher Study Design
levels of resilience compared to the younger ones (8), in This study is a cross-sectional web-based observational study,
particular regarding emotion regulation and problem-solving and it is a part of a long-term monitoring program of mental
dimensions. In contrast, younger individuals show slightly higher health outcomes in the general population and health care
social support, in the context of reduced overall resilience workers. On-line consent was obtained from the participants.
levels. According to a recent systematic review, resilience At 3 weeks after the beginning of the lockdown, an anonymous
in older adults could be operationalized as a 4-dimension survey was conducted among a self-selected sample from the
construct, that includes intrapersonal, interpersonal, spiritual Italian population. Every person living in Italy ≥ 18 years old
and experiential protective factors (9). It is noteworthy that these was eligible. Approval for this study was obtained from IRB at
factors are differentially associated with age or environmental the University of L’Aquila. This study adheres to the Declaration
circumstances, being interpersonal factors the most volatile over of Helsinki.
time and spiritual and experiential factors being associated with
older age.
Age differences in resilience levels could be associated with
the difference in mental health outcomes across the general Sampling Strategy and On-Line
population. It has been reported that during the lockdown, Questionnaire Dissemination
resilience levels were lower than normative data in younger adults For the purpose of this study, the questionnaire was spread using
aged 18–35 years (10), and this was suggested as one key factor sponsored adverts on Facebook R
, as well as using a snowball
affecting the general population’s mental health and pessimism spreading technique starting from the researchers’ acquaintances.
about the future of the pandemic. Because of the particular dissemination technique, it was not
Few studies have addressed the relation between resilience, possible to have precise data on response rate, however using the
mental health and age at the time of the COVID-19 pandemic. In Facebook Ads app, it was possible to estimate that the number
a recent study on a US sample, higher resilience was associated of link clicks was about 100.000, while the ad reached nearly 1
with lower depressive and anxiety symptoms in the general million people.

Frontiers in Psychiatry | www.frontiersin.org 2 March 2021 | Volume 12 | Article 635832

Rossi et al. COVID-19, Stress and Resilience

FIGURE 1 | Proposed moderated mediation model. PHQ, Patient Health Questionnaire; GAD, Generalized Anxiety Questionnaire; PSS, Perceived Stress
Questionnaire; RSA, Resilience Scale for Adults; IADQ, International Adjustment Disorder Questionnaire – stressful events checklist.

TABLE 1 | Sample characteristics.

Total sample <60 yr ≥60 yr Statistics (χ 2 or

Mann-Whitney as
appropriate)

Mean (SD)/N (%) Mean (SD)/N (%) Mean (SD)/N (%)

Age 38.95 (12.77) 38.07 (12.09) 63.12 (4.62)

Gender χ2(1) = 05.66; p = 0.017
Male 4,155 (19.48%) 3,984 (19.35%) 171 (22.86%)
Female 17,178 (80.52%) 16,601 (80.65%) 577 (77.14%)
History of psychiatric disorders 6,067 (28.51%) 5,860 (28.54%) 207 (27.71%) χ2(1) = 0241; p = 0.623
◦
N of Stressful events 0.81 (1.14) 0.83 (1.14) 0.35 (0.77) z = 12.32; p < 0.001
PHQ 10.67 (6.39) 10.78 (6.38) 7.79 7.79) z = 12.84; p < 0.001
GAD 9.03 (5.95) 9.13 (5.95) 6.26 (5.36) z = 13.24; p < 0.001
PSS 24.60 (8.40) 24.76 (8.37) 20.38 (8.06) z = 13.85; p < 0.001
RSA 36.96 (11.84) 37.04 (11.81) 34.68 (12.32) z = 5.40; p < 0.001

PHQ, Patient Health Questionnaire; GAD, Generalized Anxiety Questionnaire; PSS, Perceived Stress Questionnaire; RSA, Resilience Scale for Adults.

