Elsevier

Maturitas

Volume 27, Issue 2, June 1997, Pages 117-123
Maturitas

Validity and reproducibility of self-reported age at menopause in women participating in the DOM-project

https://doi.org/10.1016/S0378-5122(97)01122-5Get rights and content

Abstract

Objectives: To study validity and reproducibility of self reported age at menopause. Methods: Subjects were 1003 and 4892 Dutch women respectively aged 58–73, who participated in a population-based breast cancer screening project. The median time since menopause was 7 years for the validity study. The time span between the two questionnaires in the reproducibility study was 7–9 years. Results: Of the women with a natural menopause, 70% recalled their age at menopause correctly to within one year. For women with a surgical menopause this percentage was 80%. The validity decreased with increasing number of years since menopause. Reproducibility to within one year was 71% for women with a natural menopause and 79% for women with a surgical menopause; 95% of the women were consistent in reporting whether they had had a natural or a surgical menopause. Conclusions: As a consequence of this misclassification, the effect of age at menopause may be underestimated in studies relating self reported age at menopause to disease occurrence or mortality.

Introduction

In historical perspective there has been a shift from descriptive studies of age at menopause [1]to studies that related age at menopause to the occurrence of illnesses, such as breast cancer [2], osteoporosis [3]and especially coronary heart disease [4]. The strength of the estimated effects of age at menopause may depend on the validity of determined age at menopause.

Age at menopause can be determined by three different methods: prospectively, retrospectively (by recall) and by the status quo method. In this last method women are categorized according to their chronological age in one-year categories. Women in each category are asked if they still have menses or not. By plotting the no/yes ratio against chronological age, the median age at menopause on a population level can be determined. The individual age at menopause can be determined by the other two methods. However, the prospective determination of age at menopause can be very time-consuming, and, if large numbers are involved, very expensive. Therefore age at menopause is often determined retrospectively by interview or questionnaire. Accurate recall of age at menopause is hindered by the fact that at the time a woman has her last menstrual period, she does not know yet that it is her last period. She can only tell so in retrospect, after she has had no menses for a year or more.

Data on the validity of self reported age at menopause are scarce. The validity of self reported age at menopause has been studied in a small population of well educated American women [5]. The effect of errors in recall on breast cancer risk has been estimated by Paganini-Hill et al. [2]. The reproducibility of self reported age at menopause has previously been studied mainly in American women 6, 7, 8. Reproducibility to within one year in the study by Colditz et al. [8]was considerably higher (82–95%) than in the studies by Paganini-Hill and Ross [6](48%) and Horwitz and Yu [7](60%). Reproducibility and validity may vary between studies depending on time span between interviews (for reproducibility), time span between interview and last menses, culture and education.

In this study, we present data on the validity and reproducibility of retrospectively determined age at menopause in Dutch women participating in a population-based breast cancer screening project. The reproducibility of type of menopause (natural or artificial) is also addressed.

Section snippets

Subjects and methods

From 1974 to 1977 a total of 14 697 women, born between 1911 and 1925 and living in the city of Utrecht, participated in the DOM (Doorlopend Onderzoek Morbiditeit/Mortaliteit) project for the early detection of breast cancer. These women were invited for a second screening round in 1976–1978. The women participating in the second screening round (n=12 239) were invited for a third screening round. Only those women that responded to an invitation were invited for the next screening round. A

Validity of age at menopause

Of the 1231 women who were still menstruating at first screening, seven did not know their age at menopause at fifth screening. The date of last menstruation of 221 women could not be ascertained, because the last reported date of last menstruation was less than 12 months before the fourth questionnaire (n=187), or because menstruation started again after an absence of more than 12 months (n=34). Age at menopause could be calculated based on last reported date of menstruation for 1003 women.

Age at menopause

The main findings of this study are that the validity and the reproducibility of age at menopause to within one year are both 72%. The validity of self reported age at menopause declined quite rapidly with increasing number of years since menopause.

With respect to reproducibility, after 7–9 years the women underestimated their age at menopause by 0.3 year, irrespective of their age. Women who had undergone surgical menopause recalled their age at menopause better than women with natural

Acknowledgements

The author thanks Dr P.H.M. Peeters (Department of Epidemiology, University of Utrecht) and Professor Dr E.R. te Velde (Department of Gynaecology and Obstetrics, Academic Hospital Utrecht) for valuable comments. Mr P.A.J. Timmermans is thanked for statistical analyses.

References (17)

  • World Health Organization Technical Report Series 670. Research on the Menopause. Geneva: World Health Organization,...
  • Paganini-Hill A, Krailo MD, Pike MC. Age at natural menopause and breast cancer risk: the effect of errors in recall....
  • Luisetto G, Zangari M, Bottega F, Peccolo F, Galuppo P, Nardi A, Ziliotto D. Different rates of forearm bone loss in...
  • Schouw van der YT, Graaf van der Y, Steyerberg EW, Eijkemans MJC, Banga JD. Age at menopause as a risk factor for...
  • Bean JA, Leeper JD, Wallace RB, Sherman BM, Jagger H. Variations in the reporting of menstrual histories. Am J...
  • Paganini-Hill A, Ross RK. Reliability of recall of drug usage and other health-related information. Am J Epidemiol...
  • Horwitz RI, Yu EC. Problems and proposals for interview data in epidemiological research. Int J Epidemiol 1985; 14:...
  • Colditz GA, Stampfer MJ, Willett WC, Stason WB, Rosner B, Hennekens CH, Speizer FE. Reproducibility and validity of...
There are more references available in the full text version of this article.

Cited by (0)

1

Address for correspondence: Section of Reproductive Medicine, Department of Gynaecology and Obstetrics, Academic Hospital Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands.

View full text