The growth of the aging population might be the number one public health issue facing the world (1). The number of adults over the age of 65 will double between the years 2000 and 2030 according to the U.S. Administration on Aging, and, within the next 5 years, there will be more older adults than there will be children younger than age five for the first time in recorded history (2). In the developed world, the leading causes of mortality are nearly all age associated, marking a major shift in just one hundred years from infectious diseases as leading causes of death. Accordingly, a variety of global public health agencies have developed initiatives to promote successful aging (e.g., Centers for Disease Control’s Healthy Brain Initiative and the World Health Organization’s World Health Day focus on healthy aging), in order to reduce disparities in attaining maximum healthy life expectancy where evident. These initiatives seem to be in agreement that promotion of “successful aging” must focus more broadly than on mitigation and prevention of specific chronic diseases, targeting positive states of health and well being.
Yet, there remains little consensus about what factors comprise the essential ingredients of successful aging. As described by Rowe and Kahn three decades ago, much of the body of research on aging focuses on differentiating pathological aging from normal aging (3), and a far smaller body of research seeks to determine what differentiates older adults who seem to function better than that expected from those who follow the typical course of aging. Nonetheless, recent years have seen a surge in studies expressly and indirectly relevant to successful aging and its promotion. We review recent research on the definition, determinants, and interventions for successful aging in individuals with and without mental illnesses.
On the surface, psychiatry and successful aging would seem to have little to do with one another. By most definitions, the presence of neuropsychiatric illness would automatically preclude successful aging. However, we believe successful aging is relevant to psychiatry and to psychiatrists for several reasons. To date, phenotypic descriptions of successful aging have largely been centered on the absence of physical disabilities or systemic medical illnesses (4). As such, cognitive and emotional phenotypes are thus less well studied in the context of successful aging. However, cognitive and emotional aspects of successful aging are frequently described by older adults as central to subjective definitions of successful aging (5). In addition, there is reason to suspect that the impact of neuropsychiatric illnesses will outpace physical illnesses in accounting for total disability in coming years—for example, mental illnesses had a negative effect on gains in life expectancy in high income countries in the past decade (6). Among psychiatric disorders, there has been an increasing interest in more holistic conceptualizations of illness etiology, owing to the marked comorbidity with chronic medical problems. There has even been a suggestion that a number of chronic mental illnesses that psychiatrists routinely treat (e.g., schizophrenia, bipolar disorder, depression) are associated with courses indicative of “accelerated aging” (7, 8). Moreover, basic science points to biological mechanisms suspected as causes of accelerated aging that may bridge mental health diagnoses (e.g. oxidative stress); thus the literature on successful aging may provide guidance for novel strategies to prevent disability from chronic mental illnesses beyond the current armamentarium of psychiatric treatments. Finally, the great majority of modifiable aspects of successful aging, to date, involve volitional lifestyle choices and behaviors (9, 10). Psychiatrists treat illnesses that are associated with diminished participation in these lifestyle behaviors, and there is a growing literature on the positive impact of lifestyle-targeted interventions in mental health conditions (10). Psychiatry could thus play a key role in reducing the widening disparities in attaining healthy aging.
The most influential model of successful aging remains the one proposed by Rowe and Kahn (11, 12). This biopsychosocial model includes a set of three domains that formed the construct of successful aging: 1) freedom from disease or disability, 2) high cognitive and physical function, and 3) social and productive engagement. The MacArthur Network on Successful Aging subsequently operationalized these domains, developed a measurement approach, and recruited a sample of older adults aged 70–79 who met these criteria (13). Since the Rowe and Kahn model was introduced, a variety of studies have examined multidimensional operational definitions of successful aging in samples of older adults. Our review of the body of literature between 1987 and 2006 identified 28 studies that had reported a proportion of a sample that met criteria for successful aging (4). The great majority of these studies were cross-sectional, separating a group of participants who were “successful” from “unsuccessful”. The review revealed marked heterogeneity in definitions, with more definitions (N=29) than studies (N=28); accordingly, the rate of successful aging ranged from 0.4% to 96.0%. Beyond physical disability, which was a component in 26 of operational definitions, no other factor (e.g., life satisfaction, cognitive ability) appeared in more than half of the studies. Moreover, even among studies that were concordant in domains, the measures and criteria used to delineate “success” were highly variable. Therefore, the field has lacked consensus both in defining the essential ingredients of successful aging, as well as specifying its measurement. In addition, the field has focused somewhat narrowly on the cross-sectional differentiation of point-in-time estimates of successful aging.
