The World Health Organization (2020) defines healthy aging as “the process of developing and maintaining the functional capacity that enables well-being in older age” (p. 8). To promote healthy aging, researchers have explored the function of numerous determinants of health and well-being in old age (for a review, see Cardinal et al., 2008). Stress influences health, disease, and well-being and thus has an important place in models of healthy aging. Research to improve the management of stress responses in old age has been identified as a key priority in the field (Kennedy et al., 2014; Polsky et al., 2022).

Resilience to stress has been described as the ability to adapt and maintain one’s psychological health in the face of adversity across the lifespan (Troy et al., 2023). More recently, researchers have acknowledged the complex nature of psychological resilience, thus highlighting the multifaceted processes that it encompasses (Vella and Pai, 2019). In a process-based framework, psychological resilience involves personal competencies and characteristics, and the context and time in which they are displayed (Troy et al., 2023). The study of aging places particular emphasis on understanding resilience to stress, as individuals may undergo significant changes in physiological, psychological, and social roles during late adulthood (American Psychological Association, 2017). The scientific literature reports several types of stressors at this stage of life. In a sample of 282 adults aged 54 to 91 years (M = 68.79, SD = 5.10), Scott et al. (2011) identified the combined contribution of life events, neighborhood strain, age-related discrimination (i.e., ageism), social isolation (i.e., loneliness), financial strain, and physical health status to the experience of perceived stress. In a large study involving two samples of over 6,000 Americans aged 50 and over (M = 67 years old), the authors examined the impact of different types of stressors on the cognition. The most common sources of chronic stress included health problems (9.3 to 10.3%), physical or emotional problems in spouse or child (7.9 and 7.4%), and financial strain (4.9, 6.7%; Wang et al., 2022). A more recent study of 127 adults aged 60 to 95 years (M = 79.4, SD = 9.15) highlights the subjective aspect of stressors during aging through the contribution of individuals’ perceptions of their health and aging. Indeed, individuals who had more negative global and daily perceptions of their health and aging experienced greater perceived stress (Whitehead and Blaxton, 2021). As demonstrated by Wettstein et al. (2021), the relationship between perceptions of aging and stress is a reciprocal one, as stress also affects perceptions of aging.

Stress is a biological response that occurs when one’s external or internal balance (homeostasis) is challenged. The main hormonal circuit involved is the hypothalamic–pituitary–adrenal (HPA) axis. The end product of HPA activation is cortisol (Selye, 1936). In fact, cortisol is continuously secreted in the body according to a circadian rhythm (Brown and Wilkinson, 2015). Animal and human studies have shown a progressive increase in circadian cortisol concentrations in the first part of the night, with circulating cortisol levels increasing with age (Duffy et al., 2015). In a review, Goncharova (2020) mentioned that hyperactivation of the HPA axis has been found in older adults and in aging primate models. Another trend in aging research reported in two reviews (including studies in humans, nonhuman primates, and rodents) is a decrease in the strength of negative feedback from the glucocorticoids, such that the aging body is overexposed to cortisol while the process of regaining homeostasis is slowed, suggesting a limited psychophysiological coping capacity (Goncharova, 2020; Rao and Androulakis, 2019). Goncharova (2020) and Rao and Androulakis (2019) also report that age-related changes in the circadian rhythm of cortisol have been shown to also affect HPA reactivity, i.e., the cortisol response to stress exposure (e.g., by increasing HPA reactivity or decreasing the ability of the stress system to quickly return to a resting level).

