To the Editor
Little research has been conducted on factors that account for why some older persons recover from disability and others do not. We considered a new culture-based explanatory factor: age stereotypes (defined as beliefs about old people as a category).1 Positive age stereotypes may promote recovery from disability through several pathways: limiting cardiovascular response to stress,2 improving physical balance,3 enhancing self-efficacy,4 and increasing engagement in healthy behaviors.1,4 We hypothesized that older persons with positive age stereotypes would be more likely to recover from disability than those with negative age stereotypes. Recovery was based on 4 essential activities of daily living (ADLs; bathing, dressing, transferring, and walking) that are strongly associated with use of health care services and longevity.5,6
Methods
The sample was drawn from the Precipitating Events Project, a study with high participation (75.2%) and follow-up (95.4%) rates. Participants were recruited from a health plan in greater New Haven, Connecticut, interviewed monthly for up to 129 months, and completed home-based assessments every 18 months from March 1998 through December 2008.5,6 Inclusion criteria were age of 70 years or older, English-speaking, community-living, nondisabled (ie, independent in the 4 ADLs at baseline), responded to the baseline age-stereotype measure, and experienced at least 1 month of ADL disability during follow-up (117 participants remained nondisabled). The final sample consisted of 598 participants (79.3% of the Precipitating Events Project cohort). The Yale University Human Subjects Committee approved the study; all participants provided verbal informed consent.
Age stereotypes were assessed by asking, “When you think of old persons, what are the first 5 words or phrases that come to mind?”1,4 Responses (coded on a 5-item scale ranging from 1= most negative [eg, decrepit] to 5= most positive [eg, spry]) were averaged and dichotomized at the mean of 2.12.1,4 Interrater reliability was 94.0%. This measure has been validated and the stereotypes have been found to be stable across time.1,4 Recovery was assessed by 3 validated ADL-improvement transitions: severe disability (needing assistance in 3 or 4 of the ADLs) to no disability (independent in all 4 ADLs), severe disability to mild disability (needing assistance in 1 or 2 of the ADLs), and mild disability to no disability6 Covariates related to disability in previous studies5,6 included age, sex, race, education, living alone, number of physician-diagnosed chronic conditions, Mini-Mental State Examination score, depressive symptoms, and physical frailty (ie, >10 seconds on a rapid-gait test). The latter 5 covariates were updated every 18 months.
A competing risk Cox model for recurrent events6 evaluated the likelihood of the positive age–stereotype group to experience each recovery transition. This model adjusted for time participants spent in each ADL-disability state (severe, mild, or none), whether they recovered or worsened, and number of times and degree to which they recovered or worsened.6 Absolute risk increase percentages were calculated for the positive age–stereotype group. An individual growth model examined the association between age stereotypes and ADL-disability states over the 129 months. All models were adjusted for all covariates. We used SAS version 9.2 (SAS Institute Inc) and considered 2-tailed P values of less than .05 as significant.
Results
Participants in the positive and negative age–stereotype groups were similar at baseline (Table 1).
Table 1. Baseline Characteristics of Study Participants by Age-Stereotype Groups.
| Age Stereotypes | ||||
|---|---|---|---|---|
| Overall (n = 598) | Positive (n = 232) | Negative (n = 366) | P Valuea | |
| Age, mean (SD), y | 79 (5.2) | 79 (5.1) | 79 (5.4) | .91 |
| Female, No. (%) | 378 (63.2) | 230 (62.8) | 148 (63.8) | .81 |
| Nonwhite race/ethnicity, No. (%)b | 49 (8.2) | 22 (9.5) | 27 (7.4) | .36 |
| Physically frail, No. (%) | 267 (44.7) | 99 (42.7) | 168 (45.9) | .44 |
| Education, mean (SD), y | 12 (2.8) | 12 (2.7) | 12 (2.9) | .24 |
| Chronic conditions, mean (SD) | 2 (1.2) | 2 (1.3) | 2 (1.2) | .34 |
| Mini-Mental State Examination score, mean (SD) | 27 (2.5) | 27 (2.5) | 27 (2.5) | .67 |
| Depression, No. (%)c | 82 (13.7) | 25 (10.8) | 57 (15.8) | .10 |
| Lives alone, No. (%) | 236 (39.5) | 93 (40.1) | 143 (39.1) | .80 |
The Wilcoxon rank-sum was used to evaluate differences in means. The χ2 and Fisher exact tests were used to evaluate differences in percentages.
Ethnic and racial groups were self-determined according to US Census categories.
Defined as Center for Epidemiologic Studies Depression Scale score of 20 or greater.
