jfa journal

AND option

OR option



M. Canevelli1,2,*, F.S. Bersani1,*, F. Sciancalepore1, M. Salzillo1, M. Cesari3,4, L. Tarsitani1, M. Pasquini1, S. Ferracuti1, M. Biondi1, G. Bruno1


1. Department of Human Neurosciences, Sapienza University of Rome, Rome, Italy; 2. National Center for Disease Prevention and Health Promotion, Italian National Institute of Health, Rome, Italy; 3. Geriatric Unit, IRCCS Istituti Clinici Scientifici Maugeri, Milan, Italy; 4. Department of Clinical Sciences and Community Health, University of Milan, Milan, Italy; *These authors contributed equally to the work.

Corresponding Author: Marco Canevelli, Francesco Saverio Bersani, Department of Human Neurosciences, Sapienza University of Rome, Viale dell’Università 30, 00185, Rome, Italy, marco.canevelli@uniroma1.it, francescosaverio.bersani@uniroma1.it

J Frailty Aging 2021;in press
Published online June 25, 2021, http://dx.doi.org/10.14283/jfa.2021.29



Background: Studies increasingly suggest that chronic exposure to psychological stress can lead to health deterioration and accelerated ageing, thus possibly contributing to the development of frailty. Recent approaches based on the deficit accumulation model measure frailty on a continuous grading through the “Frailty Index” (FI), i.e. a macroscopic indicator of biological senescence and functional status.
OBJECTIVES: The study aimed at testing the relationship of FI with caregiving, psychological stress, and psychological resilience.
DESIGN: Cross-sectional study, with case-control and correlational analyses.
PARTICIPANTS: Caregivers of patients with dementia (n=64), i.e. individuals a priori considered to be exposed to prolonged psychosocial stressors, and matched controls (n=64) were enrolled.
MEASUREMENTS: The two groups were compared using a 38-item FI condensing biological, clinical, and functional assessments. Within caregivers, the association of FI with Perceived Stress Scale (PSS) and Brief Resilience Scale (BRS) was tested.
RESULTS: Caregivers had higher FI than controls (F=8.308, p=0.005). FI was associated directly with PSS (r=0.660, p<0.001) and inversely with BRS (r=-0.637, p<0.001). Findings remained significant after adjusting for certain confounding variables, after excluding from the FI the conditions directly related to psychological stress, and when the analyses were performed separately among participants older and younger than 65 years.
CONCLUSIONS: The results provide insight on the relationship of frailty with caregiving, psychological stress, and resilience, with potential implications for the clinical management of individuals exposed to chronic emotional strain.

Key words: Frailty, stress, resilience, caregiving, psychopathology, comorbidity.



Frailty has been originally conceptualized in the context of geriatric research to describe a condition of reduced homeostatic reserves and increased vulnerability to exogenous and endogenous stimuli often characterizing older adults (1). Although a definition of frailty has long been debated, it has recently been defined by the World Health Organization as “a progressive age-related decline in physiological systems that results in decreased reserves of intrinsic capacity, which confers extreme vulnerability to stressors and increases the risk of a range of adverse health outcomes” (2), and by an international consensus group as “a medical syndrome with multiple causes and contributors that is characterized by diminished strength, endurance, and reduced physiologic function that increases an individual’s vulnerability for developing increased dependency and/or death” (3).
Overall, frailty is considered to represent an estimate of organism’s biological age and, as such, it is increasingly explored in several medical areas to account for the interindividual variability in health trajectories and outcomes (4-6). Recent approaches measure frailty on a continuous grading according to the model of deficit accumulation, which postulates that the individual’s degree of frailty is related to the amount of health deficits accumulated with aging; accordingly, one’s biological and clinical complexity can be estimated by condensing such negative attributes in a single continuous variable, the Frailty Index (FI) (5, 7-11).
The concept of psychological stress is nowadays increasingly explored in virtually all fields of health-related research. It has been shown that exposure to psychological and social stress in childhood and adulthood, as well as the cumulation of psychosocial stressors over time, can have a significant negative impact on health mediated by a range of molecular mechanisms including, but not limited to, alterations in hypothalamic–pituitary–adrenal axis functioning and inflammatory responses (12).
Exposure to chronic psychological stress has been associated with increased vulnerability and worse outcomes related to cardiovascular, mental, metabolic, oncological, and infectious diseases, as well as to disability and earlier mortality (12). As a consequence, it seems possible that chronic psychological stress represents a factor contributing to the development of frailty. Such possibility is supported by the evidence linking psychological stress to the molecular mediators of biological age: while the condition of frailty has been proposed to represent a product of senescent biological age (4-6, 9, 13), studies have suggested that chronic exposure to psychological stress play a role in accelerating biological aging, i.e. leading to premature senescence, as documented, for example, by the widely replicated association of perceived psychological stress and stress-related psychiatric disorders with shorter telomere length (TL) in leukocytes (14-16).
Relatively few research has been performed so far to test the association between psychological stress and frailty. Further, to the best of our knowledge, no studies have explored such relationship among caregivers of patients with dementia, i.e. individuals exposed to severe and prolonged psychological burden (17), and operationalizing frailty through the FI as deficit accumulation. It can be hypothesized that caregiving, intended as a condition of chronic psychological stress exposure, is associated with accelerated senescence and higher accrual of health deficits, and that, among caregivers, frailty levels are directly related to the intensity of perceived psychological stress and inversely related to the individual’s capacity of psychological resilience, i.e. the capacity of maintaining positive emotional responses in the presence of psychosocial stressors. Therefore, the aim of the present study was (i) to compare biological age and functional status assessed through the FI in caregivers and matched controls, and (ii) within caregivers, to test the association of FI with measures of perceived psychological stress and resilience.