Stressors come in both explicit and implicit forms. Explicit stressors include situations that almost everyone perceives as threatening. Relative threats, as opposed to absolute ones (i.e., those arising from life-threatening events), arise from a sense of novelty, its unpredictability, a threat to one’s ego, and a sense of lack of control over a situation (N.U.T.S.; Dickerson and Kemeny, 2004; Mason, 1968). They depend on the individual’s interpretation of the situation. These relative stressors can be acute (e.g., being late for an appointment) or more insidious, such as daily annoyances (e.g., frequent arguments with a partner) or chronic stress (e.g., caregiving; Epel et al., 2018). Individuals may also experience implicit stressors in interpersonal situations that trigger an automatic physiological stress response. Stress contagion is rooted in the phenomenon of emotional contagion (Hatfield et al., 2014), which can be defined as “the tendency to automatically mimic and synchronize facial expressions, vocalizations, postures, and movements with those of another person and, consequently, to converge emotionally” (Hatfield et al., 1994, p. 153). Physiological resonance is one of the primary mechanisms involved. Buchanan et al. (2012) demonstrated the contagion of the physiological response to acute stress between a target and an observer behind one-way glass. In their study, observers showed an increase in their cortisol (stress hormone) response that was proportional to that of the individuals undergoing social stress, illustrating the resonance of the stress response at the physiological level. When the observer and the target were significant others, Engert et al. (2014) and Engert et al. (2018) found a similar and even stronger resonance. We posit that in aging, individuals may find themselves in contexts that could triggers stress contagion such as caregiving and living in residences for retirement.

In short, age itself has an impact on the functioning of the stress system, but another important element is the burden of chronic stress that some older people have experienced throughout their lives, which may have worn down their stress system. In contrast to the stress response to acute stressors, the presence of prolonged or repeated stressors results in prolonged or repeated activation of the HPA axis, which may have long-term deleterious effects (Juster et al., 2011; Turner et al., 2020). Many studies have examined the effects of chronic stress in older adults and have identified an extensive list of health concerns that are particularly vulnerable to chronic stress.

The physiological changes that occur in the face of chronic stress can promote some diseases such as hypertension, type 2 diabetes (T2D), and cardiovascular disease (Kelly and Ismail, 2015; Osborne et al., 2020; Seiler et al., 2020; Steptoe and Kivimaki, 2012). Dysfunction of the hypothalamic–pituitary–adrenal (HPA) axis leads to increased secretion of glucocorticoids. This, in turn, disrupts glucose metabolism by impairing hepatic glucose uptake and insulin secretion, thereby contributing to insulin resistance and promoting inflammation, among other consequences. This pathway increases the risk of hypertension and T2D, which are also known risks for cardiovascular disease (Osborne et al., 2020; Sharma and Singh, 2020). Several combined mechanisms are also at play: the stress response includes increased sympathetic nervous system (SNS) activity and an opposite response in parasympathetic nervous system (PSNS) activity; increased leukopoiesis (i.e., the accumulation of fats, cholesterol, and other substances in and on the walls of the arteries); altered vascular reactivity and increased coagulation, and immune dysregulation leading to inflammation and to immunosuppression. All of these responses to chronic stress could make the body more susceptible to inflammatory diseases such as cardiovascular disease (Osborne et al., 2020). Moreover, chronic stress has been recognized for its impact on cognitive function. Although findings have varied over the years due to differences in the methods used to assess stress, evidence suggests that changes in the brain—particularly the hippocampus—resulting from chronic activation of the HPA axis may be associated with an increased risk of cognitive decline (Franks et al., 2020). A systematic review and meta-analysis of 26 and 16 studies conducted between 2003 and 2020, respectively, examined the association between adult stress-including life events and perceived stress-and the risk of cognitive impairment and all-cause dementia (Franks et al., 2020). Across studies, findings from Franks et al. (2020) and Luo et al. (2023) suggest that both stressful life events and perceived stress are associated with cognitive decline and an increased risk of all-cause dementia. Finally, stress is associated with a higher risk of reporting depressive and anxiety symptoms (Duric et al., 2016; Grolli et al., 2021; Mantella et al., 2008).