Outcome Measures instrument in epidemiological research as anxiety screener. In

The Italian version of the 9-item Patient Health Questionnaire our sample, internal consistency was a = 0.91.
(PHQ-9) was used to assess depressive symptoms. PHQ- The Italian version of the 10-item Perceived Stress Scale
9 comprises nine depressive symptoms, rated on a 4-point (PSS) was used to assess perceived stress. PSS includes ten items
Likert scale, range 0–27. The total score has been taken into rated on a 0–4 Likert scale. In our sample, internal consistency
consideration as a continuous variable. PHQ-9 is a widely used was a = 0.87.
instrument in epidemiological research as a depression screener.
In our sample, internal consistency was a = 0.87.
The Italian version of the 7-item Generalized anxiety Independent Variables, Covariates, and
disorder questionnaire (GAD-7) was used to assess anxiety Confounders
symptoms. GAD-7 includes seven symptoms, rated on a 4-point Age was used both as a continuous and binary variable, with
Likert scale, range 0–21 (13). The total score has been taken into a cut-off of 60 years old as a separation between older and
consideration as a continuous variable. GAD-7 is a widely used younger adults.

Frontiers in Psychiatry | www.frontiersin.org 3 March 2021 | Volume 12 | Article 635832

Rossi et al. COVID-19, Stress and Resilience

TABLE 2 | Association between age and depression, anxiety and stress. TABLE 3 | Association between risk and protective factors and depression,
anxiety and stress.
Unadjusted Adjusted§
Unadjusted Adjusted§
b (95% CI) p b (95% CI) p
b (95% CI) p b (95% CI) p
Age → PHQ −0.11 (−0.12, −0.10) <0.001 −0.12 (−0.12, −0.11) <0.001
Total Any IADQ → PHQ Total 0.44 (0.41, 0.46) <0.001 0.42 (0.39, 0.44) <0.001
Age → GAD −0.098 (−0.10, −0.09) <0.001 −0.10 (−0.11, −0.10) <0.001 Any IADQ → GAD Total 0.42 (0.39, 0.44) <0.001 0.40 (0.37, 0.42) <0.001
Total Any IADQ → PSS Total 0.46 (0.43, 0.48) <0.001 0.43 (0.41, 0.46) <0.001
Age → PSS −0.16 (−0.17, −0.16) <0.001 −0.17 (−0.18, −0.16) <0.001 RSA → PHQ Total 0.58 (0.57, 0.59) <0.001 0.57 (0.56, 0.58) <0.001
Total
RSA → GAD Total 0.52 (0.50, 0.53) <0.001 0.51 (0.50, 0.52) <0.001
§ adjusted for gender, region and education level. PHQ, Patient Health Questionnaire; GAD, RSA → PSS Total 0.54 (0.53, 0.55) <0.001 0.54 (0.52, 0.55) <0.001
Generalized Anxiety Questionnaire; PSS, Perceived Stress Questionnaire.
PHQ, Patient Health Questionnaire; GAD, Generalized Anxiety Questionnaire; PSS,
Perceived Stress Questionnaire; RSA, Resilience Scale for Adults; IADQ, International
Adjustment Disorder Questionnaire – stressful events checklist. RSA, PHQ, GAD, and
PSS are standardized values. § adjusted for gender, region and education level.
Stressful events were assessed using the International
Adjustment Disorder Questionnaire (IADQ) checklist of stressful
events (14). The IADQ checklist explores eight different stressful TABLE 4 | Association between risk and protective factors and age.
events, namely economic, job and study difficulties, problems
related to housing, relational problems, own’s and a loved one’s Unadjusted Adjusted§
health problems, caregiving problems. In the original version, b (95% CI) p b (95% CI) p
each item has a yes/no response. We modified the response as
follows: “no/yes/yes, due to COVID-19 pandemic or lockdown” *Age → Any IADQ −0.44 (−0.47, −0.41)<0.001−0.46 (−0.49, −0.43)<0.001
in order to capture COVID-19 related stressful events. For this Age → RSA −0.14 (−0.16, −0.13)<0.001−0.15 (−0.16, −0.13)<0.001
study, a binary variable was created with 0 = “no stressful events *Old Age → Any IADQ −1.00 (−1.17, −0.83)<0.001−1.00 (−1.17, −0.82)<0.001
due to COVID-19” and 1 = “one or more stressful events due Old Age → RSA −0.20 (−0.27, −0.12)<0.001−0.21 (−0.29, −0.14)<0.001
to COVID-19.”
*logit function, dependent variable is binary. § adjusted for gender, region and education
Resilience was assessed using the 11-items Resilience Scale level. RSA, Resilience Scale for Adults; IADQ, International Adjustment Disorder
for Adults (RSA). The RSA-11 was obtained from the original Questionnaire – stressful events checklist.
33 item version (15). Participants answer on a 7-point semantic
differential scale in which each item has a positive and a negative indirect effects was ascertained inspecting the normal-based and
attribute at each end of the scale continuum. For this study, the bias-corrected confidence intervals.
RSA-11 total score was taken into account, with higher scores Finally, conditional indirect effects of COVID-19 stressful
indicating lower levels of Resilience. events on PHQ-9, GAD-7 and PSS via RSA, entering age as a
The following potential confounders were selected: Gender; moderator were tested. This model is referred to as “Moderated
Geographical Area (Northern Italy: Aosta Valley, Piedmont, Mediation” and it is largely founded on Model 59 by Hayes (12).
Lombardy, Liguria, Trentino-Alto Adige, Veneto, Friuli-Venezia In Figure 1 we show the proposed model for the moderated
Giulia, Emilia-Romagna; Center Italy: Tuscany, Umbria, Marche, mediation. The significance of the bootstrapped conditional
Lazio; Southern Italy: Abruzzo, Molise, Puglia, Campania, indirect effects was ascertained inspecting the normal-based and
Calabria, Basilicata, Sicily and Sardinia); Education level (lower bias-corrected confidence intervals.
education, undergraduate, graduate, post-graduate degree).