In the past 5 years, a number of authors have attempted to refine the construct of successful aging. Fiocco and Yaffee (14, 15) conceptualized successful cognitive aging in terms of its trajectory over time, such that successful would involve minimal or lower than expected cumulative decline from prior cognitive functioning. The move toward longitudinal trajectory-focused definitions versus point-in-time valuation avoids the imprecision associated with intraindividual variability and opens greater opportunity for prevention. Another perspective portrays successful aging as on a continuum, rather than as discrete groups representing “successful” and “unsuccessful” aging, and one that should be linked mechanistically to biological systems (16). According to Lowry et al., (16), the approach to the definition of frailty provides a model in how consensus definition of successful aging might be attained; frailty involves multiple physiological systems and varied phenotypes, yet progress has been made in adopting consensus definition that links clinical observations to physiological processes involving depletions in energy metabolism (17). If such a consensus definition of successful aging could be achieved that links observed phenotypes with biological mechanisms, the potential for translational research on successful aging would be greatly advanced. Vahia et al. (18) used a multidimensional analysis called a Bayesian belief net to examine putative causal directions among domains, acknowledging that the separate domains included in common definitions of successful aging influence one another and that a more parsimonious approach might be to understand “upstream” and “downstream” effects among domains. Thus, the concept of successful aging has evolved from a point-in-time “diagnosis” to one that is more dimensional and interdependent, mindful of changes over time, and linked to biological mechanisms.
Moreover, there has been a parallel consensus process to boost the standardization and consistency of cognitive and emotional constructs included in successful aging. The NIH Cognitive and Emotional Health Project (CEHP), through consensus among experts, produced a summary of extant evidence on the promotion of cognitive and emotional health. Subsequent efforts by CEHP have been to generate a “toolbox” of standardized and vetted measures that researchers could use to facilitate cross-compatibility among unique samples (19). Notably, the measurement framework for the CEHP includes domains that extend beyond traditional cognitive and emotional phenotypes to include positive emotion, sense of purpose, and self-efficacy.
Another trend that has emerged in considering the definition of successful aging has been the clear divergence between objective definitions and subjective definitions. It is notable that in most of the multicomponent definitions described above, a minority of older adults would be considered to be successfully aging. A recent population-based sample of older adults in the United States revealed that just 11.9% of adults 65 years and older, who were participants in the Health and Retirement Survey, met Rowe Kahn criteria for successful aging (20). In parallel, a number of studies have examined older adults perspectives on both whether or not they believe that they are aging well, and what domains they believe are the essential components of successful aging. A consistent finding across several studies is that the great majority of older adults believe they are successfully aging. Using a 10-point scale (with 10 representing “most successful”) Jeste et al. found that the great majority of 1006 randomly selected men and women in the San Diego area rated themselves an 8, 9, or 10 (21). It has been notable that, despite this upward bias, self-ratings are strongly and positively correlated with objective indicators of physical, psychological, and social health (22), and so the divergence of subjective and objective definitions of successful aging remains somewhat of a paradox. One window into this paradox is qualitative studies, employing focus group and individual interview methods that reveal that older adults’ personal definitions of successful aging frequently include phenotypes that are often not included in objective definitions (5, 23, 24). In particular, older adults are more likely to describe subjective well being and positive psychological states (maintained in the midst of physical disabilities and other losses), and are less likely to define successful aging in regard to freedom from disability, which is included in nearly all definitions of successful aging. Accordingly, Pruchno et al. proposed a two-factor model of successful aging, based on confirmatory factor analysis, that separates subjective and objective components (25). Thus, although successful aging awaits consensus definition, recent approaches have incorporated person-centered perspectives in successful aging models.
In parallel to the focus on separating successful from normal or expected aging, there have been a handful of studies that have applied multidimensional models to chronic illness populations, including mental illnesses. Some individuals with chronic psychiatric illnesses seem to exemplify recovery and “success” despite diagnoses such as schizophrenia, and public attention has been paid to individual cases, such as Nobel Prize winner John Nash, who accomplish ground-breaking works in the context of severe psychiatric symptoms (26). The goal of research in this area would be to understand which factors predict successful adaptation to illness, and to identify how some patients seem to avoid morbidity in the presence of established risk factors. In many ways research on successful aging parallels the process and challenges of attempts to define “recovery” (27)—both are positive constructs that have received great attention, yet each construct has struggled with attaining definitional consensus. As well, consumer models of recovery are somewhat divergent from scientist-derived definitions.