In contrast, research suggests that psychological resilience may be protective for older adults in the face of adversity. A large longitudinal study, involving two samples of 11,050 (mean age = 67.53, SD = 9.21) and 12,823 (mean age = 67.53, SD = 9.58) individuals aged 51 to 104 years, demonstrated that psychological resilience exerts independent and significant protective effects on health transitions and trajectories (Taylor and Carr, 2021). These effects were observed across five different health outcomes, including perceived health, functional limitations, activities of daily living, and depressive symptoms (Taylor and Carr, 2021). Despite inconsistent findings, other studies suggest that psychological resilience supports the preservation of cognitive functions during aging. For example, one study found that individuals (mean age of 55 years) who were rated as resilient had higher cognitive function 5 years later. Notably, this association was significant only among men with lower levels of education and was independent of age (Yang et al., 2021).

With increasing longevity, older adults are bound to face more challenges, underscoring the need for enhanced stress resilience among them. This highlights the importance of implementing stress management interventions in practice. The majority of stress management programs evaluated for the general population have included cognitive-behavioral stress management (CBSM; Hammerfald et al., 2006; Sharifirad et al., 2013; Tang et al., 2020) and mindfulness-based interventions (Khoury et al., 2015; Querstret et al., 2020). CBSM programs are usually based on the appraisal model of Lazarus and Folkman (1984). As for mindfulness methods, they are rooted in the practice of meditative techniques (Querstret et al., 2020). Both types of interventions frequently incorporate strategies recognized for their impact on altering the perception of stress or modulating the functioning of the HPA axis. These strategies include recognizing stressors and their effects on the mind and the body, breathing exercises, promoting problem-solving and emotion-focused coping behaviors, and cultivating the ability to identify internal and external coping resources, such as those within one’s social network. Meta-analyses of stress management interventions in different populations suggest that some interventions have been moderately successful in reducing perceived stress (Chiesa and Serretti, 2009; Querstret et al., 2020; Richardson and Rothstein, 2008). Unfortunately, few studies have evaluated the effectiveness of stress management interventions specifically in older adults. For example, a meta-analysis of 23 studies (including six randomized controlled trials) found a mindfulness-based stress reduction program to be more effective than the control group in reducing depressive symptoms in older adults immediately after the intervention, but their results provided no evidence of reduced anxiety and perceived stress. In addition, the effects on depressive symptoms dissipated at follow-up (Li and Bressington, 2019).

It’s worth noting that few programs are aligned with the past 35 years’ discoveries in the field of psychoneuroendocrinology. This field of research is concerned with understanding the causes and consequences of physiological reactivity to psychological stress in humans, such as the relative stressors (i.e., N.U.T.S.) and their effects on the body. Similar to the mindfulness approach, the Brief Training of Psychoneuroendocrinoimmunology-Based Meditation (BTP-BM) emphasizes the need for education about the stress system and its connections to emotions, cognition, and their biological aspects, while encouraging observation of their effects on behavior (Bottaccioli et al., 2014). The Center for Studies on Human Stress also developed programs rooted in the findings of psychoneuroendocrinology, combined with cognitive-behavioral strategies such as problem-solving and ways to use one’s social network as a coping strategy. These programs have demonstrated reductions in participants’ stress (both perceived and physiological), as well as in participants’ symptoms of distress among youth and social workers (Fecteau et al., 2018; Lupien et al., 2013; Massé and Plusquellec, 2018; Plusquellec et al., 2016).

The scientific literature that focuses on factors that moderate the effects of stress in older adults reports that they are less likely to experience distress symptoms when they use problem-solving strategies, social support–seeking coping styles, and positive reappraisal or cognitive reframing (Cosco et al., 2017; Kraaij et al., 2002). For example, Park and An (2016) examined the moderating effect of coping strategies on the relationship between stress (described as having a “loss experience”) and symptoms of depression in a sample of 156 women and 116 men over the age of 60. They found that problem-focused coping styles moderated the relationship between loss and depression in men.

To our knowledge, no stress management training for older adults in community settings has combined psychoneuroendocrinology findings with a cognitive-behavioral approach. The present study tested the efficacy of an adaptation of the program developed by the Center for Studies on Human Stress in a sample of older adults.

