Associations of Ageism and Health: A Systematic Review of Quantitative Observational Studies

Abstract

This systematic review synthesizes observational studies on the relationship between ageism and health. We searched 10 electronic databases and included 67 articles. The operationalization of ageism in these studies can be classified into three constructs: age stereotype, self-perceptions of aging, and age discrimination. Most ageism measures were used within a single study, and many lacked information about psychometric properties. Seven health domains—disease, mortality, physical/functional health, mental health, cognitive function, quality of life, and health behavior—have been used as outcomes. Evidence supports a significant association between ageism and health, particularly between self-perceptions of aging and health. Nine studies report moderators, which helps to identify those more vulnerable to negative effects of ageism and inform the development of interventions. The review suggests that the literature has examined limited dimensions of ageism, and that developing valid and reliable instruments for ageism-related concepts is a priority.

Keywords

age stereotypes self-perceptions of aging age discrimination

Introduction

Persons age 60 and over will, estimates indicate, triple in number between 2000 and 2050, increasing from 600 million to 2 billion worldwide (United Nations, 2015). Societies must be prepared to accommodate unprecedented growth in older adult populations. Ageism, coined by Robert Butler who defined it as “a process of systematic stereotyping and discrimination against people because they are old, just as racism and sexism accomplish this for color and gender” (Butler, 1975, p. 12), is a major hazard to older adults’ social integration and quality of life. There is no standard agreement about the age when someone becomes old. Being perceived as old, by self, others, organizations, and institutions, is probably more important than chronological age in making an individual a target of ageism. From a legal perspective, the United States’ Age Discrimination in Employment Act of 1967 protects workers age 40 and older from discrimination based on age.

Academic interest in ageism has grown in recent decades, refining the concept’s complexity and nuance. Based on a literature review of ageism definitions, Iversen and colleagues (2009) suggest that ageism should be considered a multi-dimensional concept that encompasses cognitive (stereotype), affective (prejudice), and behavioral (discrimination) components; positive and negative dimensions; conscious (explicit) and subconscious (implicit) aspects; and micro- (individual), meso- (social group), and macro- (institutional and cultural) levels.

One of the most investigated topics related to ageism is its relationship with health, and experimental studies conducted in laboratory settings have taken the lead. Several systematic reviews and meta-analyses of these experimental studies find that negative age stereotypes adversely affect older people’s physical, physiological, and cognitive performance (Horton et al., 2008; Lamont et al., 2015; Meisner, 2012), and Lamont et al. (2015) suggest that the effects of age stereotypes are stronger on cognitive performance than on other performance domains. Experimental studies are limited as they observe human behaviors outside the real-life context. Moreover, most experimental studies have examined only the cognitive dimension of ageism (age stereotypes) and its short-term effects on behavioral performance. Obviously, some health outcomes (e.g., mortality) can only be examined in observational studies.

Over the past 2 decades, an increasing number of observational studies have investigated the relationship between ageism and health. Observational studies are those in which researchers observe the association between ageism and health in a sample or population without trying to control who is or is not exposed to ageism. Warmoth et al. (2016) conducted a systematic review of observational studies that have examined the association between older people’s perceptions of aging and health. Their review included 28 studies, most of which were cross-sectional, and a majority of which (16 of 28) were judged to be of low quality. The review find that perceptions of aging have been measured in a variety of ways, and that aging perceptions are associated with health across multiple domains. One limitation of Warmoth and colleagues’ (2016) review is that it covers only a subset of the literature—studies that have examined associations of aging perceptions and health. Studies that have examined other components of ageism, such as age discrimination, were not included. In addition, their systematic review included studies published prior to March 2014. Relevant studies published after that point have not been systematically reviewed. Given the recent growth of observational studies that have examined the association of ageism and health and that few studies have reviewed this literature specifically, we conduct this systematic review to provide a comprehensive summary of extant research in this area and to identify gaps and directions for future research.

In addition, it has been suggested that moderators—variables that influence the nature/strength of the ageism-health association—play a role in understanding the health consequences of ageism (Barber & Mather, 2014; Levy, 2009). For example, Levy et al. (2012) report that negative age stereotypes are associated with memory decline over time, and the decline is greater among those for whom the age stereotypes are self-relevant (identification with an old age group) than among those for whom these stereotypes are not self-relevant. The literature on discrimination suggests that social support, coping behaviors, and group identification moderate the link between perceived discrimination and health (Pascoe & Smart Richman, 2009). From a public health perspective, understanding moderators help to identify people more vulnerable to negative health effects of ageism and inform the development of targeted interventions.

In sum, we aim to answer the following questions in this review of observational studies on ageism and health: (a) How has ageism been operationalized and what health outcomes have been examined? (b) What is the association between ageism and health? (c) What factors moderate the ageism-health association?

Methods

Search Strategy

This systematic review was conducted following PRISMA guidelines (Liberati et al., 2009). Ten major electronic databases, including CINAHL, PsycINFO, PubMed and Cochrane Library, were searched. We limited the search to peer-reviewed journal articles written in English and published between January 2000 and June 2019. We chose to begin at 2000, as a prior review suggested that ageism studies before 2000 were scant (Warmoth et al., 2016). In database searching, keywords age, aging and ageing were included with each of the ageism-related keywords (ageism, attitude, belief, expectation, perception, discrimination, stereotype, image, and prejudice), and each health keywords (health, physical health, functional health, mortality, disease, disability, subjective well-being, psychological well-being, quality of life, longevity, mental health, emotional health, depression, dementia, cognitive function, cognitive impairment, memory, health behavior, health care use, preventive care, and hospitalization.)

Only articles that met the following criteria were included in this review: (a) quantitative empirical observational studies; (b) studying ageism targeting people age 40 and older;¹ and (c) using ageism-related variables as independent variables and health-related variables as outcomes. Ageism-related variables were those related to the cognitive, affective, or behavioral components of ageism. Review-type articles, such as systematic reviews and meta-analyses, were excluded.