The study of successful aging in the context of mental health problems is nascent. Among older adults with schizophrenia, Auslander and Jeste applied criteria for sustained remission from schizophrenia and found that 10% of older adults with schizophrenia have experienced remission of symptoms from 5 years (28). Later study found that 2% of older adults with schizophrenia met Rowe and Kahn criteria for successful aging, yet when less stringent criteria are applied approximately 40% of patients may experience recovery (29). The perspectives of older adults with schizophrenia provide additional insight: a qualitative study of 32 older adults with schizophrenia revealed that most patients felt as though their symptoms had improved over time, which they attributed to better self-management skills developed over time (e.g., questioning validity of hallucinations) and greater acceptance of treatment. On balance, a much smaller subset felt as though their lives had improved over time and felt that they were actively engaged in recovery, others voiced resignation that they had missed opportunities to attain functional goals (30). As such, future conceptualizations of successful aging in schizophrenia would need to consider trajectories of both symptoms and functioning over the lifespan (31).
Another direction in applying successful aging concepts to psychiatric conditions is the study of people with Alzheimer’s-associated pathology yet normal cognitive functioning. In a postmortem study, Codispoti et al. (32) examined brains of participants in the Baltimore Longitudinal Study of Aging with significant tau and amyloid deposits, yet who did not evidence cognitive deficits in the years preceding death. This group was compared with participants who were cognitively impaired prior to death, matched on brain pathology. Findings suggested that older adults who managed to avoid cognitive impairment until death had prior evidence of alterations in cerebral blood flow that were suggestive of greater adaptive responses than did the cognitively impaired group. Thus, studying the processes by which patients adapt to brain illnesses may be fruitful.
Another group of patients who raise interesting questions about success are patients with HIV/AIDS. A subset of patients seems to avoid neurocognitive and disease morbidity associated with this illness and function well into middle age. Malaspina (33) found that 32% of middle-aged and older people with HIV met mulitidimensional criteria for successful cognitive aging—the predictors of membership in this group were lower depressive symptoms and better relationships with treatment providers. Notably CD4 counts were not related to membership in the category of successful aging, and so “success” was not simply a function of having less severe HIV-associated illness. Thus, although the application of successful aging to illnesses encountered by psychiatric providers is at preliminary stages, there appears to be some traction in understanding the characteristics of patients who appear to functional well despite illnesses.
Although there have been hundreds of studies on the genetic predictors of the lifespan, far fewer have examined genes as predictors of the “healthspan” or multidimensional models of successful aging. The four approaches that have been applied are twin studies, genome wide association studies, candidate genes studies, and gene expression studies. Glatt et al. (34) reviewed the literature in each of these types of studies and found 29 studies that examined genes and a multidimensional definition of successful aging that included phenotypes other than longevity. Twin studies revealed that successful aging phenotypes were moderately heritable, accounting for approximately the same proportion of variation in successful aging traits as in longevity (approximately 30%), providing some confirmation that further pursuit of genetic association studies is warranted. Genetic association studies frequently contrasted older adults at the extreme end of the age spectrum (e.g., >90 years old) and contrast frequency of single nucleotide polymorphisms with that of younger unselected samples. In Glatt et al. (34), six SNPs emerged as higher in frequency among classified successfully aging older adults: APOE, GSTT1, IL6, IL10, PON, and SIRT3. The frequency of SNPs in relation to subjective successful aging has also been studied, as O’Hara (35) found a relationship between the 5-HTTLPR s allele and worse self-rated successful aging. However, these studies reveal little about the mechanisms of genes on successful aging phenotypes, nor gene × environment interactions. Eaton et al. (36) advocated for integration of work on personality traits presumed to be stable over the lifespan with the twin study approach, in order to better understand the influence of genetic and epigenetic predictors of positive states of health in older age.