This study aimed to evaluate the effects of a stress management training (SMT) intervention in a sample of community-dwelling older adults on problem-focused coping strategies, mental health indicators (e.g., reductions in self-reported stress, anxiety, and depressive symptoms), and physiological indicators of stress (i.e., cortisol levels). Comparisons between the effects of SMT and a waitlist condition were made 3 weeks after the intervention and at a 3-month follow-up.

In this study, we examined the effects of a 6-week group-based stress management intervention with a stress contagion component. The intervention was adapted to meet the needs and characteristics of older adults.

First, 3 weeks after the intervention, we found a significant increase in the use of problem-solving coping strategies among participants in the experimental condition compared to the waitlist condition. Specifically, participants reported using more proactive and strategic planning coping styles after the intervention. Proactive coping includes strategies used to prevent or minimize the impact of future stressors (e.g., taking the initiative to solve a problem). Greenglass et al. (2006) found that when individuals are able to identify signs of a potential or imminent threat and have acquired skills to cope with it (e.g., planning and goal-setting), they are more likely to feel a sense of control that leads them to take adaptive actions. In a study that examined the relationship between proactive coping, daily hassles, functional disability, and other health factors in a sample of 224 community-dwelling older adults, the authors found that the use of proactive coping positively affected participants’ independence and well-being by contributing to the completion of daily tasks (Fiksenbaum et al., 2006). Our results suggest that the intervention supported the acquisition of these problem-solving skills. The development of these strategies was also among the most effective core components of meta-analyses of stress management interventions (Chiesa and Serretti, 2009; Querstret et al., 2020; Richardson and Rothstein, 2008). We postulate that, in addition to the intervention components, the group aspect may have positively contributed to our findings. Indeed, studies have demonstrated the positive influence of social support on the increased use of coping strategies (Fiksenbaum et al., 2006; Lee, 2018; Trouillet et al., 2009). The group intervention was tailored to provide participants with opportunities to talk about their stressors and to support each other. While the SMT intervention demonstrated significant improvements in proactive and strategic planning coping strategies at the 3-week posttest, these effects did not persist at the 3-month follow-up. Notably, both proactive coping strategies and strategic planning scores returned closer to baseline levels, suggesting that the initial gains from the intervention diminished over time. This decline could be attributed to several factors. First, the intervention’s impact may have been context-specific or short-lived, requiring ongoing reinforcement to maintain its effectiveness. Additionally, the absence of sustained effects may reflect the natural variability in individual coping practices or the possible lack of long-term integration of the learned strategies into daily routines. We recall that the effects of our SMT conditions were demonstrated against a waitlist condition, suggesting that we cannot disentangle the effects of the SMT component from the support group.

Second, our results showed significant effects on two measures of mental health. Exposure to the program appeared to result in decreased feelings of anxiety at 3 weeks postintervention and decreased perceived stress at the 3-month follow-up. It is notable, however, that while anxiety improved shortly after the intervention, these gains were not sustained over the longer term, as there was no significant change in anxiety levels at the 3-month mark. The paucity of research on CBT stress management interventions for older adults has obscured possible comparisons (Hofmann et al., 2012; Li and Bressington, 2019). A meta-analysis of seven randomized controlled trials comparing the efficacy of CBT (including group therapy) for late-life anxiety disorders with control conditions found significant reductions in anxiety, worry, and depressive symptoms after the intervention (Hendriks et al., 2008). In comparison, our program, that is shorter with its six sessions, suggested promising results on anxiety and stress in the short term (after 3 weeks and 3 months, respectively). The lack of sustained improvement in anxiety and the small differences in our psychological outcomes at the 3-month follow-up are in line with other experimental studies with older adults in which gains on stress, anxiety, or depression scales were rarely maintained (Hofmann et al., 2012; Li and Bressington, 2019). These findings may be due to the short duration of our intervention, coupled with the need for a sustained effort to form new habits for coping with stressors. Additionally, numerous adaptive challenges may emerge in older individuals, potentially limiting the long-term effectiveness of otherwise successful treatments and impacting their ability to maintain newly acquired skills without ongoing support.