Statistical Analysis RESULTS

All statistical analyses were conducted using STATA
R
Sample Characteristics
16 (StataCorp). Sample characteristics are reported in Table 1. Twenty-one
Firstly, the following associations were tested using a thousand three hundred thirty-four subjects participated in the
panel of linear or logistic regressions, as appropriate for the study. Of these, 17,178 (80.52%) were female. Mean age was 38.95
dependent variable: (SD = 12.77); 748 (3.51%) subjects were 60 or older.
1) association between age and PHQ-9, GAD-7 and PSS
2) association between RSA and stressful events and PHQ-9,
Associations Between Selected Variables
Tables 2–4 report the linear associations between age, resilience,
GAD-7 and PSS
COVID-19 related stressful events and psychopathology. In
3) association between RSA and stressful events and age.
our sample, age was inversely associated with PHQ-9, GAD-
Secondly, a mediation model was fitted on PHQ-9, GAD-7 and 7 and PSS (Table 2). This association held after adjusting
PSS, with stressful events as independent variable and RSA as for the selected confounders. The presence of any stressful
mediator. Mediation was tested by bootstrapping the indirect event was associated with PHQ-9, GAD-7 and PSS (Table 3).
effect at 5,000 replications. The significance of the bootstrapped Better resilience resources (i.e., a lower score on the RSA-11)

Frontiers in Psychiatry | www.frontiersin.org 4 March 2021 | Volume 12 | Article 635832

Rossi et al. COVID-19, Stress and Resilience

FIGURE 2 | Mediation analysis path diagram with direct effects and, in parenthesis, total effects. PHQ, Patient Health Questionnaire; GAD, Generalized Anxiety
Questionnaire; PSS, Perceived Stress Questionnaire; RSA, Resilience Scale for Adults; IADQ, International Adjustment Disorder Questionnaire – stressful
events checklist.

