| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
|
|
|
|||||||
|
|||||||||
a Department of Neurology, Oregon Health Sciences University, Portland
b Veterans Affairs Medical Center, Portland, Oregon
Jeri S. Janowsky, Department of Neurology CR131, Oregon Health Sciences University, 3181 SW Sam Jackson Park Rd., Portland, OR 97201-3098 E-mail: janowskj{at}ohsu.edu.
Decision Editor: William B. Ershler, MD
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Methods. Reproductive history data from informants of Alzheimer's patients were modeled by comparing responses from 40 cognitively healthy older women with that of a designated informant. The designated informants were similar in demographics to informants for patients with Alzheimer's disease.
Results. Informant data regarding reproductive history was likely to be accurate, when known. However, 30% of the subjects did not identify an informant who had personal knowledge of them. Of those informants who had personal knowledge of the subject, accuracy for those who reported that they knew the information varied depending on the aspect of reproductive history assessed (age of menarche, 29%; age of menopause, 20%; pregnancies, 63%; live births, 92%; hysterectomy, 92%; and postmenopausal estrogen use, 82%). Daughters served as the most likely and most accurate informants in this study.
Conclusion. This study demonstrates that information obtained from informants for patients with Alzheimer's disease is likely to be accurate for some but not all aspects of reproductive history. Of concern for such studies will be the 30% of patients who do not have an informant with personal knowledge about them.
THERE is recent interest in the menopausal period and in postmenopausal estrogen replacement as it relates to the risk for Alzheimer's disease (AD) in women (1)(2)(3)(4). Specifically, studies suggest that postmenopausal estrogen replacement may decrease the risk for AD in women (5)(6)(7). Data for most studies regarding the estrogen and reproductive history of AD patients are derived from informant interviews because the patient is often an unreliable historian. Informants are usually, but not always, caregivers and family members of the Alzheimer's patient. It is not clear, however, how accurate and reliable informants are with regard to historical information on the reproductive history of patients with AD. For some medical conditions, examination of medical records has been used to cross-validate informant or patient recall (8)(9). Medical record data, however, are often incomplete, particularly for remote time periods, and it is extremely time consuming to collect such information (10). Information on normal events in development, such as age of menses and menopause, is often not available in the medical record unless it was brought to a physician's attention. Genetic studies most often use family history of AD information obtained from family members (10)(11). Studies have found that cross-validation from multiple family members is required to assure accuracy (10). Other forms of cross-validating data regarding AD have included informant studies on diagnosis (12), telephone versus standardized screening instruments (13)(14), and studies where informants cross-validate information from the patient or from other informants to assess depression in patients with AD (15)(16).
Studies of depression find that caregivers and physicians rate patients with and without dementia similarly and that they rate patients as more depressed than the patient does (15)(16)(17). Studies examining cross-validation between telephone and clinical assessment measures have been used to find ways to more efficiently screen people for a more detailed cognitive and neurologic assessment (13)(14). In all of these examples, the "gold standard," that is, validation from the patient, is not available, either because the patient is no longer alive or is not an able historian. One way to investigate the knowledge of informants would be to model the same data-gathering method in healthy subjects. In this case healthy subjects give their own history, and the same data are obtained from an informant designated by the healthy subject. To our knowledge there are no studies related to AD that have examined informant accuracy by a cross-investigation of similar but healthy subjects and their informants.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Data-Collection Procedure
Index subjects responded to a questionnaire or phone interview that included questions regarding age at menarche and menopause, and a phone interview regarding postmenopausal estrogen replacement. Twenty-eight informants answered the same questions by phone interview. The 12 informants who were not contacted were those who were unavailable or had only a peripheral relationship to the index subject (friend, neighbor, or other). Informants were encouraged to respond to all questions, but "don't know" was allowed as a response. The informants did not confer with the index subject before responding to the questions. Accuracy in informant reports was considered as being full agreement with the index data unless otherwise stated. The range of responses is also shown in Table 2 . Data was analyzed in StatView 5.0, using t tests to compare differences in means between subject and informant answers for continuous variables.