The third finding of the present study concerns our program’s effect on the participants’ cortisol levels. The mean CAR was calculated to represent the increase in cortisol concentration that occurs in the first half-hour after awakening. Typically, the increase in glucocorticoids is about 50 to 75% and is thought to prepare individuals to engage with the environment to face the day ahead (Fries et al., 2009). No association was found between both groups of participants and changes in CAR over time. It should be noted that, at baseline, only 32.62% of our sample achieved an increase in cortisol concentration of at least 50%, suggesting lower cortisol levels upon awakening. However, it should be noted that we also found a large interindividual variability in the CAR results. The participants were diligent in filling in their saliva diaries, so we have access to the collection times for each sample included in the study. One possible explanation of those results is the duration required to collect the sample—information that was not compiled. For example, participants mentioned that xerostomia—a side effect of medication—caused complications in providing the 2 mL of saliva promptly. However, the size of our sample did not allow us to control for the many variables affecting CAR that could have addressed these concerns, such as smoking, illness, and medication, to name a few (Stalder et al., 2016; Stalder et al., 2022).

In contrast, participants in the SMT group showed a significant reduction in AUCg 3 weeks after the intervention, whereas cortisol levels increased in the control group. Exposure to chronic stress has been shown to alter diurnal cortisol patterns at any age (for a review, see Gaffey et al., 2016). In the context of age-related changes in the HPA axis, resilience to stress would be enhanced by the development of individual resources such as emotional regulation and social support. Conversely, a lack of these resources could exacerbate HPA axis alterations. More specifically, based on available theoretical data, the aging, stress, and resilience model (Gaffey et al., 2016) proposes that social support and coping style influence diurnal cortisol in aging. For example, older adults who experienced greater social strain from their network had elevated evening cortisol and a flatter cortisol slope across the day (Friedman et al., 2012). Rueggeberg et al. (2012) demonstrated that coping strategies also have an impact on AUCg. Their study highlighted that ill individuals who used more self-protective strategies (i.e., cognitive strategies to reevaluate a situation and to avoid self-blame) had lower AUCg when reporting loneliness. In other words, in the presence of adversity, the use of certain cognitive strategies to cope with stressors has a positive impact on diurnal cortisol. Our program sought to influence these very determinants: it aimed to teach coping strategies to reevaluate stressors through a psychoneuroendocrine lens. It also promoted strategies to enhance social support while providing emotional and instrumental support from the group modality. Although the effect size was small, the main effect found in diurnal cortisol concentration is promising. Few studies have used physiological measures to assess the efficacity of stress management interventions. At the 3-month follow-up, cortisol levels in the SMT group showed a rebound, with AUCg values returning closer to baseline levels. However, it’s important to note that despite this rebound, there was still a significant effect of the SMT intervention on AUCg values at the 3-month mark, although the effect size was smaller. This suggests that while the initial benefits of the intervention on cortisol regulation were not fully sustained over time, some residual effects remained. This finding aligns with other research indicating that short-term improvements in physiological measures can diminish if not continuously supported (Gaffey et al., 2016). The intervention’s decrease in effectiveness over time might be due to the absence of ongoing reinforcement or follow-up, as well as the challenges older adults face in maintaining stress management techniques long-term.

Given that few studies have used physiological measures to assess the efficacy of stress management interventions, our results contribute valuable insights into how such programs can impact cortisol regulation. The observed rebound in AUCg, along with the reduced but still present effect at 3 months, underscores the need for future research to explore ways to extend the benefits of stress management interventions and maintain their physiological impacts over longer periods.