A total of 128 individuals were enrolled in the study: 64 caregivers of patients with dementia (i.e. individuals a priori considered as exposed to prolonged psychosocial stressors) and 64 non-caregiver matched controls. Such amount of participants was selected as a sample size calculation performed with the G*Power 3.1 software (18) indicated that 64 subjects in each group are needed to achieve an effect size of 0.5 with power=0.80 and alpha=0.05 (two-tailed) in between-group comparisons.
Caregivers were consecutively recruited through their carereceiver’s healthcare professionals at the Memory Clinic of Policlinico Umberto I University Hospital of Rome (Italy). Non-caregiver controls were recruited among volunteers in order to be age- and sex-matched with caregivers. Inclusion criteria for caregivers were (i) being spouse, child, sibling, or parent of a patient with a major neurocognitive disorder diagnosed according to the criteria of the Diagnostic and Statistical Manual of Mental Disorders 5 (DSM5) (19), (ii) living with the carereceiver, and (iii) having the role of principal caregiver from at least two years. Caregivers and controls were not included if they had a clinically unstable and disabling medical condition at the time of the evaluation.
Participants received a complete explanation of the procedures of the study and provided consent for allowing the utilization of the collected data for research purposes. Participants were not compensated for the participation to the study.

Procedures and assessments

The frailty status of all participants (caregivers and controls, n=128) was evaluated using a 38-item FI designed according to the model of Rockwood and Mitniski, following a standard procedure (5, 9-11). As mentioned above, the FI is a health-state measure, condensing information from multiple physiological systems; it is thought to reflect individual’s biological age and vulnerability to adverse outcomes. The FI is defined as the ratio between the deficits presented by the subject and the number of deficits explored in the context of a comprehensive clinical assessment. Candidate variables include symptoms, signs, comorbidities, and disabilities meeting standardized criteria (11). Each composing item is coded as “0” or “1” depending on whether the corresponding health deficit is absent or present, respectively. The FI thus provides a continuous measure of frailty potentially ranging between 0 and 1 for each individual. A cut-off of 0.25 has been adopted to identify frail and non-frail subjects (20-22). The deficits considered for the computation of the FI used in the present study emerged from biological (e.g. oxygen saturation, systolic blood pressure, body mass index [BMI], heart rate), clinical (e.g. current and past pathologies, chronic conditions, comorbidities) and functional (e.g. difficulties in transportations, money management, household keeping) evaluations, and they are listed in Table 1. In the present study the Cronbach’s α for the overall sample (caregivers and controls) was 0.76.