was inversely associated with PHQ-9, GAD-7 and PSS. These level. In particular, having an older age (i.e., over 60-year-
associations held after adjusting for the selected confounders. old) is two-fold more negatively associated with these variables.
Finally, age was associated with lower odds of endorsing These findings highlight that older adults report lower levels of
any COVID-19-related stressful event and with better resilience depressive symptoms, anxiety and stress compared to a younger
(Table 4). These associations held after adjusting for the population. People aged 60 and over usually have a higher
selected confounders. mortality rate and are at higher risk of developing significant
complications, causing them to follow more stringent measures
Mediation Analysis than the others. For these reasons an inversed trend would have
Mediation analysis (Figure 2 and Table 5) showed that the been expected. Although these results might be considered as
impact of COVID-related stressful events on PHQ-9, GAD-7 counterintuitive, a number of authors have emphasized how
and PSS was very similar. RSA partially mediated the impact younger people tend to report higher levels of depressive and
of stressful events on the selected outcomes, as confirmed anxiety symptoms than older ages during pandemics (16–19).
by inspection of the Bootstrapped confidence intervals of the This may be explained by multiple reasons, as people aged
indirect effect through RSA. below 60 are less likely to be retired, therefore being more
preoccupied about their occupational programs and economic
Moderated Mediation Analysis incomes since they might lose their job (20, 21). Furthermore,
Table 6 reports the bootstrapped indirect effects with normal- it is a matter of fact that younger people are keener on spending
based and bias-corrected confidence intervals of the moderated a consistent amount of time on social networks or other news
mediation model. The confidence intervals of the interaction apps (22). As a result, an information overload, also defined as
term Age × RSA, which represents the conditional indirect effect, “infodemic,” where fake news, racist opinions, magic potions and
show that Age moderated the mediation effect of resilience on conspiracy theories are easily disclosed, may account for their
PHQ-9, GAD-7 and PSS. higher scores (23). Lastly, older adults are more likely to have
faced a number of major life events than their counterparts,
DISCUSSION having, therefore, a bigger wealth of experience that would
allow them to face adversities more easily. In other words, older
We presented a cross-sectional study aimed to evaluate how adults may better rely on their resilience when dealing with
resilience differs in an age-dependent manner, representing a key such situations. However, it is important to notice that mental
feature in older adults with respect to mental health outcomes health outcomes in this study were addressed using screeners
during the COVID-19 pandemic. that are more focused on the affective and cognitive components
Results showed that age is negatively associated with PHQ- of anxiety and depression, rather than somatic complaints or
9, GAD-7 and PSS scores. This was significant even controlling loneliness, which are common features of psychological distress
for confounding factors such as gender, region and education in the elderly.

Frontiers in Psychiatry | www.frontiersin.org 5 March 2021 | Volume 12 | Article 635832