|
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
All informants reported knowledge of the number of pregnancies and live births of the index subjects (1 index subject was missing data for pregnancies, and 2 were missing data for live births). Sixty-three percent of informants accurately reported the number of pregnancies, and the informants' mean responses were not significantly different from the subjects' mean responses (mean number of 2.7 and 2.4, respectively; T[26] = 1.10, p = .28). Table 2 shows that even inaccurate estimates by the informants were quite close to the actual number of pregnancies reported by the index subjects. Ninety-two percent of the informants accurately reported the number of live births, and there was no significant difference between informant and subject response [mean of 2.08 (SE = .24) and 2.0 (SE = .23), respectively; T[26] = 0, p = 1]. Similarly, Table 2 shows excellent accuracy. The relationship of the informant to the index subject did not affect accuracy for knowledge of the number of live births, but nieces, sisters, and sisters in law were much less accurate in reporting the number of pregnancies as compared with other informants.
Twenty-four of 26 informants (2 index subjects had missing data) reported knowledge of whether their index subjects had had a hysterectomy. Ninety-two percent were accurate in their knowledge. The informant's relationship to the index subject did not affect the accuracy of the hysterectomy information.
Estrogen Replacement
As displayed in Table 1 , 82% (n = 23) of the informants reported whether or not the subject had taken postmenopausal estrogen replacement in the past. Those who reported that they knew this information were 100% accurate, with 39% reporting that the index subject was on estrogen replacement when she was and 43% reporting that the index subject was not on estrogen replacement when she was not. Eighty-two percent of informants also accurately reported whether index subjects were currently taking estrogen supplementation. Overall, 18% of the available informants did not have knowledge of the index subjects' postmenopausal estrogen history. All husbands, sisters, and nieces, and 15 of 17 daughters reported knowledge of current estrogen replacement, while only 1 of 4 sons reported knowing this information.
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Seventy-five percent of informants did not know when their index subjects began menses, and almost 40% did not know their index subject's age at menopause. Most informants reported knowing the number of pregnancies, number of live births, and whether index subjects had hysterectomies, with a good level of accuracy (at 63%, 92%, and 92%, respectively). Over 80% of the informants reported knowledge concerning estrogen replacement and were 100% accurate in that knowledge.
Informant information varies somewhat by relationship with the index subject. Overall, daughters reported and had the most accurate information concerning elderly women's reproductive and estrogen history. Other female relations (sisters, sisters-in-law, and nieces) and husbands also served well as informants in this study. Of all informants, sons were the least likely to report knowledge of index subjects' estrogen and reproduction history. This is unlikely to be a major concern because daughters are usually the primary caregivers for elderly women with AD (21).
The informants in this study had the same characteristics as likely informants for patients with AD (21)(22). How to recover or estimate medically important histories for the 30% of women without close informants is problematic for AD research. However, these data suggest that the estrogen and reproductive history of elderly women who are unable to be their own historians can, for most measures, be reliably gathered from informants who are closely related to those women, especially daughters.
|
Acknowledgments |
|---|
Received August 6, 1999
Accepted March 29, 2000
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. E. Morley and H. M. Perry III Androgens and Women at the Menopause and Beyond J. Gerontol. A Biol. Sci. Med. Sci., May 1, 2003; 58(5): M409 - 416. [Full Text] [PDF]
|
|
|
|
|
|
W. A. Banks and J. E. Morley Memories Are Made of This: Recent Advances in Understanding Cognitive Impairments and Dementia J. Gerontol. A Biol. Sci. Med. Sci., April 1, 2003; 58(4): M314 - 321. [Full Text] [PDF]
|
|
|
|
|
|
S. Asthana Estrogen and Cognition: The Story So Far J. Gerontol. A Biol. Sci. Med. Sci., April 1, 2003; 58(4): M322 - 323. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| All GSA journals | The Gerontologist |
| Journals of Gerontology Series B: Psychological Sciences and Social Sciences | |