Table 1. Deficits considered in the computation of the Frailty Index


In addition to the FI, 62 of the caregivers were also evaluated with the 10 Items Perceived Stress Scale (PSS) (23-24), and with the Brief Resilience Scale (BRS) (25-27). In the PSS, respondents are asked to rate how often they experienced psychological stress in the past month on a 5-point Likert-type scale ranging from “Never = 0” to “Very Often = 4”, and a total score is calculated such that higher score reflects higher perceived psychological stress (23-24). In the present study, the Italian version of the scale was used (24), and the Cronbach’s α was 0.85. In the BRS, respondents are asked to rate their ability to recover from psychosocial stress by answering to 6 questions on a 5-point Likert-type scale ranging from “strongly disagree=1” to “strongly agree= 5”, and a total score is calculated such that higher score reflects higher resilience (25-27). Further, the authors of BRS proposed the following scores to qualitatively differentiate levels of resilience: below 3.00=low resilience, above 4.30=high resilience (27). In the present study, the Italian version of the scale was used (25), and the Cronbach’s α was 0.89.

Statistical methods

The Statistical Package for the Social Sciences (SPSS) was used for statistical calculations. All tests were 2-tailed with alpha=0.05. Quantitative data are expressed as means ± standard deviations (SD).
Parametric tests were performed as data were normally distributed (skewness and kurtsosis between -2 and 2). Analysis of variance (ANOVA) was used to test inter-group differences in FI between caregivers and controls, while analysis of covariance (ANCOVA) was used to control for confounding variables. Pearson correlations and partial correlations were performed to test the association of PSS and BRS with FI within caregivers.
ANOVA, Pearson correlation, and chi-square (Χ²) tests were also used to perform the following sensitivity analyses: (i) analyses performed using a modified FI computed after the exclusion of variables directly related to psychological stress; (ii) analyses performed separately among participants older and younger than 65 years; and (iii) analyses performed to further stratify/investigate the characteristics of the enrolled sample).



Almost all caregivers were spouses/partners (n=42, 65.6%) or children (n=20, 31.2%) of their carereceivers. The duration of their role as caregivers ranged between six and ten years in almost a half of cases (48.3%) and was lower than five years in 33.3% of cases. There were no significant differences between caregivers and non-caregiver controls by age, BMI, sex, and use of medications (Table 2).

Table 2. Characteristics of caregivers and controls

Abbreviations: BMI=Body Mass Index; ANOVA= Analysis of variance; X²=Chi squared; BMI data were available for 54 caregivers and 51 controls, information on medications was available for 33 caregivers and 34 controls.


Consistently with previous studies, the distribution of the adopted 38-item FI showed right skewness with a maximal score below 0.7 (scores ranged between 0 and 0.58).

In the overall study population, the adopted 38-item FI had a characteristic (11) right-skewed distribution with scores ranging between 0 and 0.58 (Figure 1). The mean FI value was 0.19±0.12, the median value was 0.16 (interquartile range [IQR] 0.11-0.26), and the 99th percentile was 0.56. The FI scores were higher in women than in men (0.20±0.12 vs. 0.16±0.09; F=5.708, p=0.018). Across all participants, FI scores were positively correlated with chronological age (r=0.419, p<0.001) (Figure 2). Based on the above-mentioned qualitative FI cut-off of 0.25, 28.1% (n=36) of participants could be classified as frail.

Figure 1. Distribution of FI values in caregivers and controls. Data are shown as %. FI scores are on the X-axis

Consistently with previous studies, the distribution of the adopted 38-item FI showed right skewness with a maximal score below 0.7 (scores ranged between 0 and 0.58).


The mean FI value was 0.21±0.12 in caregivers and 0.16±0.11 in controls. One-way ANOVA determined highly significant group differences between caregivers and controls (F[1, 127] = 8.308, p = 0.005) (Figure 3), and an extended model using age and sex as covariates did not alter the significance of this result (F[1, 124] = 11.247, p = 0.001). Using the above-mentioned FI cut-off, higher frailty prevalence was observed among caregivers compared with controls (24 vs 12, i.e. 37.5% vs. 18.8%; Χ²=5.565; p=0.018). Resilient caregivers (n=17), i.e. those with high resilience according to the above-mentioned BRS cut-off, had mean FI similar (non significantly lower) than controls (0.11±0.06 vs 0.16±0.11, F=2.247; p=0.138).

Figure 2. Scatterplot showing the significant (r=0.419, p<0.001) association of FI with age across all participants (caregivers are in black, controls are in white)

Figure 3. Graph showing the significant difference in FI between caregivers and controls (F=8.308, p=0.005). Bars indicate standard error


Within caregivers, mean scores of PSS and BRS were 17.94±8.90 and 3.36±1.07, respectively. FI was significantly positively associated with PSS (r=0.660, p<0.001), and it was significantly negatively associated with BRS (r=-0.637, p<0.001) (Figure 4). These correlations remained statistically significant (both p≤0.001) when age, sex, education, BMI, years of caregiving, and type of relationship with the carereceiver (i.e. being spouses/partners, children, siblings, or parents of carereceivers) were included as covariates. The FI was not significantly different between caregivers who were spouses/partners of the carereceivers (n=42) and caregivers who were children of the carereceivers (n=20) (F=0.866; p=0.356).