It remains somewhat of a mystery why aging occurs at all. There are over 40 theories of aging, with explanations of the aging process that range from genetic mutations (e.g., DNA damage due to free radicals), cellular senescence (e.g., telomere theory, oxidative stress), and physiological systems (e.g., dysregulation of the immune system, metabolic changes) (37). These theories each have the challenge of accounting for the myriad age-associated changes within the skin, outside of the skin, and their interaction over time, in addition to specifying whether processes are causal and which biomarkers are reliable and valid indicators of the theory. Perhaps the most excitement in recent years has come from research on telomeres, the region of the chromosome that protects the end of the chromosome from damage. Pioneering research suggests that the length of the telomere provides an indication of cellular aging (38), and that people who are subject to chronic stress have shorter telomeres and evidence of advanced cellular aging (39). Atzmon et al. (40) examined telomere length in centenarians and found that greater telomere length, even among people whose histories of robust health made them likely to attain such advanced age, was predictive of better cognitive ability, lower likelihood of diabetes, and greater physical functioning. The observation that telomere length is associated broadly across disparate phenotypes in aging suggests that it represents a potentially fundamental marker of aging (although it is notable that recent study did not find an association between telomere length and optimism or self-rated health (41)). It is also notable that telomere deterioration has been found in depression (42), schizophrenia (43), and people with extreme childhood adversity (44), suggesting a mechanistic link between these disorders and observations of “accelerated aging” across psychiatric diagnoses. In a pilot study in depression, telomere length was associated with markers of inflammation and oxidative stress, as well as a more chronic course (45). There remain many questions surrounding telomere biology (46), yet research in this area links fundamental biological aging process with the negative impacts of stress, which seems to undergird several common psychiatric conditions.
In addition to biomarkers that bridge multiple aging phenotypes, there may be fundamental mechanisms by which interventions reduce age-associated morbidity and mortality. Ironically, the science again points to stress, yet as a positive influence. Mattson (47) has described the process of “hormesis,” in which physiological responses to mild stressors increases protective and strengthening effects on the organism. Physical activity, cognitive stimulation, and caloric/dietary restriction involve quite different behaviors, yet each are mild stressors that may produce similar adaptive responses that promote neurogenesis and cellular protection. As with telomeres, biological processes may account for similar protective effects observed across different interventions.
Cognitive reserve is the ability to maintain performance in the presence of brain deterioration. A recent meta-analysis appeared to confirm the long-held notion that people with higher educational attainment experience reduced risk of dementia, and that, once diagnosed, experience a more rapid rate of decline (48). However, there are notable exceptions to the literature on education effects (46).There is also related evidence that reduced risk of dementia is also associated with lifetime engagement in complex cognitively demanding occupations (49) and activities (50) that confer cognitive stimulation. Conversely, there is some evidence that participation in cognitively “sedentary” activities, such as television, increases risk of cognitive decline (51).
Of great interest is the understanding of the mechanisms of cognitive reserve, which might occur by carrying forward structural advantages developed earlier in life (52, 53) (brain reserve) or by functional changes in brain networks that enable maintenance of performance (cognitive reserve). Additionally, within cognitive reserve, functional changes may involve compensation, which could be observed by greater brain activation (i.e., recruitment of other brain regions to maintain task performance) or increased efficiency of the same brain networks, which would lead to observations of reduced activation. Studies have yet to fully resolve whether compensation, efficiency, or structural explanations fully account for the capacity for some individuals to maintain high performance in the presence of typical age-associated brain deterioration. Kaup et al. (53) found that 83% of 35 studies in older adults found that bigger brain structures were associated with better cognitive performance. Eyler at al. (54) reviewed 80 functional magnetic resonance imaging studies and found that 70% of studies showed greater activation as related to better performance and that effects were stronger in older adults and were most prominent in the frontal cortex. Steffener et al. (55) used path modeling to indicate that both compensation and efficiency hypotheses could be supported. Although a long way from resolution, the observation that older adults retain neuroplasticity into older age provides for enthusiasm that interventions initiated in older age may produce cognitive protection (described later in this paper).
A related field of study has sought to explore the neurobiological underpinnings of “wisdom,” a construct that is historically associated with improvement with aging. Meeks and Jeste (56) provided a theoretical synthesis that integrates research on the components of wisdom: 1) prosocial behavior and attitudes, 2) pragmatic knowledge and effective social decision making, 3) emotional homeostasis, 4) reflectiveness and self-understanding, 5) value relativism and tolerance, and 6) toleration of ambiguity. Each of these subdomains contains abilities traditionally associated with cognitive aging as well as aspects of emotion regulation and social cognition; accordingly, the neural basis of the model involves brain circuitry and top-down regulation of frontal (e.g., prefrontal cortex) and limbic structures. Although speculative, this model provides a foundation for studying the neurobiology of gains observed in aging as well as avoidance of losses.
There have been a number of studies that have associated positive psychological traits (e.g., optimism, sense of purpose, life satisfaction, self-efficacy, personal control) and states (e.g. happiness) and reduced mortality in healthy samples of older adults (57). What sets these predictors apart from other determinants of successful aging is that, on average, aging is positively associated with levels of these psychological constructs. Stone et al. (58) reported data from 340,847 Americans and found that after young adulthood, age was positively associated with happiness and interest, and, across the lifespan, negatively associated with feeling stressed, worried, or angry. Branchflower and Oswald (59) harmonized multinational surveys on life satisfaction and appeared to confirm the existence of a “u-shaped” curve, with young adults and older adults experiencing highest levels of life satisfaction, and middle-aged adults reporting lower life satisfaction.
