Age at menarche: Accuracy of recall after thirty-nine years.

Abstract

Among 143 women whose menarcheal age was documented during a longitudinal growth study, recall 39 years after the event gave the following results: menarche was recalled as 0.2 years earlier than the actual date (p<0.05), the standard deviation of recalled age was 0.3 years larger (p < 0.01), and the coefficient of correlation, r, between actual and recalled age was 0.60 ± s.e. 0.05. Accuracy of recalled age at menarche is of considerable clinical and scientific moment. There have been two reports to date of women whose ages at menarche were documented during the course of a longitudinal growth study, and who were queried some years thereafter. Livson and McNeill (1962) found, among 43 women questioned after 17 years, no differences in the means or variances of recalled and actual ages at menarche, which were correlated with an r of 0.75. Damon et al. (1969) obtained similar results for 60 women questioned some 19 years after menarche, with an r of 0.78. We report here the accuracy of recall for 143 white women whose menarcheal ages were documented during a growth study between 1922 and 1934, and who were queried, on the average, 39 years after menarche. Subjects and Methods In 1922, W. F. Dearborn of Harvard Graduate School of Education began a study of all children entering the first grade of three Boston suburbs. Each September for the next 12 years, or as long as they remained in these school systems, standard physical and mental exami1 Department of Anthropology and Center for Population Studies, Harvard University, Cambridge, Massachusetts 02138. Dr. Damon died on July 19, 1973. 2 Department of Biology, Grand Valley State College, Allendale, Mich. 49401. Human Biology , September 1974 , Vol. 46 , No. 3, pp. 381-384. © Wayne State University Press, 1974 This content downloaded from 207.46.13.21 on Tue, 27 Sep 2016 04:57:38 UTC All use subject to http://about.jstor.org/terms 382 Albert Damon and Carl Jay Bajema nations were performed on the children. The data have been published in detail, including ages at menarche ( in months ) for some of the girls in two of the three towns (Dearborn et al. 1938). In 1967 and 1968 we followed up, by questionnaire, as many of the survivors as possible. Women were asked, among other things, the age in months at which menarche had occurred. This question was answered by 143 women whose menarchael ages had been recorded at a mean age of 12.98 ± s.e. 0.08 years (S.D. = 1.0 ± s.e. 0.06 years). Their mean age at the time of query was 52.0 ± s.e. 0.07 years. Table 1 Accuracy of Age of Menarche, Recalled after 39 Years. Difference Between Recalled Age and Actual Age at Menarche of 143 Females in the Third Harvard Growth Study Difference in Years Number (recalled age Reporting minus Such a Per Cent actual age ) * Difference of Sample -4 1 0.7 -3 3 2.1 -2 11 7.7 -1 29 20.3 0 72 50.4 + 1 22 15.4 +2 2 1.4 +3 3 2.1 Total 143 100.0 * The integral row headings represent the midpoints of the class intervals -3.50 to -4.49, -2.50 to -3.49, etc. The data base from which the present subjects were drawn was as follows. Of 3,592 participants in the study, we tried to locate 3,198 who had been observed annually for as long as they remained in the school systems, up to a maximum of 12 years. Of these, 1,222 were alive in 1968 and returned questionnaires; 204 were deceased; 849 did not respond; and 923 could not be traced with the limited time and funds available. Of the 1,222 persons returning questionnaires, 613 were This content downloaded from 207.46.13.21 on Tue, 27 Sep 2016 04:57:38 UTC All use subject to http://about.jstor.org/terms