Figure 4. Scatterplots showing the significant association of FI with perceived psychological stress (PSS, r=0.660, p<0.001) and resilience (BRS, r=-0.637, p<0.001) within caregivers


As four of the 38 variables included in the FI are directly related to psychological stress (i.e. irritability, anhedonia, fatigue, sleep disorders), we created a second modified FI (defined as 34-item FI) excluding such variables from the computation of the score to be tested in sensitivity analyses. Performing between-group comparisons (caregivers vs controls) on 34-item FI and correlation analyses between 34-item FI, PSS and BRS within caregivers still gave significant findings (data not shown).
For further sensitivity, we performed between-group and within-group analyses (i) only including participants with an age higher than 64 years (caregivers n=38, controls n=39), and (ii) only including participants with an age in the range of 18-64 years (caregivers n=26, controls n=25). The mean FI values were again significantly higher in caregivers than in controls (older adults: 0.25±0.12 vs 0.19±0.11, F=4.090, p=0.047; adults: 0.17±0.10 vs 0.10±0.07, F=7.216, p=0.010), and, within caregivers, they were again significantly directly associated with PSS and inversely associated BRS in both groups (all p ≤0.003 by zero-order correlations).



Consistently with the hypotheses, in the present research we observed that caregivers had more pronounced frailty, as assessed by the FI, than controls (Figure 3), and that within caregivers the degree of frailty was directly associated with the degree of perceived psychological stress and inversely associated with the degree of resilience capacity (Figure 4). The statistical significance of such findings remained when certain potentially confounding variables were controlled for, when certain conditions directly related to psychological stress (irritability, anhedonia, fatigue, sleep disorders) were excluded from the computation of FI, and when the analyses were performed separately among participants older and younger than 65 years.
Previous evidence suggested that caregivers can show more pronounced health deficits and biological senescence than non-caregivers as a consequence of the prolonged psychological burden of caregiving (17, 28), that psychological stress has a deteriorating effect on health (12), that resilience skills can contribute to reduce the negative impact of psychological stress on health (17, 29), and that frail older adults have more stress-related psychological symptoms than non-frail older adults (30). Further, studies suggested that psychological stress is associated with accelerated/premature biological senescence measured through molecular indicators of ageing, i.e. molecular systems which are thought to reflect, mediate or promote cellular ageing and biological senescence (e.g. TL, mitochondrial DNA copy number, epigenetic signatures), while psychological resilience can have a protective role (9, 14, 16, 31-33). The findings of the present research are consistent with such data, and add novelty and specificity to the field as (i) the enrolled sample had a mean age of 67.71±11.57 with ages ranging between 34 and 89, while previous studies linking FI with psychopathology were mainly focused on older adults (i.e. individuals with age ≥65) (34-37); (ii) the health and biological measures were conceptualized within the multidimensional construct of frailty evaluated on a continuous grading (FI) rather that dichotomously (i.e. frail vs non frail subjects); (iii) information was collected multimodally, i.e. by integrating senescence- and health-related dimensions emerging from biological (e.g. oxygen saturation, systolic blood pressure, BMI, heart rate), clinical (e.g. current and past pathologies, chronic conditions, comorbidities) and functional (e.g. difficulties in transportations, money management, household keeping) evaluations (Table 1).
In relation to frailty, the findings of the present study are in line with previous observations emerging from studies based on FI in which different health deficits were considered: FI scores were significantly associated with chronological age (Figure 2) (7); significantly higher FI values were observed in women than in men, consistently with previous observations of sex-specific differences in FI scores, which are often defined in the literature as the “male-female health-survival paradox” (38); the FI exhibited a right-skewed distribution with a maximal value of 0.58 (Figure 1), consistently with previous data showing that FI had an upper limit of 0.7, which have been explained by the fact that people cannot tolerate and survive health deficits above a certain threshold (39).