One of the primary hypotheses as to why these positive effects exist is that older adults appear more adept at adaptively managing emotions than younger or middle-aged adults. Older adults are more likely to be able to experience positive and negative emotions simultaneously, to mitigate the lasting impact of negative experience, and to selectively attend to positive information (60). A recent study by Brassen et al. (61) examined a sequential risk-taking task in younger adults, depressed older adults, and putatively healthy older adults; this task was designed to induce regret. Using a multimodal approach integrating fMRI and autonomic response, the authors found that healthy older adults engaged in adaptive responses to regret after choices that led to failure, whereas depressed older adults and younger adults did not.
Such affective neuroscience approaches dovetail with recent clinical studies on resilience in older adults. Lamond et al. (62) found that self-rated resilience, as rated on the Connor Davidson Resilience Scale, was composed of two primary factors in older women: one which involved personal control and active problem-solving, and another that involved tolerance of negative affect and stressors. The latter factor was more strongly associated with indicators of emotional health, consistent with basic tenets of lifespan developmental theories of aging involving adaptive management of emotions.
Understanding the determinants of successful aging should pave the way for interventions to impact these determinants. In the next section, we consider the current evidence for interventions that are relevant to the promotion of successful aging, focusing on cognitive and emotional aging. In particular, we focus on interventions that target the improvement of functioning in older adult samples who were not selected for specific illnesses.
Physical activity and exercise
Recent evidence indicates substantial benefits of physical activity that extend across physical, cognitive, and emotional domains in later life. Current recommendations from the Center for Disease Control and Prevention (CDC) suggest that individuals 65 years of age and older get 150 minutes of moderate-intensity or 75 minutes of vigorous-intensity aerobic activity each week and two or more days of muscle-strengthening activities (http://www.cdc.gov/physicalactivity/everyone/guidelines/olderadults.html). Physical fitness, including aerobic power, flexibility, and muscle strength, decline with age (63, 64). Older adults are the most sedentary segment of the population, with 28% to 35% of older adults reporting no leisure-time physical activity (http://www.ahrq.gov/ppip/activity.htm).
Longitudinal studies indicate a relationship between physical fitness and cognitive ability (65, 66). Physical fitness appears to provide some level of protection from cognitive impairments, possibly by offsetting age-related structural changes in the brain as people age (67). In a meta-analysis of 18 physical activity intervention studies with human subjects published between 1966 and 2001, older individuals who were randomized to training interventions that included either aerobic (cardiovascular) training or combination training (cardiovascular and strength) scored higher on measures of executive function, cognitive control, and spatial and processing speed tasks by about a half a standard deviation (65).
There are a number of theoretical mechanisms for the influence of physical activity on cognitive function. Aerobic fitness is associated with increased expression of brain-derived neurotrophic factor (BDNF) (68) as well as N-acetylasparatate (NAA) in the frontal cortex (69). NAA levels decline with age and are associated with poorer working memory performance; higher levels of aerobic fitness may offset this age-associated decline influencing working memory capacity (69). In addition, physical exercise may reduce oxidative stress (70) and increase supplementary growth factors that support angiogenesis, neurogenesis, and synaptic plasticity (71). In a review of human intervention studies Kramer and Erikson (72) proposed that physical fitness produces dynamic changes in patterns of event-related brain potentials (ERPs) in older adults, mirroring those of younger adults, particularly increasing P3 amplitude and decreasing P3 latency.
Physical activity may also have antidepressant effects in older adults, as seen in a number of studies (73, 74) showing that increases in exercise training significantly improves depressive symptoms in older adults (75). In one study of the National Health and Nutrition Examination Survey (NHANES) 2005–2006, adults ages 20 and older who participated in moderate levels of physical activity, for 30 minutes a day/three or more days a week, were 28% less likely to show signs of depression than individuals who do not participate in any type or low levels of physical activity (76). Stathopoulou et al. (77) reported a meta-analysis of 11 physical activity treatment outcomes studies with individuals with depression, a small subset of studies reviewed were conducted with older adults. The authors found that high intensity exercise had greater improvement in treating depression, improving quality of life and sleep quality; aerobic and anaerobic exercise appeared to yield the same gains in all studies reviewed. Interestingly the analysis suggested that exercise-alone showed lower relapse rates than combination (exercise and pharmacotherapy) (77). The mechanisms of physical activity on emotional health are unclear, but physical activity may mitigate the harmful effects of prolonged stress on cellular, neural, immunologic, and emotional health (68). Regular physical activity may augment endogenous antioxidant systems, which, in turn, decreases oxidative stress (78). In addition, research suggests that endogenous opioids (i.e. endorphins, dynorphins), which are increased during and directly following exercise, contain biochemical characteristics that have been associated with mood regulation (79).