The study management and screening process was managed in Covidence, a Cochrane-recommended tool for systematic reviews. The initial search retrieved 4,987 journal articles (Figure 1). After removing duplications, a total of 2,824 articles remained for initial screening. Two authors screened all titles and abstracts to identify articles satisfying the above-mentioned criteria, and their agreement rate was 98%. After excluding 2,752 that did not pass the initial screening, we mined the citations of the rest and the review-type articles mentioned in the introduction, resulting in an additional 39 articles. The round of abstract screening and hand search yielded 111 articles. The full texts of the 111 articles were then reviewed for eligibility by two authors. Any articles in doubt were discussed by all authors. The full-text assessment yielded a total of 67 articles (full citations are provided in Supplement A.) Figure 1 shows the flow of study screening.

Figure 1.

Systematic review flowchart. Note. Each column represents the number of times a statistically significant association between an ageism construct (e.g., age stereotype) and a health domain (e.g., disease) was reported, included those reported by good quality (darker shade) and fair/poor quality (lighter shade) studies. No bar is shown if the category has zero studies. For example, one good quality and three fair/poor quality studies reported significant associations between age stereotype and cognitive function. Columns with an alphabetical letter on top had insignificant findings reported. The percentage of non-significant findings, out of all the times that the association was examined, for a = 50%, b = 16%, c = 14%, and d = 9%. Columns without any letter on top mean that the association was reported to be statistically significant 100% of the times. For example, out of all the times that the association between self-perception of aging and physical/functional health was examined, 84% were statistically significant (n = 16).

Data Analysis and Synthesis

Two authors independently extracted the following information from each study using a pre-defined template: date of publication, research question/purpose, country of participants, sampling method, sample size and characteristics (age, gender, race, and education), study design, independent variables and measures, health outcomes and measures, moderators, and major findings. Discrepancies between the two reviewers were resolved through consensus.

We conducted a descriptive analysis (count, mean, percentage) to describe study characteristics and other results below. We deferred to the original study when defining statistical significance. We used codes and categories to help discern patterns in the operationalization of ageism and health outcomes. Because most “effect sizes” in the included studies were regression coefficients, we intentionally chose not to do a meta-analysis as recommended by numerous methodological studies (Becker & Wu, 2007; Fernandez-Castilla et al., 2019; Kim & Peiris, 2019). Key arguments opposing meta-analyzing regression coefficients include metric issues and accuracy issues. Specifically, regression coefficients are statistical point estimates of the relationship between an independent variable and a dependent variable given a set of covariates. When the set of covariates changes, the regression coefficient will also change, resulting in a change in its metric. Thus, regression coefficients from different models are not on the same metric (Becker & Wu, 2007; Fernandez-Castilla et al., 2018). Moreover, while there are a few statistical methods such as an iterative generalized least square or the multivariate Bayesian approach that can potentially address the above-mentioned problem, these methods hold strict distributional assumptions of regression modeling strategies that our current dataset does not qualify. As a result, a meta-analysis will likely produce an incorrect overall estimate of an “effect size.”

Quality Assessment

We used the National Heart, Lung, and Blood Institute (NHLBI)’s Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies to conduct quality appraisals of the included studies ( National Institutes of Health, 2014). The tool includes 14 questions that help to evaluate the internal validity of a study. We used 12 of its 14 questions as two questions were not applicable to the studies we reviewed. The NHLBI suggests using the questions to assess the potential for bias, rather than tallying them up to arrive at a summary judgment of quality. Accordingly, we rated each study to be good, fair and poor based on the risk of potential bias (Supplementary Table 2). A good study has the least risk of bias and results are valid, a fair study is susceptible to some bias that is not sufficient to invalidate its results, and a poor study indicates significant risks of bias. Operationally, good studies included those that were either checked ‘yes’ (meeting the standard) on all the 12 questions or had ‘no’ only on two questions (sample size justification and repeated exposure assessment) that were deemed non-fatal flaws. Studies that had flaws in the measurement of the exposure/outcome or in the adjustment of potential confounders were likely to be rated as poor quality if the bias was serious, or if they also had problems in other aspects (e.g., selection bias). Fair studies were those not rated as good or poor.

Results

Characteristics of Included Studies

Our final analytical sample included 67 studies with a total of 164,726 participants. Some of the included studies used the same dataset, so there was a possibility of duplicate participants. For example, seven studies were based on the Health and Retirement Study (HRS), and their average sample size was 5,205 (range = 2,966 to 7,712). If excluding potential duplicate participants and using the average sample size to represent the number of participants in studies that used the same dataset, the total number of participants involved in this systematic review was 66,325.

Of the 67 included studies, 72% (n = 47) were published after 2010 (see Supplement Figure 1 for publication trends). The studies were conducted in 17 different countries, with a majority in the United States (n = 27), followed by Germany (n = 11), Australia (n = 3), Canada (n = 3), Korea (n = 3), and other countries (n = 15). In addition, three were comparison studies across multiple countries/regions. (See Supplement Table 1 for a list of all included studies and their characteristics.)

The average sample size was 2,459 and the average age of respondents was 69. On average, 57% of the samples were female. Three studies included only female participants. Forty-two studies reported education levels of their samples, and the average education was 12 years. Of the 27 studies conducted in the United States, 16 reported racial/ethnic compositions of their samples. On average, 77% of their samples were Caucasian, 9.5% African American, 8.4% Asian, and 5.1% Hispanic.

