Abstract
Persons without dementia residing in a biracial community completed a brief scale of proneness to psychological distress, and 1,064 were subsequently examined for incident Alzheimer disease (AD) 3 to 6 years later. In analyses controlling for selected demographic and clinical variables, persons prone to distress were 2.4 times more likely to develop AD than persons not distress prone. This effect was substantially stronger in white persons compared to African Americans.
In a recent study of a group of predominantly white clergy members, the personality trait of neuroticism, an indicator of proneness to psychological distress, was associated with risk of Alzheimer disease (AD).1 In this study, we tested for the presence of the association in a stratified random sample of older residents of a biracial community.
Methods.
Participants.
We interviewed 6,158 persons aged 65 years or older (79% of those eligible) from a geographically defined area of Chicago. Based on brief cognitive tests in the interview and detailed clinical evaluation of a stratified random sample, a disease-free cohort was identified. Three years later, a sample of these persons was evaluated for incident disease, and another disease-free cohort was identified, from which a second sample was chosen and evaluated for incident disease 3 years after that. The study was approved by the Rush Institutional Review Board.
These two samples were combined, yielding 1,145 persons, with analyses based on 1,064 with data on key study variables. At entry into the study’s incidence portion, mean age was 73.8 years (SD = 9.6); mean education was 12.9 years (SD = 6.3); 61.9% were women and 49.8% were African American.
Clinical evaluation.
Persons sampled for ascertainment of disease incidence had a medical history, neurologic examination, cognitive testing, and brain MRI in those with possible stroke. AD was clinically diagnosed according to National Institute of Neurological and Communicative Disorders and Stroke–Alzheimer’s Disease and Related Disorders Association criteria,2 and major depressive disorder was classified using the criteria of the Diagnostic and Statistical Manual of Mental Disorders III-R, as reported in more detail elsewhere.3,4
At entry into the study’s incidence portion, we assessed APOE genotype, depressive symptoms (with the 10-item Center for Epidemiologic Studies Depression [CES-D] scale), and frequency of participation in cognitively stimulating activities, as previously described.3,4
Assessment of distress proneness.
Distress proneness was assessed at the initial population interview with four items from the Neuroticism scale of the NEO Five-Factor Inventory (item numbers: 1, 21, 36, 51).5 Participants rated agreement with each item on a five-point scale (0 to 4) with higher scores indicating more distress proneness. The item scores were summed and multiplied by three to make the total score more comparable to the original 12-item scale, as previously described.6 In a separate group of 102 older persons without dementia from the Rush Memory and Aging Project7 (approved by the Rush Institutional Review Board), the four-item scale had a correlation of 0.86 (p < 0.001) with the standard 12-item scale, supporting the validity of the brief measure.
Data analysis.
Logistic regression was used to estimate the odds of developing AD as a function of level of distress proneness with model estimates weighted to account for the stratified sampling.3 Variance estimation was based on jackknife repeated replication.
Results.
At baseline, distress proneness scores ranged from 0 to 42 (mean = 15.5, SD = 11.3). Level of distress proneness (t = −0.09, p = 0.932) and its correlation with selected demographic and clinical variables associated with risk of AD (table 1) were similar in African American and white persons. Women had higher scores than men among African Americans (t = 3.73, p < 0.001) but less evidently among whites (t = 0.58, p = 0.561). Distress proneness was positively related to depressive symptomatology in both subgroups, supporting the construct validity of the scale. Its internal consistency was low, however, with a coefficient alpha of 0.46 in the entire disease-free cohort (0.44 in African Americans, 0.51 in whites), reflecting the scale’s brevity.
Table 1 Correlation of the distress proneness measure with other covariates in African Americans (above diagonal) and whites (below diagonal)
During follow-up, 170 persons developed AD. In a logistic regression model that controlled for age, sex, race, education, possession of an APOE-ε4 allele, and follow-up time, each one point on the distress proneness scale increased the odds of AD by 6% (OR = 1.060; 95% CI: 1.012, 1.109). Thus, a distress prone person (score = 24, 90th percentile) was 2.4 times more likely to develop disease than someone not prone to distress (score = 9, 10th percentile).