Rossi et al. COVID-19, Stress and Resilience

TABLE 5 | Path analysis results. In line with our hypothesis and consistent with previous
literature, RSA scores are linearly associated with PHQ-9, GAD-7
Coef Confidence intervals
and PSS. This means that higher levels of resilience act through a
Direct effects “buffering effect” on such variables, therefore mitigating COVID-
RSA → PHQ 0.56 (0.55, 0.57) (0.55, 0.57) 19 related stressors (2).
IADQ → PHQ 0.30 (0.27, 0.32) (0.27, 0.32) Supporting the hypothesis that older adults have better
RSA → GAD 0.50 (0.49, 0.51) (0.49, 0.51)
abilities to manage calamities and to get by during difficult times,
IADQ → GAD 0.29 (0.27, 0.31) (0.27, 0.31)
regression analyses showed that age is inversely associated with
RSA → PSS 0.52 (0.51, 0.53) (0.51, 0.53)
any of IADQ items and RSA scores. Results are even higher when
IADQ → PSS 0.33 (0.30, 0.35) (0.30, 0.35)
old age is set as the independent variable, meaning that people
aged 60 and over perform remarkably better at successfully
IADQ → RSA 0.25 (0.22, 0.28) (0.22, 0.28)
overcome stressful life events such as COVID-19 pandemic.
Total effects
These findings are in line with previous research, as a number
RSA → PHQ 0.56 (0.55, 0.57) (0.55, 0.57)
of authors have recognized how older people generally have high
IADQ → PHQ 0.42 (0.41, 0.46) (0.41, 0.46)
levels of resilience during difficult times, despite their own state of
RSA → GAD 0.50 (0.49, 0.51) (0.49, 0.51)
health, socioeconomic status and past personal experiences (26).
IADQ → GAD 0.42 (0.39, 0.44) (0.39, 0.44)
Our simple mediation analysis highlighted the role of
RSA → PSS 0.52 (0.51, 0.53) (0.51, 0.53)
COVID-19-related stressful events on resilience, reporting a
IADQ → PSS 0.46 (0.43, 0.48) (0.43, 0.48)
positive association with RSA scores. These findings show that
IADQ → RSA 0.25 (0.22, 0.28) (0.22, 0.28)
bigger stressful events are capable of affecting one’s ability to
Bootstrapped indirect effects
cope with adversities. Moreover, other than having a direct
IADQ → (RSA) → PHQ 0.14 0.12, 0.15 (N)
effect on PHQ, GAD-7 and PSS scores, IADQ acts indirectly on
0.12, 0.16 (BC)
these three variables. Indeed, when setting RSA as a mediating
IADQ → (RSA) → GAD 0.12 0.11, 0.14 (N) variable, regression coefficients still show a positive association
0.11, 0.14 (BC) with PHQ, GAD-7 and PSS scores. In other words, COVID-
IADQ → (RSA) → PSS 0.13 0.11, 0.14 (N) 19-related traumatic events (e.g., losing loved ones, lockdown
0.11, 0.14 (BC) stringent measures, poor economic incomes) may impact directly
(N), Normal-based (95% Conf. Interval); (BC), Bias-corrected confidence interval; PHQ, on enhancing depressive symptoms, anxiety and perceived stress,
Patient Health Questionnaire; GAD, Generalized Anxiety Questionnaire; PSS, Perceived or indirectly, lowering one’s buffering effect of resilience.
Stress Questionnaire; RSA, Resilience Scale for Adults; IADQ, International Adjustment The present mediating effect of RSA is different depending on
Disorder Questionnaire – stressful events checklist.
the age of the subjects. In older people, resilience influences the
psychopathological outcome more strongly compared to younger
TABLE 6 | Conditional Indirect effects of the moderated mediation model. adults, meaning that their buffering effect is higher on depressive
and anxiety symptoms and stress than their counterparts.
Bootstrapped indirect effects Coef Confidence intervals Nevertheless, stressful events might impact differently on people
aged 60 or over, with their resilience turning out to be more
IADQ → (RSA) → PHQ 0.13 0.08, 0.17 (N)
fragile compared to people aged under 60.
0.06, 0.17 (BC)
Our findings gather a strong relevance, as even if older people
IADQ → (RSA) → GAD 0.10 0.07, 0.14 (N)
have stronger aptitudes to cope with difficult situations, they
0.05, 0.13 (BC)
might experience what has been defined as a “double-burden,”
IADQ → (RSA) → PSS 0.10 0.06, 0.13 (N)
having their buffering abilities compromised by the disease
0.08, 0.14 (BC)
itself (27). Indeed, in addition to their physical vulnerability,
PHQ, Patient Health Questionnaire; GAD, Generalized Anxiety Questionnaire; PSS, in terms of infection risk, morbidity and mortality, they might
Perceived Stress Questionnaire; RSA, Resilience Scale for Adults; IADQ, International acquire a mental vulnerability, which would then lead to alarming
Adjustment Disorder Questionnaire – stressful events checklist. scenarios with worse clinical outcomes.
The present work suffers from a number of important
limitations – notably related to the on-line sampling technique
According to the latest researches, COVID-19 pandemic has and due to its cross-sectional fashion. Firstly, social network-
caused a significant increase in the prevalence of anxiety, stress based recruitment carries a significant selection bias, as people
and depressive symptoms (3, 4, 24). Indeed, it is not surprising are self-selected, without inclusion criteria, and measures are
that our findings showed a positive association between stressful self-reported. Moreover, on-line sampling may introduce a
events and depressive symptoms, anxiety and perceived stress. selection bias excluding subjects with poor informatic literacy
Notably, even controlling for potentially confounding factors like or even cognitive deficits. As a matter of fact, internet use is
education levels, results were still significant. Indeed, according associated with a number of factors that promote successful
to previous studies, people with higher levels of education are aging, including cognitive functioning and higher socioeconomic
more likely to develop depression and anxiety as they might be status (28). Secondly, although follow-up data will be collected
more aware of their own state of health (25). across time, the cross-sectional design of this study does not