Twenty-three studies (35%) used cross-sectional and 44 (66%) used longitudinal designs. Over half (n = 40) were secondary analyses based on existing datasets, including the HRS (n = 7), German Ageing Survey (DEAS; n = 7), Ohio Longitudinal Study of Aging (OLSA; n = 4), Baltimore Longitudinal Study of Aging (BLSA; n = 3), Precipitating Events Project (PEP; n = 3), Interdisciplinary Longitudinal Study of Adult Development and Aging (ILSE; n = 3), Chinese Longitudinal Healthy Longevity Survey (CLHLS; n =3), Midlife Development in the United States (MIDUS; n = 2), Australian Longitudinal Study of Aging (ALSA; n = 2), and others. In 27 studies, authors collected their own data.

Operationalization of Ageism

The included studies used various instruments to operationalize ageism. Based on the construct they measure, these instruments can be classified into three categories: age stereotype, self-perceptions of aging, and age discrimination (Table 1). Age stereotype refers to positive and/or negative stereotypes of older adults as a group. These instruments primarily seek respondents’ attitude toward older people in general (e.g., most old people get set in their ways and are unable to change.). Self-perceptions of aging refers to attitude toward one’s own aging. Instruments belonging to this category have explicit reference to the respondents themselves (e.g., do you feel that as you get older you are less useful?). Age discrimination refers to self-reported experiences of age-based discrimination. Below we report the measurement of the three ageism constructs, including their psychometric properties, based on information provided by the included studies.

Table 1.

Measurement of Ageism.

Category	Instrument	# Studies¹	Reliability²
Age Stereotype	Aging Stereotypes and Exercise Scale (Chalabaev et al., 2013)	1	0.80³
	Attitudes Toward Old People Scale (Tuckman & Lorge, 1953)	3	0.80
	Facts on Aging Quiz (Palmore, 1980)	1	0.66
	Image of Aging Scale (Levy et al., 2004)	1	0.84
	Older People Scale (Lu & Kao, 2009)	1	0.93
	One item “Please tell me where most people would place the status of people over 70?” (Age UK, 2011)	1	NR
	One item “What are the first five words or phrases that come to mind when you think of an older person?” (Levy & Langer, 1994)	3	0.84 – 0.94
	Five-item self-developed questions on attitude toward aging (Lai, 2009)	1	0.61
Self-Perceptions of Aging	Age-based Rejection Sensitivity Questionnaire (Kang & Chasteen, 2009)	1	0.95
	AgeCog Scale (Wurm et al., 2007)	3	0.77 – 0.82
	Aging Perceptions Questionnaire (Barker et al., 2007)	1	NR
	Attitudes to Ageing Questionnaire (Laidlaw et al., 2007)	5	0.62 - 0.78
	Attitudes Toward Own Aging (Lawton, 1975)	17	0.62 - 0.82
	Brief Aging Perceptions Questionnaire (Sexton et al., 2014)	1	0.79
	Expectations Regarding Aging (Sarkisian et al., 2005)	6	0.74 – 0.86
	Fear of Aging Scale (Sarkisian et al., 2005)	1	0.92
	Negative Belief in Aging in Spanish (Blanca et al., 2005)	1	0.64
	Personality in Intellectual Aging Contexts Scale (Lachman et al., 1982)	1	0.78
	Korean Symptom Management Beliefs Questionnaire (Yeom & Heidrich, 2009)	1	0.81
	One-item illness attribution (Stewart et al., 2012)	1	NR
	One-item “With age, do you feel more useless?” (Gu et al., 2016)	3	NR
	Four-item self-developed questions on attitude toward aging (Momtaz et al., 2013)	1	0.80
	Personal Aging Experience Scale (Steverink et al., 2001)	3	0.74 - 0.77
Age Discrimination	Everyday Discrimination Scale (Williams et al., 1997)	7	0.79 – 0.94
	One-item whether you had been discriminated against, or placed at a disadvantage due to their age, in the past 12 months (yes/no; Avidor et al., 2017)	1	NR
	Four-item self-developed scale (Garstka et al., 2005)	1	0.77
	Five-item scale adapted from measures of racial and gender discrimination (Sabik, 2013)	1	0.75

Note. ¹Number of studies that have used the instruments. One study used two instruments. ²Reliability is based on reports of the included studies. ³Figures in regular font are Cronbach’s α, in italic font are inter-rater reliability. NR = not reported.

Measurement of age stereotype

Twelve studies (18%) focused on age stereotype, which has been assessed by eight different instruments (Table 1). Of these, the Attitudes Toward Old People Scale and the single-item “What are the first five words or phrases that come to mind when you think of an older person?” were used most frequently (three studies each). The remaining measures have been used by one study only. Five of the eight age stereotype measures were reported as demonstrating good reliability (Cronbach’s α or inter-rater reliability ≥ .8) (Chalabaev et al., 2013; Levy & Langer, 1994; Levy et al., 2004; Lu & Kao, 2009; Tuckman & Lorge, 1953; see Table 1). The included studies provided no information about the validity of the age stereotype instruments.

Measurement of self-perceptions of aging

More than two-thirds (n = 45, 67%) of the included articles focused on self-perceptions of aging. Within these studies, a total of 16 different measures was used (Table 1). The five-item Attitudes Toward Own Aging scale (ATOA) was the most commonly used measure, used in 17 studies. Expectations Regarding Aging (ERA) and Attitudes to Ageing Questionnaire (AAQ) were used by six and five studies, respectively. The remaining measures were used by three or fewer studies. Seven of the 16 measures demonstrated good reliability (Cronbach’ α > =.8) (Kang & Chasteen, 2009; Lawton, 1975; Momtaz et al., 2013; Sarkistan et al., 2005; Wurm et al., 2007; Yeomn & Heidrich, 2009; see Table 1). Some evidence was provided to support the factor structure of three scales (ATOA, ERA, AgeCog Scale). No information related to the validity of other instruments was provided.