A common question practitioners face is which treatment models best motivate exercise? In a review of 38 randomized controlled trials (RCTs) of physical activity that included a total of 57 interventions, three types of interventions were compared: home-based, group-based, and educational (80). In home-based physical activity interventions, participants performed their prescribed exercise routine in their home; group-based physical activity interventions included supervised group-based exercise; while educational physical activity interventions included an education piece with information on exercise and health as well as encouragement to participate in physical activity. Results from this review suggested that while home-based and group-based interventions had the best outcomes, sustainability remains a problem (80).
Technology provides for evolving means of motivating exercise, which may increase access and motivation to exercise. Exergames combine computer-simulated environments and physical activity in the context of pleasant activity. In a pilot study, 12 weeks of exergames participation was associated with significant reductions in subsyndromal depressive symptoms and improvements in health-related quality of life, and cognitive performance (74). Van Stralen et al. (81) developed a computerized intervention for physical activity that focused on the environmental issues that are often barriers to physical activity in older adults. Intervention strategies included targeting motivation, premotivation, and postmotivation, as well as environmental issues. Results suggested that targeting environment increased participants’ physical activity by one hour per week.
Nutrition and dietary interventions
The U.S. National Health and Nutrition Examination Study (NHANES), a national level cross-sectional study of 11,145 participants ages 18 years and older conducted between 2007 and 2010, found that older adults ages 60 and older had higher levels of abdominal obesity and hypertension than any of the younger age groups, with the highest levels being in the 80 and older group (82). Nutrition plays a significant role in prolonged health and health promotion throughout the life course. Owing to relationships identified between nutritional intake and mild cognitive impairment (83, 84), depression (85, 86), physical deterioration (87), and quality of life (88), promotion of successful aging in older adults should include healthy dietary practices (89).
In a study of unique nutrition interventions from 1950 to 2008, 15 high quality randomized control trials were found: 10 nutrition education and counseling interventions and five nutrition supplement interventions (88). Within this review nutrition counseling interventions that incorporated active participation demonstrated the best nutrition-related outcomes. These interventions invited participants to develop personalized health plans, set goals, and develop higher levels of self-efficacy. In an older review of nutrition interventions and education for older adults, 25 studies met the criteria for inclusion (89). Studies were more successful if they included simple, practical nutrition messages, provided education to increase motivation and nutrition knowledge, offered incentives for attendance and change, and when behavioral interventions produced readiness to change and self actualized goals.
Calorie restriction (CR) or dietary restriction, a 10%–40% reduction in nutritional energy intake, is frequently considered an avenue for extending lifespan and health span (90, 91). Calorie restriction has been investigated in a number of species, including rodents, rhesus monkeys, and humans (90, 92). There is some controversy as to how much calorie restriction is necessary and the magnitude of the effect, as Mendelsohn and Larrick (91) discussed the impact of confounds such physical activity and muscular fatty acid metabolism as influencing variation of in outcomes. The mechanisms of calorie restriction are of course of great interest, in particularly whether it is affecting the rate of aging or postponing age-related disease (91). Calorie restriction may indeed reduce risk for cardiovascular disease, cancer, and diabetes in humans (92). For example, in a recent review of studies conducted with various animal models, it appears that calorie restriction is associated with a lower risk of cancer due to a antitumor effect associated with a decrease in anabolic hormones, oxidative markers, inflammatory cytokines (93).