Frontiers in Psychiatry | www.frontiersin.org 6 March 2021 | Volume 12 | Article 635832

Rossi et al. COVID-19, Stress and Resilience

leave enough room for causal inferences. For these reasons, and anxiety symptoms and perceived stress. Our findings suggest
caution must be taken into account when generalizing results that resilience plays a central role in protecting older adults from
to the population. Thirdly, limitations concern the inability to psychological distress and should therefore be taken into account
address cognitive deficits using an on-line survey. Indeed, it is in general health policies as well as treatment strategies.
worth noticing how a number of relevant factors may negatively
influence the psychological distress in older adults, including DATA AVAILABILITY STATEMENT
physical comorbidities, functional and cognitive impairments
as well as loneliness and neglect. The counterpart of resilience The raw data supporting the conclusions of this article will be
is in fact frailty, defined as a condition of both physical and made available by the authors, without undue reservation.
mental dysregulation that leads to a higher vulnerability and
therefore worse health outcomes (29). However, data on physical ETHICS STATEMENT
comorbidities and dependency, that could definitely influence the
selected outcomes in the elderly, were not collected. The studies involving human participants were reviewed and
Key strengths of our research are represented by its large approved by University of L’Aquila. The patients/participants
sample size and its distinctive timing in collecting data, which was provided their written informed consent to participate in
gathered when lockdown measures were implemented in Italy. this study.
In conclusion, we found that older age is associated with
higher levels of resilience. This would allow them to face weighty AUTHOR CONTRIBUTIONS
adversities such as COVID-19 pandemic more powerfully than
the others. However, stressful life events may act stronger on RR, VS, FP, and GL: conceptualization. RR: methodology, formal
them, compromising their buffering coping abilities, with higher analysis, and data curation. RR and TJ: writing–original draft.
rates of depression, anxiety and stress. As COVID-19 pandemic RR, VS, TJ, FP, and GL: writing– review and editing. All authors
might have brought us in a new era of communication and contributed to the article and approved the submitted version.
technological progress, it will be important to develop more
home-based agendas that would improve older people well-being ACKNOWLEDGMENTS
and therefore enhance their resilience.
This study was supported by Territori Aperti, a project funded by
CONCLUSION the Fondo Territori Lavoro e Conoscenza of the Confederazione
Generale Italiana del Lavoro, the Confederazione
This study addressed the age-dependent effect of resilience in Italiana Sindacati Lavoratori, and the Unione Italiana
mediating the impact of COVID-19 stressful events on depressive del Lavoro.

REFERENCES 8. Gooding PA, Hurst A, Johnson J, Tarrier N. Psychological resilience