Measurement of age discrimination

Ten studies (15% of included articles) examined age discrimination. Of these, seven assessed the construct using the Everyday Discrimination Scale which was reported to have good reliability (Cronbach’ α > =.8; Williams et al., 1997; see Table 1). The EDS asked respondents to first report experiences of discrimination and then attribute the experiences to a cause. If age was reported as the cause, age discrimination was coded positive. The other three remaining studies each had its own self-reported age discrimination measure. Two of them had acceptable levels of reliability (α >= .7) and one used a single item with no reliability reported (see Table 1). Validity was not reported for any of the age discrimination measures.

Quality of Included Studies

Twenty-six studies (39%) were judged to be of good quality (marked * in Supplement Table 1), meaning the risk of bias is minimal, based on the NHLBI’s Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies (National Institutes of Health, 2014). Of these, all used longitudinal designs. Nineteen (28%), including twelve longitudinal and seven cross-sectional studies, were evaluated as fair in quality. Twenty-two studies (35%)—seven longitudinal and fifteen cross-sectional—were assessed to be of poor quality. The potential for measurement bias was the primary reason for study quality variation, followed by potential confounding and selection bias. (See Supplement Table 2 for study quality ratings.)

Associations of Ageism Constructs and Health Domains

The 67 studies included a total of ninety health-related outcome variables, which we categorized into seven domains: disease, mortality, physical/functional health, mental health, cognitive function, quality of life, and health behavior.² The most-studied health domains were physical/functional health (26 times), quality of life (15 times), and mental health (13 times). The least studied were disease (3 times) and mortality (8 times). Some studies included multiple health domains as outcomes.

Below we report findings pertinent to the association between each ageism construct and each health domain, first in all included studies (n = 67) and then in studies judged to be of good quality (n = 26). Unless otherwise noted, the terms statistically significant associations or significant associations in the following refer to associations assessed to be statistically significant by the original studies and in the expected direction. The majority of studies used p ≤ .05 to indicate statistical significance.

In all included studies

In total, seven longitudinal and five cross-sectional studies examined the association between age stereotype and health; thirty-one longitudinal and fourteen cross-sectional studies investigated the relationship between self-perceptions of aging and health, and six longitudinal and four cross-sectional studies analyzed the relationship between age discrimination and health.

Figure 2 shows the number of times a significant association between each ageism construct and each health domain has been reported. With respect to age stereotype which was examined in 12 articles, significant associations with physical/functional health (4 times), cognitive function (4 times), mental health (2 times), quality of life (2 times), health behavior (1 time), and disease (1 time) have been reported (see Figure 2). No study has examined the association between age stereotype and mortality.

Figure 2.

Associations of ageism and health.

Self-perceptions of aging was the focus of 45 articles and has been found to be significantly associated with all seven health domains, including physical/functional health (16 times), quality of life (10 times), health behavior (10 times), mortality (8 times), cognitive function (6 times), mental health (6 times), and disease (1 time) (see Figure 2). Non-significant associations of self-perceptions of aging with physical/functional health (3 times), cognitive function (1 time), quality of life (1 time), and disease (1 time) have also been reported.

Ten studies investigated the relationship between age discrimination and health. They reported that age discrimination was significantly associated with mental health (6 times), physical/functional health (3 times), and quality of life (3 times). Non-significant relationships were reported with cognitive function (1 time) and mental health (1 time). No studies have examined the association of age discrimination with diseases, mortality, and health behavior.

In good quality studies

If analyzing only the studies judged to be of good quality (n = 26), age stereotype was significantly correlated with diseases (1 time) and cognitive function (1 time) (Figure 2); self-perceptions of aging had significant associations with physical/functional health (8 times), mortality (5 times), cognitive function (4 times), health behavior (3 times), mental health (3 times), quality of life (1 time), and disease (1 time); and age discrimination was significantly associated with mental health (2 times). Two good-quality studies reported non-significant associations between self-perception of aging and cognitive function (1 time), and between age discrimination and mental health (1 time).

Overall, very few non-significant associations between ageism and health were reported (8 of 91 associations ≈ 9% in all included studies; 2 of 29 associations ≈ 7% in good quality studies). Evidence for a significant relationship between self-perceptions of aging and health is strong, considering the number of good quality studies that have reported such a relationship and the wide range of health outcomes involved. Significant associations between age stereotype and health, and between age discrimination and health, have also been reported by multiple studies. Few of these studies, however, were judged to be of good quality.

Moderators in Ageism-Health Associations

Moderators specify differences in the association between ageism and health by group or the value of a third variable. Of the 67 included studies, nine (13%) examined moderating effects and all reported significant findings (marked † in Supplement Table 1).). One study found that self-relevance moderated the association between age stereotype and memory decline. That study measured self-relevance as whether one’s own age was at or older than the age that he/she perceived to be old. Four studies reported that age, race/ethnicity, gender and subjective social status, respectively, moderated the association between self-perceptions of aging and health. Specifically, the first study found that the relationship between positive views on aging and physical limitations was stronger among older adults than among middle-aged adults. The second found racial/ethnic group differences: age expectation was significantly associated with depression among Latino older adults, but not among African American, Chinese, and Korean older adults. The third study showed that the association between self-perceptions of aging and fluid intelligence was stronger in men than in women. The fourth study found that subjective social status moderated the effects of cognitive aging beliefs on cortisol reactivity among older adults.

7With respect to the relationship between age discrimination and health, one study reported that perceived age discrimination was negatively associated with psychological well-being for older (age 65+) but not young (age 17-25) adults, whereas another reported that middle-aged adults (age 40-64) were more susceptible to the negative effects of age discrimination than older adults (age 65-93). The latter study also found that among middle-aged adults, perceived age discrimination was associated with a greater decline in subjective well-being for those who expected to live for more years. A third study found that the association between perceived age discrimination and mental health was stronger for women than for men, and that sense of control buffered the stress of age discrimination.