Dietary patterns also have influence on the likelihood of successful aging. The Mediterranean diet (87, 94) is characterized by the consumption of plant foods, fresh fruits, olive oil, fish, along with low amounts of red meat, a low to moderate intake of dairy, and wine consumed with meals (94, 95). The Mediterranean diet has been associated with decreased mortality, a reduction in cardiovascular and chronic diseases, as well as a reduction in cognitive decline, and risk of Alzheimer’s disease (AD) (95, 96). The Health, Aging, and Body Composition Study (Health ABC) that used a sample of 3,075 older adults aged 70 to 79 indicated that older adults with highest conformance to the Mediterranean diet (87) evidence less decline in mobility over eight years. In a sample of 1,410 older adults, Feart et al. (97)found that while adherence to the Mediterranean diet was associated with a slower decline on MMSE scores, it was not significantly related to performance on other cognitive tests. In another study, 1,393 cognitively normal participants were followed until they were diagnosed with mild cognitive impairment (MCI) or until last follow up 4.5 (± 2.7) years (84). Higher adherence to the Mediterranean diet was associated with a lower risk of MCI; compared with those in the lowest tertile of adherence, participants in the middle tertile of adherence had 17% less risk for MCI and participants the highest tertile had 28% less risk of MCI (84). It is unclear what the “active ingredients” in the Mediterranean diet, yet suspected ones are Vitamin B12 and folate, antioxidants, monounsaturated fatty acids, and moderate amounts of alcohol (87, 96, 97).
Vitamin D deficiency (83) appears to predict decreases in cognitive functioning and the development of Alzheimer’s disease. In a study of 498 older adult women who did not take vitamin D supplements, women who developed Alzheimer’s disease during the following 7 years had the lowest baseline vitamin D intakes compared with those who did not develop dementia (83). Vitamin D supplementation has been shown to reduce postural sway, increase lower extremity strength gain, and decrease the time it takes to complete the Timed Up and Go test in a review of 13 articles on vitamin D supplementation in older adults (98). In a large study (N=81,189) of postmenopausal women ages 50–79, vitamin D supplementation was also associated with a 20% reduced risk of depression from baseline to 3-year follow-up (99).
Finally, the supplement Resveratrol has also been promoted as a preventive agent for several diseases and potential therapy (100). While tests on laboratory animals have suggested antiaging, anticarcinogenic, anti-inflammatory, and antioxidant properties associated with Resveratrol, only recently have human clinical trials been conducted showing matched potential to prevent chronic disease and improve health in humans (100). Further research needs to be conducted to increase knowledge about the totality of benefits of this supplement as well as other supplements whose claims of positive effects are often not subject to empirical study or governmental oversight (101).
There is, of course marked interest in interventions that might slow the rate of cognitive decline and reduce risk of dementia (70). There are a variety of approaches that are aimed at enhancing cognitive ability, including traditional classroom-based cognitive training, computerized cognitive training, and cognitive stimulation. In a review of RCTs of cognitive interventions published between August 2007 and February 2012 that involved guided practice of cognitive tasks, cognitive training appeared to be beneficial for numerous facets of objectively measured cognitive functioning, including fluid intelligence, attention, processing speed, executive functioning, and memory performance (102). For example, the Advanced Cognitive Training in Active and Vital Elderly (ACTIVE) study included 2,168 older adults with good functional and cognitive status randomized across four treatment groups, memory, reasoning, speed of processing, and control (103). Five years postintervention all three groups retained a significant effect on cognitive outcomes, although, as is typical the effects were restricted to the targeted outcome (e.g., memory training only significantly improved memory) and effects on distal outcomes (e.g., functioning) were weaker than that associated with cognitive abilities. Nonetheless, at 5-year follow up, the participants in the reasoning group had significantly less difficulties with independent activities of daily living.
One question is whether training should target multiple versus single domains. A sample of 270 community-dwelling older adults, 65 to 75 years old, who evidenced high functional capacity and no reported cognitive impairment, were enrolled into one of three treatment groups (multiple domain, single domain, or control group) (104). Single-domain training centered around reasoning training such as numerical reasoning and verbal reasoning, while multiple-domain training focused on reasoning, visuospatial map reading skill development, memory, problem solving strategies, handcraft making, and health and physical exercise. Both single- and multiple-domain training showed significant training effects on visual reasoning; single-domain training demonstrated significant effects on work interference and visuospatial/constructional score, and multiple-domain training showed effects on immediate and delayed memory. Multiple-domain training was more effective in follow-up analyses, suggesting it may be more advantageous for long-term effects on cognitive decline when compared with single-domain training.
