Screening for Cognitive Impairment in Older Adults Attending an Eye Clinic: METHOD

21 Oct


Following approval from the University of Texas Medical Branch (UTMB) institutional review board, we collected social, demographic and health information from 100 consecutive Hispanic, African-American and white adults, aged >55, who came to receive ophthalmologic care at the UTMB Eye Center. Patients included in the study had no history of dementia, were able to give informed consent for the interviews and physical examination, and were able to communicate in English or Spanish. The eye examination was done by a board-certified ophthalmologist. Patients with significant cognitive, physical and emotional disorders were referred to appropriate professionals for further management.
The data collection was performed over a period of eight months (November 2003 to June 2004). In addition to sociodemographic and medical information and using standardized questionnaires, we collected information on physical activity level and cognition. All participants also had a comprehensive eye examination as well as blood pressure measurements. The assessments lasted approximately one hour. Interviews were done in English or Spanish, depending on the preferences of participants. The bilingual interviewers were thoroughly trained in the administration and scoring of the questionnaires.


Cognitive Impairment
Saint Louis University Mental Status Examination (SLUMS) is a new and easy to use 11-item scale with good reliability and validity for screening for mild and severe cognitive impairments. SLUMS was developed to detect mild cognitive impairment and to distinguish it from dementia. SLUMS scores range from 0 to 30, with lower scores indicating increasing severity of cognitive impairments in the cognitive domains of orientation, memory, attention and executive functions. actos medication
Our primary outcome was the presence of any cognitive impairment (mild or severe) using the SLUMS scale. Mild cognitive impairment was defined as a SLUMS score of 20-27 for subjects with high-school education and above, and as a score of 15-19 for those with less than high-school education. Scores of 1-19 (for those with high-school education) and 1-14 (for those with less than high-school education) depicted severe cognitive impairment (dementia) on the SLUMS scale.

Vision Impairment
The Visual Data Battery (VDB) was used to assess vision and screen for ocular pathologies with the use of the following instruments:

1. Snellen letter chart for 20-foot distance vision (with correction, if any). Best corrected vision was obtained using established techniques for refraction.

2. Rosenbaum pocket vision screener at 14 inches for near (reading) vision

3. Refractor

4. Ishihara plates for color vision testing using handheld pictorial book

5. Amsler grid, a simple grid for assessing central field and for other visual field examination using confrontation method

6. Haag Streit 900 or equivalent slit lamp to view the anterior part of the eye, including eyelid, conjunctiva, cornea, anterior chamber, iris, pupil and the lens

7. Goldman tonometer attached to a slit lamp to measure intraocular pressure

8. Indirect and direct ophthalmoscope with dilated fundus examination to look at the retina and optic nerve using slit lamp and handheld +78 & +90 D lens and 20 D lens for indirect ophthalmoscopy.

Other Measures
Demographic and health measures included age, gender, ethnicity, years of education, marital status, and systolic blood pressures. For selected medical condition, subjects were asked if they have ever had a physician diagnosis of stroke, diabetes and hypertension. Physical Activity Scale for the Elderly (PASE), a 10-item instrument, was used to measure level of self-reported occupational, household and leisure activities during a one-week period.

Statistical Analysis
For screening covariates, we examined social, demographic, health and lifestyle factors associated with cognitive impairment with descriptive contingency tables (Chi-squared and Fisher’s exact test). Multivariate logistic regression models were then used to identify factors associated with cognitive impairment as a function of ethnicity (white vs. non-white) by comparing cognitively impaired subjects to those with normal cognition. The factors adjusted for in the regression models included age (prevalence of cognitive impairment increases with age) and all variables that had a p value <0.3 in the univariate analyses. The unadjusted model included white and nonwhite ethnicity. Because of small numbers of nonwhites, African-American and Hispanic adults were grouped in the same category of nonwhite ethnicity. Adjusted model 1 included age, years of education and systolic blood pressure.  While vision (normal visual acuity in both eyes, yes versus no) was added to the fully adjusted model 3. All analyses were performed using the SAS System for Windows®, Version 8.12 (SAS Institute, Cary, NC).