Discussion

This systematic review includes sixty-seven quantitative observational studies that have investigated the association between ageism and health. Findings reveal that ageism has been operationalized using a variety of instruments, which can be classified into three categories: age stereotype, self-perceptions of aging, and age discrimination. A majority of instruments was used within a single study, and the psychometric properties of most have not been established. About ninety health-related variables, which we classified into seven health domains—diseases, mortality, physical/functional health, mental health, cognitive function, quality of life, and health behavior—have been used as outcomes. Overall, the findings suggest that ageism is negatively associated with health. The strongest findings are for self-perceptions of aging which has been found to be significantly associated with all seven health domains. Age stereotype and age discrimination have also been found to be significantly associated with multiple health domains, though the evidence is weaker due to few good quality studies focusing on these two ageism constructs. Findings also suggest that several variables related to social-group membership (age, gender, race/ethnicity) and psychological attributes (self-relevance, sense of control, subjective social status, subjective life expectancy) moderate the relationship between ageism and health. This systematic review is the first to focus specifically on observational studies on ageism and health, and the first to provide a comprehensive summary of this field.

Operationalization of Ageism

Our findings make it clear that measurement is a major issue in the study of ageism. While a number of ageism measures has been developed, their psychometric properties—particularly validity—remain unclear. Some researchers developed their own ageism scales, others used single-item measures without providing justification. Nearly 30% of included articles reported neither the reliability nor validity of the ageism measures they used. As shown in Table 1, there is a lack of standardization in the measurement of ageism constructs. Our findings are consistent with Warmonth et al.’s (2016) review of the literature on aging perceptions and health. They noted that “the validation of measures used (e.g., testing of validity)” was a reason for variation in study quality. A recent systematic review of ageism scales shows that many existing scales have never undergone psychometric assessment, and that only one (Expectations Regarding Aging), among 11 ageism scales being analyzed, met requirements for psychometric validation (Ayalon et al., 2019).

Under-Studied Dimensions of Ageism

Even though a variety of instruments has been used to examine ageism, the dimensions of ageism that have been investigated are very limited. More than 80% of the included articles investigated the cognitive component of ageism, including attitude toward self (self-perceptions of aging) and toward older adults in general (age stereotype). Affective (prejudice) and behavioral (discrimination) dimensions remained under-studied. Further, most instruments assessed explicit, negative, and individual-level ageism. Given ageism’s multi-dimensionality (Iversen et al., 2009), current research offers a limited understanding of how ageism affects individual health. To date, observational studies have not examined implicit ageism, which researchers suggest to be more prevalent than explicit ageism in daily life (Hummert, 2011; Levy & Banaji, 2002). Few have examined positive ageism such as positive age stereotypes. Experimental studies suggest that implicit and explicit age priming (a technique used to elicit age-stereotype responses) have similar effects (Horton et al., 2008), and that the effects of negative age stereotypes are stronger than those of positive stereotypes (Meisner, 2012). We do not know whether these findings apply to observational studies. In addition, almost all included articles investigated individual-level ageism, and of these, most examined intra-individual ageism (self-perceptions of aging). Ageism at the institutional and cultural levels has rarely been studied.

Associations between Ageism and Health

The included studies provide support for a significant association between ageism and health. Overall, very few non-significant findings have been reported. However, several issues should be noted when interpreting such findings. First, many significant associations were reported by fair- to poor-quality studies only. The total number of good-quality studies is relatively small (n = 26). Second, some studies were based on data from the same source. For example, 6 of 45 articles that investigated associations between self-perceptions of aging and health analyzed data from the German Ageing Survey (DEAS), and 4 of 10 that examined the relationship between age discrimination and health used data from the Health and Retirement Survey (HRS). The opportunity to identify statistically significant relationships may have been inflated for studies using the same dataset. Third, publication bias may partially explain the dominance of significant findings in the included articles, as studies with insignificant findings are less likely to be published.

The relationship between self-perceptions of aging and health was investigated the most and received the strongest support. Self-perceptions of aging deserve some discussion. As Wurm et al. (2017) have suggested, self-perceptions of aging is likely to be affected by personality, relationships, and health conditions, in addition to the influence exercised by societal views of aging. Future research should distinguish between the health effects of self-perceptions of aging grounded in beliefs about aging and health effects grounded in personality, social relationships, and health conditions.

Only about 17% of all included studies investigated associations between age stereotype and health. We classified the ageism instruments into age stereotype and self-perceptions of aging based on the extent to which they assessed attitude toward older people in general or toward the self. According to the stereotype embodiment theory, age stereotypes are internalized and directed at the self as one ages (Levy, 2009). Others argue that older individuals do not necessarily “internalize”; rather, they resist age stereotypes (Zebrowitz, 2003). Theoretically, it is important to separate the two concepts—age stereotype and self-perceptions of aging—and examine their independent associations with health. However, none of the included studies examined both concepts simultaneously. It is possible that the measures of age stereotype in some included articles have actually assessed self-perceptions of aging.

Relatively few studies (15%) examined the relationship between age discrimination and health. Data from the United States, Britain, and Canada show a high prevalence of perceived discrimination based on age—higher than that based on sex and race (Lozon & Barratt, 2012; Rippon et al., 2014; Sutin et al., 2015). It is surprising that the relationship between age discrimination and health has not received more research attention. One reason may be that age discrimination usually is in subtle forms and may even be well-meaning, making it hard to be taken seriously. Given the long recognition in the scientific literature about adverse health consequences of perceived discrimination (Pascoe & Smart Richman, 2009), the study of age discrimination seems critically important.