Relative to the negative age–stereotype group, the positive age–stereotype group had the following hazard ratios for recovery transitions: severe to none, 1.44 (95% CI, 1.01 to 2.06); severe to mild, 1.23 (95% CI, 1.03 to 1.46); and mild to none, 1.15 (95% CI, 1.03 to 1.29). The positive age–stereotype group showed an advantage in the associated absolute risk increase percentages (Table 2). As indicated by the individual growth model, the age–stereotype group × time interaction (β = −0.0004 [95% CI, −0.0006 to −0.0002]; P =.001) showed that the positive age–stereotype group had a significantly slower rate of ADL decline.
Table 2. Association of Positive Age Stereotypes With Recovery From Disability.
| Recovery Transitiona | Increase in Absolute Risk of Recovery for Positive Age–Stereotype Group, %b | Likelihood of Recovery by Positive Age–Stereotype Groupc | |
|---|---|---|---|
| Hazard Ratio (95% CI)d | P Value | ||
| Severe disability to no disability | 7.7 | 1.44 (1.01-2.06) | .04 |
| Severe disability to mild disability | 8.1 | 1.23 (1.03-1.46) | .02 |
| Mild disability to no disability | 1.1 | 1.15 (1.02-1.29) | .02 |
The number of person-months at risk for the transition from severe disability to no disability was 4704, for the transition from severe disability to mild disability it was 4704 and for the transition from mild disability to no disability it was 8411.
The absolute risks were calculated from models that included age-stereotype group and the 9 covariates from Table 1. The time-dependent covariates were updated every 18 months. Values refer to the absolute risk of making the specific transitions between month t and t + 1. The absolute risks were calculated using coefficients obtained from a set of pooled logistic regression models (1 for each transition).
The competing risk model adjusted for time participants spent in each activity of daily living disability state (severe, mild, or none), whether they recovered or worsened, and number of times and degree to which they recovered or worsened.6
The hazard ratios were calculated from models that included age-stereotype group and 9 covariates. The time-dependent covariates were updated every 18 months.
Comment
Older persons with positive age stereotypes were 44% more likely to fully recover from severe disability than those with negative age stereotypes. The study was limited by recruitment from 1 community; however, its demography is similar to the US population, except for an underrepresentation of black participants. Further research is needed to determine whether interventions to promote positive age stereotypes could extend independent living in later life.
Acknowledgments
Funding/Support: Research for this report was funded by National Institutes on Health grants R01AG032284 and R01HL089314 (Dr Levy) and R37AG17560, R01AG023993, and K24AG021507 (Dr Gill). In addition, Dr Levy was supported by an investigator award from the Patrick and Catherine Weldon Donaghue Medical Research Foundation. The Precipitating Events Project was conducted at the Yale Claude D. Pepper Older Americans Independence Center (grant P30AG21342).
Role of the Sponsor: The sources of funding and support had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.
Footnotes
Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Author Contributions: Dr Levy had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Levy, Slade, Gill.
Acquisition of data: Levy, Gill.
Analysis and interpretation of data: Levy, Slade, Murphy, Gill.
Drafting of the manuscript: Levy, Slade.
Critical revision of the manuscript for important intellectual content: Levy, Slade, Murphy, Gill.
Statistical analysis: Levy, Slade, Murphy.
Obtained funding: Levy, Gill.
Study supervision: Levy, Gill.
References
- 1.Levy B, Langer E. Aging free from negative stereotypes: successful memory in China and among the American Deaf. J Pers Soc Psychol. 1994;66(6):989–997. doi: 10.1037//0022-3514.66.6.989. [DOI] [PubMed] [Google Scholar]
- 2.Levy BR, Hausdorff JM, Hencke R, Wei JY. Reducing cardiovascular stress with positive self-stereotypes of aging. J Gerontol B Psychol Sci Soc Sci. 2000;55(4):205–213. doi: 10.1093/geronb/55.4.p205. [DOI] [PubMed] [Google Scholar]
- 3.Levy BR, Leifheit-Limson E. The stereotype-matching effect: greater influence on functioning when age stereotypes correspond to outcomes. Psychol Aging. 2009;24(1):230–233. doi: 10.1037/a0014563. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Levy BR. Stereotype embodiment: a psychosocial approach to aging. Curr Dir Psychol Sci. 2009;18(6):332–336. doi: 10.1111/j.1467-8721.2009.01662.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hardy SE, Gill TM. Recovery from disability among community-dwelling older persons. JAMA. 2004;291(13):1596–1602. doi: 10.1001/jama.291.13.1596. [DOI] [PubMed] [Google Scholar]
- 6.Gill TM, Allore HG, Gahbauer EA, Murphy TE. Change in disability after hospitalization or restricted activity in older persons. JAMA. 2010;304(17):1919–1928. doi: 10.1001/jama.2010.1568. [DOI] [PMC free article] [PubMed] [Google Scholar]