in young and older adults. Int J Geriatr Psychiatry. (2012) 27:262–70.
1. Wister A, Speechley M. COVID-19: pandemic risk, resilience and doi: 10.1002/gps.2712
possibilities for aging research. Can J Aging. (2020) 39:344–7. 9. Wilson CA, Walker D, Saklofske DH. Developing a model of resilience
doi: 10.1017/S0714980820000215 in older adulthood: A qualitative meta-synthesis. Ageing Soc. (2020). 1–23.
2. Barzilay R, Moore TM, Greenberg DM, DiDomenico GE, Brown LA, White doi: 10.1017/S0144686X20000112
LK, et al. Resilience, COVID-19-related stress, anxiety and depression during 10. Killgore WDS, Taylor EC, Cloonan SA, Dailey NS. Psychological resilience
the pandemic in a large population enriched for healthcare providers. Transl during the COVID-19 lockdown. Psychiatry Res. (2020) 291:113216.
Psychiatry. (2020) 10:1–8. doi: 10.1038/s41398-020-00982-4 doi: 10.1016/j.psychres.2020.113216
3. Rossi R, Socci V, Pacitti F, Mensi S, Di Marco A, Siracusano A, et al. 11. Karasar B, Canli D. Psychological resilience and depression during the
Mental health outcomes among healthcare workers and the general COVID-19 pandemic in turkey. Psychiatr Danubina. (2020) 32:273–9.
population during the COVID-19 in Italy. Front Psychol. (2020) 11:3332. doi: 10.24869/psyd.2020.273
doi: 10.3389/fpsyg.2020.608986 12. Hayes AF. Introduction to Mediation, Moderation, and Conditional Process
4. Rossi R, Socci V, Talevi D, Mensi S, Niolu C, Pacitti F, et al. Analysis: A Regession Approach. New York, NY: The Guilford Press (2018).
COVID-19 pandemic and lockdown measures impact on mental health 13. Spitzer RL, Kroenke K, Williams JBW, Löwe B. A brief measure for
among the general population in Italy. Front Psychiatry. (2020) 11:790. assessing generalized anxiety disorder. Arch Intern Med. (2006) 166:1092.
doi: 10.3389/fpsyt.2020.00790 doi: 10.1001/archinte.166.10.1092
5. Kang SJ, Jung SI. Age-related morbidity and mortality among patients with 14. Shevlin M, Hyland P, Ben-Ezra M, Karatzias T, Cloitre M, Vallières F,
COVID-19. Infect Chemother. (2020) 52:154–64. doi: 10.3947/ic.2020.52.2.154 et al. Measuring ICD-11 adjustment disorder: the development and initial
6. Jeste DV, Savla GN, Thompson WK, Vahia IV, Glorioso DK, Martin AS, validation of the international adjustment disorder questionnaire. Acta
et al. Association between older age and more successful aging: critical Psychiatr Scand. (2020) 141:265–74. doi: 10.1111/acps.13126
role of resilience and depression. Am J Psychiatry. (2013) 170:188–96. 15. Capanna C, Stratta P, Hjemdal O, Collazzoni A, Rossi A. The Italian validation
doi: 10.1176/appi.ajp.2012.12030386 study of the resilience scale for adults (RSA). Appl Psychol Bull. (2015)
7. Southwick SM, Bonanno GA, Masten AS, Panter-Brick C, Yehuda R. 272:16–24.
Resilience definitions, theory, and challenges: interdisciplinary perspectives. 16. González-Sanguino C, Ausín B, Castellanos MÁ, Saiz J, López-Gómez A,
Eur J Psychotraumatol. (2014) 5:25338. doi: 10.3402/ejpt.v5.25338 Ugidos C, et al. Mental health consequences during the initial stage of the 2020

Frontiers in Psychiatry | www.frontiersin.org 7 March 2021 | Volume 12 | Article 635832