The wide range of health outcomes found to be associated with ageism in the included articles is quite remarkable and suggests that ageism should be treated as a public health risk. However, the direction of effects between ageism and health may be complicated. For example, using longitudinal data, Ayalon (2018) found that depressive symptoms preceded perceived age discrimination, not the reverse. Other studies reported a bidirectional relationship between views of aging and physical health, but they determined that the impact of views on aging was stronger than reverse effects (Wurm et al., 2007). Future research is needed to understand the dynamic relationship between ageism and health. Further, even though 17 countries were represented in the included articles, most studies were conducted in Western countries. There are socio-cultural differences in attitude toward older people (North & Fiske, 2015). Some researchers have argued that differences in ageist attitudes between Eastern and Western cultures depend on the questions asked (Vauclair et al., 2017). Most ageism scales were developed in Western countries. To study the effects of ageism on health in non-western cultures, more work is needed to develop culturally appropriate measures of ageism. In addition, local socio-cultural context—including family and welfare norms—should be considered as these may affect how older people respond to different components of ageism (Wilińska et al., 2018).

Moderators in Associations of Ageism and Health

Of included studies, we identified nine (13%) that examined moderators in the association between ageism and health. The significant moderating effects imply that ageism, whether self-directed (self-perceptions of aging) or other-directed (age discrimination), does not have uniform effects across members of different social groups. Moreover, psychological characteristics such as sense of control may buffer the negative effects of ageism on health. Given the small number of studies that have investigated moderating effects, the findings should be treated as preliminary. However, the findings suggest that further investigations to examine moderators in the relationship between ageism and health are warranted.

Study Limitations

This systematic review is limited in several aspects. First, it included only quantitative observational studies written in English. Other research designs including qualitative observational studies and community-based controlled trials, and non-English publications were excluded. We also did not include the gray literature. Second, despite our effort to identify all relevant published studies, our search process may have overlooked eligible studies. Third, it is possible that errors were made when extracting data from the studies. Fourth, we did not conduct a meta-analysis or calculate effect sizes, hence, were not able to provide an overall quantitative estimate of the association between ageism and health. While it was an intentional decision to avoid producing biased statistical results, a meta-analysis would be warranted if advanced statistical methods become available. Fifth, we depended on the original studies’ report of statistical significance, which may have over-biased toward significant findings.

Directions for Future Research

The discussion above has identified many avenues for future research. First, developing valid and reliable instruments to assess cognitive, affective, and behavioral components of ageism should be a top priority. Second, future studies should focus more on the affective and behavioral components (prejudice and discrimination against older people) of ageism and expand investigations into implicit and positive ageism, as well as ageism at the institutional and cultural levels. Third, the relationship between different components of ageism and different dimensions of health should be investigated using multi-wave longitudinal data. Fourth, the pathways linking different ageism components and health outcomes should be examined. We noticed that some included articles have begun exploring physiological (e.g., C-reactive protein; Levy & Bavish, 2018), psychological (e.g., self-efficacy; Tovel et al., 2019), and behavioral (e.g., leisure activity engagement; Hicks & Siedlecki, 2017) pathways. Finally, greater efforts must be made to ensure the quality of studies in this field. Rigorous methodological approaches will allow future research to reveal the complex role that ageism plays in individual and population health. With the global rise of the older adult population, the relationship between ageism and health has become an important area of research. Our review of observational studies on this topic suggests that there is an urgent need to develop valid and reliable ageism measures and to expand investigations across multiple dimensions of ageism.

Supplemental Material

Supplemental Material, Supplemental_Materials - Associations of Ageism and Health: A Systematic Review of Quantitative Observational Studies

Supplemental Material, Supplemental_Materials for Associations of Ageism and Health: A Systematic Review of Quantitative Observational Studies by Rita Xiaochen Hu, Mengsha Luo, Anao Zhang and Lydia W. Li in Research on Aging

Footnotes

Acknowledgments

The authors thank Ms. Beth Zambone for copy editing the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Rita Xiaochen Hu

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Age UK. (2011). Grey matters—A survey of ageism across Europe: EU briefing and policy recommendations. https://www.ageuk.org.uk/Documents/EN-GB/For-professionals/ageism_across_europe_report.pdf?dtrk=true

*Avidor

Ayalon

Palgi

Bodner

(2017). Longitudinal associations between perceived age discrimination and subjective well-being: Variations by age and subjective life expectancy. Aging & Mental Health, 21(7), 761–765. https://doi.org/10.1080/13607863.2016.1156050

Ayalon

(2018). Perceived age discrimination: A precipitator or a consequence of depressive symptoms? The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 73(5), 860–869. https://doi.org/10.1093/geronb/gbw101

Ayalon

Dolberg

Mikulionienė

Perek-Białas

Rapolienė

Stypinska

Willińska

de la Fuente-Núñez

(2019). A systematic review of existing ageism scales. Ageing Research Reviews, 54, 100919. https://doi.org/10.1016/j.arr.2019.100919

Barber

S. J.

Mather

(2014). Stereotype threat in older adults: When and why does it occur, and who is most affected. In Verhaeghen

Hertzog

(Eds.), The Oxford handbook of emotion, social cognition, and problem solving during adulthood (pp. 302–320). Oxford University Press.

Barker

O’Hanlon

McGee

H. M.

Hickey

Conroy

R. M.

(2007). Cross-sectional validation of the aging perceptions questionnaire: A multidimensional instrument for assessing self-perceptions of aging. BMC Geriatrics, 7, 9. https://doi.org/10.1186/1471-2318-7-9

Becker

B. J.

M. J.

(2007). The synthesis of regression slopes in meta-analysis. Statistical Science, 22(3), 414–429. https://doi.org/10.1214/07-STS243

Blanca

M. J.

Sánchez

Trianes

M. V.

(2005). Cuestionario de evaluación de estereótipos negativos hacia la vejez [Negative Stereotypes Towards Aging Questionnaire]. Revista Multidisciplinar de Gerontotologia, 15, 212–220.

Butler

R. N.

(1975). Why survive? Being old in America. Harper & Row.

10.