Another question is whether computerized interventions can produce similar effects, which, in light of their portability and ease of administration, could provide for greater reach of interventions. A recent review found that computer-based cognitive interventions showed comparable results to traditional cognitive interventions (105). Online video games have been described as an intervention for improving cognitive control in older adults (105). In a study of the impact of video games on cognitive control, 54 older adults were randomized to play strategy games for 30 minutes per day for 7 weeks, while a control group answered test questions about documentaries online each day (106). Results suggested that using online cognitive training games can increase inhibition and inductive reasoning, while selective attention was significantly increased in the documentary group as measured by the useful field of view test. A recent strategy has been combining cognitive stimulation with physical activity. For example, a clinical trial indicated that exergame cyber cycling for 3 months was associated with greater gains in executive function than participants who spent comparable time exercising on a stationary bike, indicating that there may be an incremental benefit of interactive decision making and virtual reality imagery in exergames (107). While many questions exist about the optimal approach to cognitive remediation and its combination with other kinds of interventions, recent work has shown substantial promise (106).
Social isolation or loneliness is an important and prevalent health concern in the growing population of community-dwelling older adults (see Nicholson  or Routasalo et al. ). In a RCT of psychosocial group rehabilitation for older adults ages 75 and older, participants were randomized to one of three groups as an intervention participant or control: therapeutic writing and group psychotherapy, exercise and health related discussions, and art and inspiring activities (see Pitkala  or Routasalo et al., 2008 ). Individuals in the intervention groups met once a week for 3 months, a total of 12 times, for 5 to 6 hours including breaks and lunch. Individuals in the intervention groups showed improved health compared with the individuals in the control groups at the 1-year follow-up, as well as reduced rates of mortality and reduced cost of health services (110). In addition, intervention group participants showed improved psychological well-being scores, and a larger number had acquired new friends at the 1-year follow up compared with control group participants (109). While the above investigation shows promising results for group interventions that target social isolation, unfortunately at this time interventions for social isolation are still underdeveloped and not well researched (108).
Volunteering is another intervention for social isolation that is currently understudied. Volunteering offers extensive benefits to the community, as well as enhanced well being and positive health and psychological outcomes to older volunteers (111). The Baltimore Experience Corps Trial (BECT) (112) involves pairing older adults with academically at-risk youth, and found that volunteering in this way, in addition to benefitting younger adults, was associated with better cognitive ability in older adults who volunteered compared with a control group. One of the main premises of this trial is that older adults are more likely to want to volunteer their time when the aim of the experience is to contribute to younger generations and to leave a legacy, versus focusing on personal benefits of volunteering on physical and cognitive health and social activity. In a study of community-dwelling older adults in Australia, individuals who engaged in moderate and high levels of volunteering (1 to 7 hours per week) showed higher levels of life satisfaction and positive effect than nonvolunteers (112).
Positive psychological interventions
Positive psychological interventions have been examined as a route to increasing happiness and well-being (113) by strengthening the expression of positive emotions and character traits (114, 115). A new focus has been on positive aging approaches designed to assist older adults with coping and responding to age-related transitions with flexibility (116). Positive psychological approaches include interventions that focus on characteristics such as gratitude, forgiveness, or altruism. While there has been little research on such interventions with older adults specifically, positive psychology interventions have been readily implemented and tested in wide age-range populations (see for example Mongrain and Anselmo-Matthews ). In a meta-analysis of 51 positive psychology interventions with a total of 4,266 individuals, Sin and Lyubomirsky (115) found that positive psychology interventions appear to decrease depressive symptoms and significantly enhance well-being and may be sustained by individuals over time. For example, a 1-year follow up of a clinical trial of loving-kindness meditation as a method for increasing positive emotions, one third of participants continued to use this meditation technique on their own (117).
While there may not be consensus in the definition of successful aging, it is agreed upon that this construct extends beyond living a long life and avoiding physical disability. Psychiatry can and should play a key role in defining the research and promotion of successful aging, in light of the influence of cognitive and emotional health on outcome trajectories into later life. Moreover, there seems to be significant overlap in the fundamental biology of aging and the mechanisms that underlie the course of major mental illnesses. Research on the subset of older adults with histories of or substantial risk factors for mental illnesses, yet who manage to avoid or adapt to morbidity, is one way in which psychiatry research can be integrated with work on successful aging. Much of the literature on interventions to promote successful aging has focused on modifications to lifestyle behaviors. It is notable that mental illnesses are a risk factor for diminished practice of healthy lifestyles, and lifestyle modification presents an important approach to augmenting standard psychiatric care. Although the task of altering the rate of aging is formidable and infrequently, if ever, met, a variety of interventions improve cognitive, emotional, and social functioning in normative samples of older adults. Many of these interventions are or could be adapted to increase the likelihood of successful aging in older adults with mental health diagnoses. Given that the absolute number of older people with mental illnesses will triple by the year 2030, the importance of psychiatry to successful aging has never been greater.