Rossi et al. COVID-19, Stress and Resilience

coronavirus pandemic (COVID-19) in Spain. Brain Behav Immun. (2020) 24. Salari N, Hosseinian-Far A, Jalali R, Vaisi-Raygani A, Rasoulpoor
87:172–6 doi: 10.1016/j.bbi.2020.05.040 S, Mohammadi M, et al. Prevalence of stress, anxiety, depression
17. Nwachukwu I, Nkire N, Shalaby R, Hrabok M, Vuong W, Gusnowski A, et al. among the general population during the COVID-19 pandemic: a
COVID-19 pandemic: age-related differences in measures of stress, anxiety systematic review and meta-analysis. Global Health. (2020) 16:1–11.
and depression in Canada. Int J Environ Res Public Health. (2020) 17:6366. doi: 10.1186/s12992-020-00589-w
doi: 10.3390/ijerph17176366 25. Zhang Y, Ma ZF. Impact of the COVID-19 pandemic on mental health
18. Ozamiz-Etxebarria N, Dosil-Santamaria M, Picaza-Gorrochategui M, and quality of life among local residents in Liaoning Province, China:
Idoiaga-Mondragon N. Niveles de estrés, ansiedad y depresión en la a cross-sectional study. Int J Environ Res Public Health. (2020) 17:2381.
primera fase del brote del COVID-19 en una muestra recogida en doi: 10.3390/ijerph17072381
el norte de España. Cadernos de Saúde Pública. (2020) 36:e00054020. 26. MacLeod S, Musich S, Hawkins K, Alsgaard K, Wicker ER. The
doi: 10.1590/0102-311x00054020 impact of resilience among older adults. Geriatr Nurs. (2016) 37:266–72.
19. Su TP, Lien TC, Yang CY, Su YL, Wang JH, Tsai SL, et al. Prevalence doi: 10.1016/j.gerinurse.2016.02.014
of psychiatric morbidity and psychological adaptation of the nurses 27. Chen LK. Older adults and COVID-19 pandemic: resilience matters. Arch
in a structured SARS caring unit during outbreak: a prospective and Gerontol Geriatr. (2020) 89:104124. doi: 10.1016/j.archger.2020.104124
periodic assessment study in Taiwan. J Psychiatr Res. (2007) 41:119–30. 28. Angevaare MJ, Monnier AA, Joling KJ, Smalbrugge M, Schellevis FG,
doi: 10.1016/j.jpsychires.2005.12.006 Hertogh CMPM, et al. The application of the concept of resilience in aging
20. Huang Y, Zhao N. Generalized anxiety disorder, depressive symptoms research and older adult care: a focus group study. Front Med. (2020) 7:365.
and sleep quality during COVID-19 outbreak in China: a web- doi: 10.3389/fmed.2020.00365
based cross-sectional survey. Psychiatry Res. (2020) 288:112954. 29. Hale M, Shah S, Clegg A. Frailty, inequality and resilience. Clin Med J R Coll
doi: 10.1016/j.psychres.2020.112954 Phys Lond. (2019) 19:219–23. doi: 10.7861/clinmedicine.19-3-219
21. Wang Y, Di Y, Ye J, Wei W. Study on the public psychological
states and its related factors during the outbreak of coronavirus Conflict of Interest: The authors declare that the research was conducted in the
disease 2019 (COVID-19) in some regions of China. Psychol absence of any commercial or financial relationships that could be construed as a
Health Med. (2020) 26:13–22. doi: 10.1080/13548506.2020.17 potential conflict of interest.
46817
22. Tsitsika AK, Tzavela EC, JanikianM, Ólafsson K, Iordache A, Schoenmakers Copyright © 2021 Rossi, Jannini, Socci, Pacitti and Lorenzo. This is an open-access
TM, et al. Online social networking in adolescence: Patterns of use in six article distributed under the terms of the Creative Commons Attribution License (CC
European countries and links with psychosocial functioning. J Adoles Health. BY). The use, distribution or reproduction in other forums is permitted, provided
(2014) 55:141–7. doi: 10.1016/j.jadohealth.2013.11.010 the original author(s) and the copyright owner(s) are credited and that the original
23. Rathore FA, Farooq F. Information overload and infodemic in the publication in this journal is cited, in accordance with accepted academic practice.
COVID-19 pandemic. J Pakis Med Assoc. (2020) 70 (Suppl. 3):S162–5. No use, distribution or reproduction is permitted which does not comply with these
doi: 10.5455/JPMA.38 terms.

Frontiers in Psychiatry | www.frontiersin.org 8 March 2021 | Volume 12 | Article 635832