Ageism is a complex phenomenon that encompasses stereotypes, prejudice and discrimination against older adults, old age and aging (1). Age discrimination is becoming increasingly important in a time of rapid population aging and there is a growing literature on ageism focusing on the negative effects of age discrimination both at the individual and societal level (2, 3). According to World Health Organization (4), ageism is the most frequent social form of discrimination and COVID-19 perpetuated and triggered several forms of public ageism (5, 6).

A recent review reported that most of the studies explored other-directed ageism and only a small number investigated self-directed ageism and subjective perceived ageism (7).

Indeed, it has been reported that subjective perceived age discrimination (PAD) produces detrimental effects (8) with harmful consequences for physical, mental, and social health (9, 10). Research has generally focused on PAD in everyday situations (9, 11, 12). Different studies showed that it was associated with older age (12–14), gender (15), years of education (12, 13, 16, 17), lower socioeconomic status (12, 13, 16, 17), ethnicity (13), not being married (17–19), being unemployed (16, 19) and depressive symptoms (20). Ageism is also related to social victimization (21) as well as elder abuse (22).

Frailty is an age-related clinical condition characterized by a decline in physiological capacity across several systems, that increase the vulnerability to stressors resulting in an increased risk of negative outcomes, such as hospitalization, falls, institutionalization and mortality (23). Frailty is a multidimensional condition, with physical and psychosocial factors playing an important role in its development (24). Indeed, frailty can be adequately captured by the comprehensive geriatric assessment (CGA) (25), by exploring several domains such as physical and cognitive function, nutrition, multimorbidity, polypharmacy and social determinants that are key factors in the assessment and management of frailty in older people (26). The Multidimensional Prognostic Index (MPI) is a prognostic tool based on a standard CGA that showed a good discrimination and accuracy for mortality and other negative outcomes both in hospital (27) and community settings (28). Nowadays, the MPI has becoming one of the most used frailty instruments in older adults (29). Recently, a self-administered version of the MPI (SELFY-MPI) (30) and its short-form version (SELFY-MPI-SF) (31) were developed and validated in community-dwelling older people both showing strong agreement and precision when compared with the standard version of the MPI based on the CGA.

Ageism and frailty are conditions that share the possibility of being associated with negative outcomes for the older people; however, very few studies included frailty as a factor potentially related to ageism (32, 33) and little is known of the possible relationship between these two conditions. Recently, increased attention was given to ageism in healthcare setting since it is widely recognized that behavioral and attitudinal approach toward older adults requires more awareness among health-care professionals (34).

The aim of this study was to investigate the association between frailty and PAD in a community-dwelling older population.

The study sample included a total of 1,337 community-dwelling older people aged 65 years or over. with a mean age of 77.4 years (SD: 7.4, range 65–107 years). Women were 56.0% and almost a third of the subjects lived alone (n = 429; 31.5%).

Eighty-three participants (6.2%) declared to have been discriminated against for their old age (PAD-group) during the 5 years before the interview. Table 1 shows the characteristics of the study participants, grouped according to the presence or absence of PAD: the subjects included in the PAD-group were older (p = 0.03), more frequently women (p = 0.046), with greater economic difficulties (p = 0.001), lower level of cultural fruition (p = 0.022), social network (on average about two points less on the Lubben scale, p < 0.001), higher prevalence of negative affectivity (on average about 3.5 points more on the PANAS-negative scale, p < 0.001), and lower prevalence of positive affectivity (about two points less on the PANAS-positive scale, p = 0.032) compared to participants who did not declared PAD.

As shown in Table 1; Figure 1, a significantly higher level of frailty was observed in participants included in the PAD group than in subjects who did not report PAD: median SELFY-MPI-SF score = 0.25 (IQR 0.06–0.88) vs. 0.19 (IQR 0.06–0.25), respectively (p < 0.001). After adjustment for age and gender, the logistic binary regression analysis demonstrated a significant association between PAD and the SELFY-MPI-SF score (OR 1.19, 95%CI 1.03–1.37) and the PANAS-negative scale (OR 1.06, 95%CI 1.03–1.10), thus showing that only the degree of frailty and the extent of negative affect were independent predictors of PAD (Figure 2).

In this study we explored the association between ageism and self-assessed frailty in a large sample of community-dwelling older people. A relatively small proportion of the study participants enrolled in the PRESTIGE project experienced PAD (6.2%), a percentage lower than that reported in other studies (12, 13, 47, 48). This finding can be explained by the characteristics of the participants included in the study, i.e., socially and culturally active older people with a low degree of physical and cognitive limitations. Nevertheless, this study contributes to provide relevant information on the relative role of frailty and several socio-economic, cultural, and psychological factors in fostering the subjective perception of age-based discrimination in older people.

Most studies address people's attitudes toward older people, but few explores the subjective PAD of older subjects themselves (7, 46). In our study, a greater degree of frailty and a poor psychological well-being constitute the personal characteristics with the strongest association with subjective perceived ageism. While the association with depressive symptoms had already been stressed in other studies (20), the association of PAD with frailty is a new finding not previously highlighted. One study (32) investigated the cross-sectional association of ageist attitudes with frailty in Veterans 50 years and older but neither explicit nor implicit ageist attitudes were found associated with frailty. In our sample each increase of one decimal point in the SELFY-MPI-SF value is associated with about 20% increase in the likelihood of feeling a victim of ageism.

It should be also emphasized that the extent of the association of frailty and negative affects with PAD attenuate in our survey the role of other factors that emerged in previous studies (12–15), such as age, gender, cultural factors, and economic conditions.

The present study has some limitations. First, the study sample including active seniors attending U3A can be considered mostly a “low risk” group according to the MPI risk of frailty categories (26, 27). Therefore, a selection bias in our findings is possible. Second, the choice of ascertaining the presence of subjective ageism with an explicit question that only admitted “yes” or “no” as an answer, may not have caught more hidden forms of ageism. Third, multidimensional frailty was assessed only through a self-reported tool: even if SELFY-MPI has a good agreement with standard MPI, some differences could be present using a standard form of MPI. Finally, this study does not allow to define the mechanisms of association between frailty and PAD.

As researchers, we are interested in potential interventions against ageism and the first step is to develop appropriate strategies and to target interventions on specific more vulnerable older populations. For this reason, further studies are needed to verify whether interventions aimed at preventing or treating frailty are also capable of reducing PAD.

Perceived age discrimination (PAD) in community-dwelling older people is associated with the degree of individual frailty, target of possible interventions aimed both at reducing ageism and improving the well-being of the older adults.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Comitato Etico della Liguria. The patients/participants provided their written informed consent to participate in this study.

AP planned the study, supervised the data analysis, and the paper submission. SP planned the study, contributed to revising the paper, and wrote the section Design and Participants. AC performed all statistical analysis and wrote the remaining sections of the paper. NV supervised all statistical analysis and contributed to revising the paper. SZ wrote the remaining sections of the paper. VP helped to plan the study and wrote the section Statistical Analysis. CT planned the study and wrote the section Frailty. EZ and PG collected data. All authors contributed to the article and approved the submitted version.

This work was supported by Fondazione CARIGE (Stronger, less frail GRANT 2018).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.