In relation to the health status of caregivers, our results support the relevance of resilience to psychosocial stress for such at-risk individuals: (i) among caregivers, higher levels of frailty were significantly associated with higher perceived psychological stress and with lower capacity of psychological resilience; (ii) across all subjects, while the FI of caregivers was significantly higher than that of controls, the FI of those caregivers with high levels of resilience was similar (non significantly lower) than that of controls. Such data can contribute to extend to psychological resilience the evidence suggesting that physical resilience (which has been defined as the capacity of function maintenance or recovery following biomedical or pathological challenges) plays a protecting role towards the development of frailty (40). Further, consistently with previous studies (17, 41), these findings support the possibility that being the caregiver of a chronically ill person does not inevitably lead to a more deteriorated health profile, and that the way caregivers respond to adversities play a relevant role in determining how psychosocial stressors affect health. From a clinical perspective, it is thus possible that interventions on caregivers focused on decreasing perceived psychological stress and increasing psychological resilience skills can play a role in the improvement of their frailty and general health status. Of relevance, as this was a cross-sectional study, longitudinal and causal relationships between such variables cannot be established, and both directions of causality are potentially possible.
Overall, while the study of the relationship between psychological stress, health and ageing through the analysis of molecular markers of ageing is closely related to pathophysiology, the study of such link through a FI can be more closely related to clinical and functional state, thus possibly expanding the knowledge originating from such area of investigation (9). Relatedly, recent studies in older adults have observed FI to be significantly associated with TL and epigenetic clocks in leukocytes (9, 42-44), and preliminary studies have integrated various mlecular indicators of ageing within a biomarker-based FI (45). Among the molecular indicators of ageing, increasing attentions is nowadays given to the so-called “inflammaging”, i.e. chronically increased levels of inflammatory cytokines which are thought to underlie the progression of senescence-related processes (9, 46). While the FI adopted in the current study did not incude the assessment of cytokine levels, it did consider certain conditions which are tightly related to increased inflammation, such as cancer, osteoporosis, diabetes, and previous episodes of transient ischemic attack or stroke; subsequently, the adopted deficit accumulation approach may capture certain aspects of inflammation even in absence of molecular measures of inflammatory processes.
The present study has several limitations, among which: (i) the cross sectional nature of the research does not allow to establish causal relationships , if any, between correlated variables and the potential direction of causality; (ii) some pieces of information were not fully collected for all participants: complete BMI data were available for 54 caregivers and 51 controls, data on medication use were available for 33 caregivers and 34 controls, information on PSS, BRS, and education was collected in 62 of the 64 caregivers and not in controls, information on years of caregiving was collected in 60 of the 64 caregivers; (iii) the reliability of self-report measures (such as PSS and BRS) can be affected by several biases (e.g. social desirability bias, response bias) (47); (iv) mean age of caregivers and controls was 67.72 ± 11.59 and 67.70 ± 11.63, so the findings may not be applicable to cohorts of different ages, although when the analyses were performed separately among participants older and younger than 65 years the significance of the main results did not change; (v) the study sample size (n = 128) was adequate to test between-group differences on FI (as indicated by an a priori power analysis), but not to test within group correlations, although the observed significant associations of FI with PSS and BRS had large effect sizes (r=0.660 and -0.637 respectively); (vi) a selection bias may have occurred as recruiting caregivers through their carereceiver’s healthcare professionals may have favoured the inclusion of caregivers with higher levels of psychological stress; (vii) we performed a 1:1 match between cases and controls, while epidemiological studies suggest to enrol more than one control for every case to obtain the best methodology (48); (viii) although we statistically controlled for certain potential confounders, other residual confounding factors may have influenced the results of the study. Among the strengths, (i) participants of both groups were in good general health at the moment of the evaluation, i.e. they were not included if they currently had a clinically unstable and disabling medical condition; (ii) the FI was developed according to a well established procedure (11), it showed adequate reliability (Cronbach’s α=0.76), and it comprehensively assessed information arising from biological, clinical, and functional evaluations (the participants were thus assessed multimodally); (iii) we used PSS and BRS, which are extensively used and validated assessment instruments for psychological stress and resilience, and showed satisfactory Cronbach’s α was in the current sample (0.85 and 0.89 respectively).
In conclusions, our findings provide pieces of insight on the complex relationships of frailty, conceptualized as a measure of deficit accumulation and an indicator of functional status and biological age, with caregiving, psychological stress and resilience, with potential implications for the psychological and medical management of individuals exposed to chronic emotional strain.