Our study has several limitations. The first is the lack of proportionality in the recruitment in the specific settings chosen (i.e., high-end PRs, services for older caregivers, and day centers). The sample includes a higher proportion of people living in high-end PRs, which are not representative of most seniors in the general population: in Québec, 17% of older adults live in PRs; in our study, they represented about 75% (SCHL, 2021). This limits the generalizability of the results. However, this population is rarely studied because of its presumed socioeconomic privilege. Nevertheless, our results showed a significantly higher rate of distress in our sample compared to the rates found in other studies. The proportion of participants above the clinical threshold on the total HADS scale (score ≥ 16) in our sample was high: 43.9%. Several reasons may explain the observed levels of distress. Some participants were juggling multiple responsibilities, including caregiving and supporting family members. Others had just moved into their new homes because many residences opened just weeks before the study began. For some, the move was prompted by a serious health problem, a partner’s loss of autonomy, or other stressful situations. This suggests that the recruitment location was important in identifying at-risk individuals who could benefit from the intervention. Future research should examine other populations of older adults to determine the generalizability of the findings.

Second, although the attrition rate at the postintervention assessment and at the 3-month follow-up may seem high, it is consistent with what is typically seen in studies of older people testing the efficacy of a treatment. The attrition rate was not different between our two groups: 25.9 and 28.8%, respectively. A total of 62.9% of the sample responded to all three measures. Participants who dropped out by the time of the follow-up had significantly higher levels of combined anxiety and depressive symptoms at baseline. Although this is common in geriatric research, it can introduce significant bias into the results and affect the magnitude of the findings. In a systematic review of the literature on attrition among older adults in longitudinal studies, Chatfield et al. (2005) reported attrition rates ranging from 2 to 50% between measurement waves. In studies of older adults with chronic conditions, attrition rates as high as 30% are not uncommon (Chatfield et al., 2005; Gardette et al., 2007). Although we initially recruited 170 participants, which is generally an adequate number for a two-arm CRT, analysis of physiological data—in particular, cortisol—often requires even more participants. Therefore, we may have lacked statistical power.

Although this study has some limitations, it is important to note that the O’stress program is a brief preventive intervention program focused on stress and stress contagion that appears to increase coping strategies and decrease levels of anxiety and perceived stress. O’stress also appeared to reduce participants’ diurnal cortisol levels, which has not often been demonstrated in previous studies. In fact, our study is one of the few to demonstrate the efficacy of a stress management intervention in older adults on both subjective and objective measures of stress. However, some of these effects on coping and anxiety were not maintained over time, highlighting the importance of adding booster sessions. Booster sessions review the procedures and strategies learned during the main intervention and are used to increase the magnitude or persistence of the intervention’s benefits. Given that, similar to our findings, the effects of most resilience programs in older adults tend to diminish over time, it seems essential to plan booster sessions in adavance, especially for short interventions. These booster sessions would be beneficial because the intervention requires the development of new habits, which can be time consuming. Furthermore, aging is generally not a period of stability, as older adults often face frequent adaptive challenges, which may contribute to the limited long-term effectiveness of otherwise impactful treatments.

We should emphasize that our results suggest that this type of brief preventive program could reduce the levels of basal cortisol in older adults, which could represent a significant protective factor against health outcomes that has been related to HPA function. The program, which was offered in a preventive approach in community settings to all older adults interested, had significant effects on a nonclinical sample. A larger implementation of the program could be ensured by training psychosocial workers and even peers in the community. This approach is used for the dissemination of DeStress for Success©, with a training session offered twice a year, and thousands of kids and teenagers have been able to benefit from the program.

Increasing life expectancy challenges us to find solutions to ensure that these extra years are synonymous with health and well-being. Such interventions, which take little time to implement and appear promising, have the advantage of being scalable. For example, the DeStress for Success© program, developed for adolescents, has already been distributed to more than 3,000 practitioners and teachers for use with their students. We believe that our study provides sufficient evidence to warrant further research to improve the effectiveness of O’stress in different settings. In addition, we plan to adapt the stress contagion component to older adults’ formal and informal networks. We argue that improving resilience to stress in aging should become part of a universal prevention intervention and be embedded in every setting that provides services to older people.

This article is part of MJR’s thesis work.