Chalabaev

Emile

Corrion

Stephan

Clément-Guillotin

Pradier

d’Arripe-Longueville

(2013). Development and validation of the aging stereotypes and exercise scale. Journal of Aging and Physical Activity, 21(3), 319–334. https://doi.org/10.1123/japa.21.3.319

11.

Fernández-Castilla

Aloe

A. M.

Declercq

Jamshidi

Onghena

Natasha Beretvas

Van den Noortgate

(2019). Concealed correlations meta-analysis: A new method for synthesizing standardized regression coefficients. Behavior Research Methods, 51(1), 316–331. https://doi.org/10.3758/s13428-018-1123-7

12.

*Garstka

T. A.

Hummert

M. L.

Branscombe

N. R.

(2005). Perceiving age discrimination in response to intergenerational inequity. Journal of Social Issues, 61(2), 321–342. https://doi.org/10.1111/j.1540-4560.2005.00408.x

13.

*Gu

Brown

B. L.

Qiu

(2016). Self-perceived uselessness is associated with lower likelihood of successful aging among older adults in china. BMC Geriatrics, 16, 172. https://doi.org/10.1186/s12877-016-0348-5

14.

*Hicks

S. A.

Siedlecki

K. L.

(2017). Leisure activity engagement and positive affect partially mediate the relationship between positive views on aging and physical health. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 72(2), 259–267. https://doi.org/10.1093/geronb/gbw049

15.

Hummert

M. L.

(2011). Age stereotypes and aging. In Schaie

K. W.

Willis

S. L.

(Eds.), Handbook of the psychology of aging (7th ed., pp. 249–262). Elsevier.

16.

Horton

Baker

Pearce

G. W.

Deakin

J. M.

(2008). On the malleability of performance: Implications for seniors. Journal of Applied Gerontology, 27, 446–465. https://doi.org/10.1177/0733464808315291

17.

Iversen

T. N.

Larsen

Solem

P. E.

(2009). A conceptual analysis of ageism. Nordic Psychology, 61, 4–22. https://doi.org/10.1027/1901-2276.61.3.4

18.

Kang

S. K.

Chasteen

A. L.

(2009). The development and validation of the age-based rejection sensitivity questionnaire. The Gerontologist, 49, 303–316. https://doi.org/10.1093/geront/gnp035

19.

Kim

Peiris

T. B.

(2019). Meta analysis of regression: A review and new approach with application to linear-circular regression model. Communications in Statistics—Theory and Methods. https://doi.org/10.1080/03610926.2019.1679183

20.

Lachman

M. E.

Baltes

Nesselroade

J. R.

Willis

S. L.

(1982). Examination of personality-ability relationships in the elderly: The role of the contextual (interface) assessment mode. Journal of Research in Personality, 16, 485–501. https://doi.org/0.1016/0092-6566(82)90007-1

21.

*Lai

D. W. L.

(2009). Older Chinese’ attitudes toward aging and the relationship to mental health: An international comparison. Social Work in Health Care, 48(3), 243–259. https://doi.org/10.1080/00981380802591957

22.

Laidlaw

Power

M. J.

Schmidt

(2007). The Attitudes to Ageing Questionnaire (AAQ): Development and psychometric properties. International Journal of Geriatric Psychiatry: A Journal of the Psychiatry of Late Life and Allied Sciences, 22, 367–379. https://doi.org/10.1002/gps.1683

23.

Lamont

R. A.

Swift

H. J.

Abrams

(2015). A review and meta-analysis of age-based stereotype threat: Negative stereotypes, not facts, do the damage. Psychology and Aging, 30, 180–193. https://doi.org/10.1037/a0038586

24.

Lawton

M. P.

(1975). The Philadelphia geriatric center morale scale: A revision. Journal of Gerontology, 30, 85–89. https://doi.org/10.1093/geronj/30.1.85

25.

Levy

B. R.

(2009). Stereotype embodiment: A psychosocial approach to aging. Current Directions in Psychological Science, 18, 332–336. https://doi.org/10.1111/j.1467-8721.2009.01662.x

26.

*Levy

B. R.

Banaji

M. R.

(2002). Implicit ageism. In Nelson

T. D.

(Ed.), Ageism: Stereotyping and prejudice against older persons. The MIT Press.

27.

*Levy

B. R.

Bavishi

(2018). Survival advantage mechanism: Inflammation as a mediator of positive self-perceptions of aging on longevity. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 73(3), 409–412. https://doi.org/10.1093/geronb/gbw035

28.

Levy

Langer

(1994). Aging free from negative stereotypes: Successful memory in China among the American deaf. Journal of Personality and Social Psychology, 66(6), 989. https://doi.org/10.1037//0022-3514.66.6.989

29.

Levy

B. R.

Kasl

S. V.

Gill

T. M.

(2004). Image of aging scale. Perceptual and Motor Skills, 99, 208–210. https://doi.org/10.2466/PMS.99.4.208-210

30.

*Levy

B. R.

Zonderman

A. B.

Slade

M. D.

Ferrucci

(2012). Memory shaped by age stereotypes over time. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 67(4), 432–436. https://doi.org/10.1093/geronb/gbr120

31.

Liberati

Altman

D. G.

Tetzlaff

Mulrow

Gøtzsche

P. C.

Ioannidis Mike Clarke

J.P.A.

Devereaux

P.J.

Kleijnen

Moher

(2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. PLOS Medicine, 6(7), e1000100. https://doi.org/10.1371/journal.pmed.1000100

32.

Lozon

J. C.

Barratt

(2012). Revera report on ageism. Retrieved, 9(22), 2015.

33.

Kao

S. F.

(2009). Attitudes towards old people in Taiwan: Scale development and preliminary evidence of reliability and validity. Journal of Education and Psychology, 32(1), 147–171.

34.

Meisner

B. A.

(2012). A meta-analysis of positive and negative age stereotype priming effects on behavior among older adults. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 67, 13–17. https://doi.org/10.1093/geronb/gbr062

35.