Because depressive symptoms, a common form of distress, have also been associated with AD,4 we repeated the analysis controlling for CES-D score and CES-D score squared to account for linear and nonlinear effects of depressive symptomatology, and the effect of distress proneness was not substantially changed (OR = 1.057; 95% CI: 1.009, 1.107). Results were also comparable when persons with major depression at the incident disease evaluation (n = 27 [2.5%]) were excluded (OR = 1.063; 95% CI: 1.014, 1.115).
Cognitive activity is related to risk of AD3 and was correlated with distress proneness. Therefore, we repeated the analysis controlling for cognitive activity level.3 The effect of distress proneness was unchanged (OR = 1.052; 95% CI: 1.004, 1.101).
Because knowledge about distress proneness and dementia in racial or ethnic subgroups is limited, we repeated the analysis with a term for the interaction of distress proneness with race. There was an interaction (p = 0.0498). The association of distress proneness with disease incidence was robust in white persons but was less substantial in African Americans (figure). To confirm this effect, we constructed separate models for African American and white participants (table 2). With each point on the distress proneness scale, the odds of developing AD increased by 12% in whites (p = 0.001) compared to 2% in African Americans (p = 0.409).
Figure. Probability of developing incident Alzheimer disease (AD) as a function of distress proneness in older African American (solid line) and white (dotted line) persons, adjusted for age, sex, education, ε4, and follow-up time.
Table 2 Association of distress proneness with incident AD in white and African American persons
In subsequent analyses, distress did not interact with age, sex, education, or APOE genotype (each p > 0.4).
Discussion.
In a stratified random sample of older persons from a biracial community, we found that higher distress proneness was associated with an increased risk of incident AD during a 3- to 6-year period of observation. On average, the odds of developing AD were more than doubled in persons prone to distress compared to those not distress prone. The results support the hypothesis that distress proneness is a risk factor for AD in old age.
The association between distress proneness and incident AD observed in this population-based sample of older persons is consistent with the only previous investigation of which we are aware, a study of a selected group of predominantly white clergy members.1 Of note, however, the association of distress proneness with risk of AD was robust in whites but less substantial in African Americans. This dissociation was unexpected. In a previous study of the population from which these samples were drawn, race did not modify the association of distress with cognitive decline, but the inverse association of distress with baseline cognitive level was reduced in African Americans compared to whites.6 The distribution of the distress proneness measure, its internal consistency, and its association with depressive symptoms were similar in African American and white persons, suggesting that the dissociation is probably not due to racial differences in distress level or in the psychometric properties of the scale. Although African Americans have been disproportionately exposed to social conditions considered to be particularly stressful, there do not appear to be strong racial differences in negative emotional states or traits.8 Older African Americans may have developed personal9 or social10 resources that help to reduce psychological distress or its deleterious effect on health, but the nature and neurobiologic impact of such resources remain to be determined. Further research on the association of distress proneness with loss of cognition in older African Americans is needed.
Study strengths include the large, population-based sample and use of widely accepted AD criteria implemented by an experienced physician following a uniform clinical evaluation. The principal study limitation is that we used a brief measure of distress proneness. Because this brief measure lacks the precision of longer scales,5 the present results may underestimate the magnitude of the association of distress proneness with risk of AD.
Acknowledgment
The authors thank the residents of Morgan Park, Washington Heights, and Beverly who participated in the study. They also thank Ann Marie Lane for community development and oversight of project coordination, Michelle Bos, Holly Hadden, Flavio LaMorticella, and Jennifer Tarpey for coordination of the study, Hye-Jin Nicole Kim for analytic programming, and the staff of the Rush Institute for Healthy Aging.
Footnotes
-
Supported by National Institute on Aging grants R01 AG11101, RO1 AG17917, and P30 AG10161 and by National Institute of Environmental Health Sciences grant R01 ES10902.
Received June 3, 2004. Accepted in final form September 7, 2004.
Disputes & Debates: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.