Conflicts of Interest: The authors do not have conflicts of interest to declare.

Ethical standards: Participants provided consent for allowing the utilization of the collected data for research purposes.



1. Clegg, A., et al., Frailty in elderly people. Lancet, 2013. 381(9868): p. 752-62.
2. World Health Organization, World report on ageing and health. 2015: World Health Organization.
3. Morley, J.E., et al., Frailty consensus: a call to action. J Am Med Dir Assoc, 2013. 14(6): p. 392-7.
4. Hoogendijk, E.O., et al., Frailty: implications for clinical practice and public health. Lancet, 2019. 394(10206): p. 1365-1375.
5. Mitnitski, A.B., A.J. Mogilner, and K. Rockwood, Accumulation of deficits as a proxy measure of aging. ScientificWorldJournal, 2001. 1: p. 323-36.
6. Cesari, M., B. Vellas, and G. Gambassi, The stress of aging. Exp Gerontol, 2013. 48(4): p. 451-6.
7. Rockwood, K. and S.E. Howlett, Age-related deficit accumulation and the diseases of ageing. Mech Ageing Dev, 2019. 180: p. 107-116.
8. Cesari, M., et al., The frailty phenotype and the frailty index: different instruments for different purposes. Age Ageing, 2014. 43(1): p. 10-2.
9. Bersani, F.S., et al., Frailty Index as a clinical measure of biological age in psychiatry. J Affect Disord, 2020. 268: p. 183-187.
10. Canevelli, M., et al., Promoting the Assessment of Frailty in the Clinical Approach to Cognitive Disorders. Front Aging Neurosci, 2017. 9: p. 36.
11. Searle, S.D., et al., A standard procedure for creating a frailty index. BMC Geriatr, 2008. 8: p. 24.
12. Epel, E.S., et al., More than a feeling: A unified view of stress measurement for population science. Front Neuroendocrinol, 2018. 49: p. 146-169.
13. Junius-Walker, U., et al., The essence of frailty: A systematic review and qualitative synthesis on frailty concepts and definitions. Eur J Intern Med, 2018. 56: p. 3-10.
14. Bersani, F.S., et al., Accelerated aging in serious mental disorders. Curr Opin Psychiatry, 2019. 32(5): p. 381-387.
15. Mathur, M.B., et al., Perceived stress and telomere length: A systematic review, meta-analysis, and methodologic considerations for advancing the field. Brain Behav Immun, 2016. 54: p. 158-169.
16. Puterman, E. and E. Epel, An intricate dance: Life experience, multisystem resiliency, and rate of telomere decline throughout the lifespan. Soc Personal Psychol Compass, 2012. 6(11): p. 807-825.
17. Harmell, A.L., et al., A review of the psychobiology of dementia caregiving: a focus on resilience factors. Curr Psychiatry Rep, 2011. 13(3): p. 219-24.
18. Faul, F., et al., Statistical power analyses using G*Power 3.1: tests for correlation and regression analyses. Behav Res Methods, 2009. 41(4): p. 1149-60.
19. American Psychiatric Association, Diagnostic and statistical manual of mental disorders (5th ed.). 2013: American Psychiatric Publishing.
20. Song, X., A. Mitnitski, and K. Rockwood, Prevalence and 10-year outcomes of frailty in older adults in relation to deficit accumulation. J Am Geriatr Soc, 2010. 58(4): p. 681-7.
21. Hoogendijk, E.O., et al., Operationalization of a frailty index among older adults in the InCHIANTI study: predictive ability for all-cause and cardiovascular disease mortality. Aging Clin Exp Res, 2020. 32(6): p. 1025-1034.
22. Rockwood, K., M. Andrew, and A. Mitnitski, A comparison of two approaches to measuring frailty in elderly people. J Gerontol A Biol Sci Med Sci, 2007. 62(7): p. 738-43.
23. Cohen, S. and G. Williamson, Perceived stress in a probability sample of the United States, in The social psychology of health: Claremont Symposium on applied social psychology, S. Spacapan and S. Oskamp, Editors. 1988, Sage.
24. Mondo, M., C. Sechi, and C. Cabras, Psychometric evaluation of three versions of the Italian Perceived Stress Scale. Curr Psychol, 2019.