Momtaz

Y. A.

Hamid

T. A.

Masud

Haron

S. A.

Ibrahim

(2013). Dyadic effects of attitude toward aging on psychological well-being of older Malaysian couples: An actor–partner interdependence model. Clinical Interventions in Aging, 8, 1413–1420. https://doi.org/10.2147/CIA.S51877

36.

National Institutes of Health. (2014). Quality assessment tool for observational cohort and cross-sectional studies. National Heart, Lung, and Blood Institute . https://www.nhlbi.nih.gov/health-topics/study-quality-assessment-tools

37.

North

M. S.

Fiske

S. T.

(2015). Modern attitudes toward older adults in the aging world: A cross-cultural meta-analysis. Psychological Bulletin, 141(5), 993–1021. https://doi.org/10.1037/a0039469

38.

Palmore

(1980). The facts on aging quiz: A review of findings. The Gerontologist, 20, 669–672. https://doi.org/10.1093/geront/20.6.669

39.

Pascoe

E. A.

Smart Richman

(2009). Perceived discrimination and health: A meta-analytic review. Psychological Bulletin, 135(4), 531–554. https://doi.org/10.1037/a0016059

40.

Rippon

Kneale

de Oliveira

Demakakos

Steptoe

(2014). Perceived age discrimination in older adults. Age and Ageing, 43(3), 379–386. https://doi.org/10.1093/ageing/aft146

41.

*Sabik

N. J.

(2015). Ageism and body esteem: Associations with psychological well-being among late middle-aged African American and European American women. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 70(2), 191–201. https://doi.org/10.1093/geronb/gbt080

42.

Sarkisian

C. A.

Steers

W. N.

Hays

R. D.

Mangione

C. M.

(2005). Development of the 12-item expectations regarding aging survey. The Gerontologist, 45, 240–248. https://doi.org/10.1093/geront/45.2.240

43.

Sexton

King-Kallimanis

B. L.

Morgan

McGee

(2014). Development of the Brief Ageing Perceptions Questionnaire (b-apq): A confirmatory factor analysis approach to item reduction. BMC Geriatrics, 14(1), 44. https://doi.org/10.1186/1471-2318-14-44

44.

Steverink

Westerhof

G. J.

Bode

Dittmann-Kohli

(2001). The personal experience of aging, individual resources, and subjective well-being. The Journals of Gerontology, Series B: Psychological Sciences, 56, 364–373. https://doi.org/10.1093/geronb/56.6.P364

45.

*Stewart

T. L.

Chipperfield

J. G.

Perry

R. P.

Weiner

(2012). Attributing illness to “old age”: Consequences of a self-directed stereotype for health and mortality. Psychology & Health, 27(8), 881–897. https://doi.org/10.1080/08870446.2011.630735

46.

Sutin

A. R.

Stephan

Carretta

Terracciano

(2015). Perceived discrimination and physical, cognitive, and emotional health in older adulthood. The American Journal of Geriatric Psychiatry: Official Journal of the American Association for Geriatric Psychiatry, 23(2), 171–179. https://doi.org/10.1016/j.jagp.2014.03.007

47.

*Tovel

Carmel

Raveis

V. H.

(2019). Relationships among self-perception of aging, physical functioning, and self-efficacy in late life. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 74(2), 212–221. https://doi.org/10.1093/geronb/gbx056

48.

Tuckman

Lorge

(1953). Attitudes toward old people. Journal of Social Psychology, 37, 249–260.

49.

United Nations, Department of Economic and Social Affairs, Population Division. (2015). World population ageing 2015 (ST/ESA/SER.A/390). https://www.un.org/en/development/desa/population/publications/pdf/ageing/WPA2015_Report.pdf

50.

Vauclair

C. M.

Hanke

Huang

L. L.

Abrams

(2017). Are Asian cultures really less ageist than Western ones? It depends on the questions asked. International Journal of Psychology/Journal international de psychologie, 52(2), 136–144. https://doi.org/10.1002/ijop.12292

51.

Williams

D. R.

Jackson

J. S.

Anderson

N. B.

(1997). Racial differences in physical and mental health: Socio-economic status, stress and discrimination. Journal of Health Psychology, 2, 335–351. https://doi.org/10.1177/135910539700200305

52.

Wilińska

de Hontheim

Anbäcken

E. M.

(2018). Ageism in a cross-cultural perspective: Reflections from the research field. In Ayalon

Tesch-Römer

(Eds.), Contemporary perspectives on ageism (pp. 425–440). Springer.

53.

Warmoth

Tarrant

Abraham

Lang

I. A.

(2016). Older adults’ perceptions of ageing and their health and functioning: A systematic review of observational studies. Psychology, Health & Medicine, 21, 531–550. https://doi.org/10.1080/13548506.2015.1096946

54.

Wurm

Diehl

Kornadt

A. E.

Westerhof

G. J.

Wahl

H.-W.

(2017). How do views on aging affect health outcomes in adulthood and late life? Explanations for an established connection. Developmental Review, 46, 27–43. https://doi.org/10.1016/j.dr.2017.08.002

55.

*Wurm

Tesch-Römer

Tomasik

M. J.

(2007). Longitudinal findings on aging-related cognitions, control beliefs, and health in later life. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 62(3), P156–164. https://doi.org/10.1093/geronb/62.3.p156

56.

Yeom

H. E.

Heidrich

S. M.

(2009). Effect of perceived barriers to symptom management on quality of life in older breast cancer survivors. Cancer Nursing, 32(4), 309–316. https://doi.org/10.1097/NCC.0b013e31819e239e

57.

Zebrowitz

L. A.

(2003). Aging stereotypes: Internalization or inoculation? A commentary. The Journals of Gerontology Series B: Psychological Sciences, 58, P214–P215. https://doi.org/10.1093/geronb/58.4.P214

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.47 MB