On the use of surrogate respondents for controls in a case-control study of Alzheimer's disease.

OBJECTIVE: To examine the presence and extent of bias introduced by using surrogate respondents for healthy controls in a case-control study of Alzheimer's disease (AD). DESIGN: Comparative study of matched responses to questionnaire ascertaining lifestyle issues. SETTING: University Hospitals/Case Western Reserve University Alzheimer Center. PARTICIPANTS: Controls (n = 50) were identified through the Research Registry. Surrogates (n = 50) were their healthy relatives or friends. MEASUREMENTS: Answers in the areas of demographic and occupational history, smoking habits, medical history, dietary intake, and leisure and work activities were recorded. The analysis was based on methods for paired data. Continuous variables were analyzed, focusing on paired differences between self and surrogate responses. RESULTS: For occupations and exposures, over 80% of the surrogates agreed with the subjects on over 80% of the questions. On smoking history, over 90% of the surrogates agreed with the subjects on over 70% of the questions. On leisure and work activities, over 70% of the surrogates agreed with the subjects on over 50% of the questions. There was less agreement regarding medical history. For continuous variables, most paired t-tests of zero mean difference between self and surrogate responses resulted in nonrejection of this hypothesis. Computed mean differences were not always positive or always negative. CONCLUSION: We did not find systematic under- or overreporting by the surrogates of the controls. Therefore, if there are biases in the responses of surrogates of the AD cases in our case-control study, they would not be canceled out by using surrogates for the controls.

Association of Alzheimer's disease and smoking: the case for sibling controls.

OBJECTIVES: To study the association between cigarette smoking and Alzheimer's Disease (AD). DESIGN: Intrafamily case-control, with sibling controls and a variable number of controls per case. SETTING: AD cases were identified through the Research Registry of the University Hospitals of the Cleveland/Case Western Reserve University Alzheimer Center. PARTICIPANTS: Study subjects were 86 probable AD cases (index cases) and all of their full siblings, alive or dead, aged > or = 50 years, a total of 238 subjects. MEASUREMENTS: Exposure for each individual was ascertained with a questionnaire answered by several informants. Cognitive status of siblings of the AD cases (impaired or intact) was ascertained by telephone using validated instruments. Diagnosis of dementia in cognitively impaired siblings of index cases was not attempted. RESULTS: Chi-square analysis tested for departure from a random distribution of disease across smokers and nonsmokers within families. No significant departure was found utilizing all families (P > .40) nor utilizing those families where only the index case was affected (P > .90). Conditional logistic regression evaluated the association within families, controlling for age, sex, and education. Analyses both included and excluded secondary cases of cognitive impairment. No association was found between smoking and disease (OR = 1.26; 95% confidence interval, 0.66-2.42, and OR = 1.42; 95% confidence interval, 0.69-2.89, respectively). CONCLUSIONS: No significant association was found between smoking and cognitive status. Further, analyses based on the comparison of persons with AD with their unaffected siblings also suggest that smoking does not decrease the risk of AD.

Age effects on random-array letter cancellation tests.

OBJECTIVE: This study was designed to determine whether young and older adults differ in the spatial pattern of omission errors on random-array letter cancellation tasks. BACKGROUND: Aging is associated with declines in the speed or efficiency of visual information processing. It is unclear whether the spatial characteristics of visual exploration also change with aging. METHOD: Thirty young adults and 30 older adults each completed 21 random-array cancellation forms. Forms were systematically varied in paper size, target-to-distractor ratio, stimulus density, and target number. RESULTS: The spatial distribution of errors was not random for older adults. Younger adults expressed a trend toward nonrandom error location, but the spatial distribution did not differ between groups. There was also a strong trend toward more errors per subject in the older group. Older subjects required more time for task completion. CONCLUSIONS: The findings are consistent with a generalized age-related decline in the speed or efficiency of visual search, but the spatial properties of directed attention do not appear to be different between young and older adults.

Validation of a Telephone Cognitive Assessment Battery.

OBJECTIVE: To present and evaluate an instrument, the Telephone Cognitive Assessment Battery (TCAB), designed to be administered over the telephone to assess the cognitive status of older individuals. The TCAB addresses mental status, reasoning and executive ability, primary and secondary memory, and language. It consists of six neuropsychological tests and takes approximately 15 to 20 minutes to complete. DESIGN: The instrument is evaluated with a comparative cross-sectional design, with data collected both prospectively and retrospectively. SETTING: The University Hospitals of Cleveland/Case Western Reserve University Alzheimer Center Research Registry. PARTICIPANTS: Forty Alzheimer's Disease cases selected from among those most recently recruited into the Registry and 40 cognitively intact Registry controls. Controls were selected randomly so that the two groups had similar distributions of age, sex, and education. MEASUREMENTS: The cognitive status of all participants was assessed utilizing both the TCAB and the usual in-person Registry evaluation, which includes medical history data and in-person assessment of cognitive status. In order to measure the potential learning effect of repeated testing, half of the cases and half of the controls were recruited and assessed over the telephone with the TCAB before their in-person Registry evaluation (with a waiting period of at least 2 weeks between evaluations), whereas the other two halves received the TCAB after they had become part of the Registry. The TCAB was administered to all participants by a single investigator. Two clinical evaluators, blinded to the Registry diagnosis of the subjects, independently classified the subjects as cognitively impaired, normal, or questionable on the basis of the results of the TCAB and a brief listing of medical illness and depressive symptoms. A final classification was achieved through consensus and subsequently compared with the Registry diagnosis, taken here to be the gold standard. RESULTS: Test scores of subjects assessed by TCAB before receiving the in-person assessment were compared with those of subjects receiving the in-person assessment first. There were no significant differences between mean scores of the two groups (those with TCAB first and those with TCAB last) for either cases or normal controls. High values of the kappa statistic were obtained for the two initial evaluators of the TCAB classification, demonstrating excellent interrater reliability. Regarding the reconciled TCAB classification, the ability of the TCAB to correctly classify subjects according to cognitive status, while controlling for potential confounders such as age and educational level, was assessed by means of discriminant analysis techniques. Knowledge of the TCAB classification and age allowed the correct classification of 95% of the participants; this was not significantly improved by knowledge of other potential determinants. Sensitivity and specificity were calculated under two schema for classifying those subjects in the "questionable" category. Positive and negative predictive values of the TCAB were computed assuming a prevalence of cognitive impairment of 10% in the older population. High negative predictive values (over 99%) were obtained under both schema, whereas the positive predictive values were seen to be more dependent on the classification of questionables. CONCLUSION: Research studies involving ascertainment of cognitive status of older people, particularly those that require periodic follow-up, such as those focusing on healthy aging, commonly suffer from lack of representativeness of subjects, often brought about by problems related to mobility of potential participants. It is also crucial that normal individuals who are recruited initially to serve as controls in epidemiologic studies of dementing illnesses be reevaluated periodically, and this may be hindered by the same obstacles. (ABSTRACT TRUNCATED)