25. Laudadio, A., L. Mazzocchetti, and F.J. Fìz Perez, Gli stumenti disponibili in lingua italiana, in Valutare la resilienza: teorie, modelli e strumenti. 2011, Carocci Editore.
26. Smith, B.W., et al., The brief resilience scale: assessing the ability to bounce back. Int J Behav Med, 2008. 15(3): p. 194-200.
27. Smith, B.W., et al., The Foundations of Resilience: What Are the Critical Resources for Bouncing Back from Stress?, in Resilience in Children, Adolescents, and Adults, S. Prince-Embury and D.H. Saklofske, Editors. 2013, Springer.
28. Damjanovic, A.K., et al., Accelerated telomere erosion is associated with a declining immune function of caregivers of Alzheimer’s disease patients. J Immunol, 2007. 179(6): p. 4249-54.
29. Palacio, G.C., et al., Resilience in Caregivers: A Systematic Review. Am J Hosp Palliat Care, 2020. 37(8): p. 648-658.
30. Desrichard, O., et al., Frailty in aging and its influence on perceived stress exposure and stress-related symptoms: evidence from the Swiss Vivre/Leben/Vivere study. Eur J Ageing, 2018. 15(4): p. 331-338.
31. Bersani, F.S., et al., Association of dimensional psychological health measures with telomere length in male war veterans. J Affect Disord, 2016. 190: p. 537-542.
32. Epel, E.S., et al., Accelerated telomere shortening in response to life stress. Proc Natl Acad Sci U S A, 2004. 101(49): p. 17312-5.
33. Verner, G., et al., Maternal Psychological Resilience During Pregnancy and Newborn Telomere Length: A Prospective Study. Am J Psychiatry, 2021. 178(2): p. 183-192.
34. Benraad, C.E.M., et al., Frailty, multimorbidity and functional status as predictors for health outcomes of acute psychiatric hospitalisation in older adults. Aging Ment Health, 2018: p. 1-10.
35. Benraad, C.E.M., et al., Frailty as a Predictor of Mortality in Older Adults within 5 Years of Psychiatric Admission. Int J Geriatr Psychiatry, 2020.
36. Lohman, M., L. Dumenci, and B. Mezuk, Depression and Frailty in Late Life: Evidence for a Common Vulnerability. J Gerontol B Psychol Sci Soc Sci, 2016. 71(4): p. 630-40.
37. Aprahamian, I., et al., Frailty in geriatric psychiatry inpatients: a retrospective cohort study. Int Psychogeriatr, 2020: p. 1-9.
38. Gordon, E.H., et al., Sex differences in frailty: A systematic review and meta-analysis. Exp Gerontol, 2017. 89: p. 30-40.
39. Rockwood, K. and A. Mitnitski, Limits to deficit accumulation in elderly people. Mech Ageing Dev, 2006. 127(5): p. 494-6.
40. Whitson, H.E., et al., Physical Resilience: Not Simply the Opposite of Frailty. J Am Geriatr Soc, 2018. 66(8): p. 1459-1461.
41. Biondi, M., F.S. Bersani, and M. Pasquini, The Role of Integrated Interventions in Psychosomatic Diseases, in Person Centered Approach to Recovery in Medicine, L. Grassi, M.B. Riba, and T. Wise, Editors. 2019, Springer.
42. Araujo Carvalho, A.C., et al., Telomere length and frailty in older adults-A systematic review and meta-analysis. Ageing Res Rev, 2019. 54: p. 100914.
43. Kim, S., et al., The frailty index outperforms DNA methylation age and its derivatives as an indicator of biological age. Geroscience, 2017. 39(1): p. 83-92.
44. Breitling, L.P., et al., Frailty is associated with the epigenetic clock but not with telomere length in a German cohort. Clin Epigenetics, 2016. 8: p. 21.
45. Mitnitski, A., et al., Age-related frailty and its association with biological markers of ageing. BMC Med, 2015. 13: p. 161.
46. Franceschi, C., et al., Inflammaging: a new immune-metabolic viewpoint for age-related diseases. Nat Rev Endocrinol, 2018. 14(10): p. 576-590.
47. Demetriou, C., B.U. Ozer, and C.A. Essau, Self-Report Questionnaires, in The Encyclopedia of Clinical Psychology, R.L. Cautin and S.O. Lilienfeld, Editors. 2015, John Wiley & Sons.
48. Lewallen, S. and P. Courtright, Epidemiology in practice: case-control studies. Community Eye Health, 1998. 11(28): p. 57-8.