Institutional members access full text with Ovid®

Share this article on:

Psychometric properties of a menopausal symptom list

Freeman, Ellen W. PhD1,2; Sammel, Mary D. ScD3; Liu, Li MD, MS4; Martin, Paula MS4


Objective: To determine psychometric properties of a brief menopause symptom list and its sensitivity to menopausal status in a population-based cohort of late reproductive-age women.

Design: A 12-item menopause symptom list (MSL) administered in a cohort of African American and Caucasian women aged 38 to 52 years (N = 350) was psychometrically evaluated. Menopausal status of the cohort was determined by menstrual cycle dates obtained in interviews and participants' daily symptom records. Results of factor analysis were applied to longitudinal assessments of the cohort over a 3-year period. Convergent validity with other standard measures of mood, stress, health, and quality of life was determined.

Results: Internal consistency was found for the MSL items. Item total correlations are reported. Factor analysis identified three dimensions (psychological, somatic, and vasomotor). Multivariate analysis of cohort data over a 3-year interval showed that the menopausal symptoms increased over time (P = 0.0004) and that the identified factors were differentially associated with menopausal status. Psychological symptoms increased in the premenopausal and early transition groups but decreased in the late menopausal-postmenopausal groups (P = 0.0046 for the interaction). Vasomotor symptoms increased in both the early transition and late menopausal-postmenopausal groups (P = 0.0309 and P = 0.0543, respectively). Psychological symptoms (factor 1) had high correlations with other standard symptom measures (Center for Epidemiologic Studies-Depression Scale, r = 0.59; Zung Anxiety Scale, r = 0.65), whereas factors 2 and 3 did not, suggesting that the somatic and vasomotor symptoms were not associated with mood or health problems.

Conclusions: The MSL provides a brief questionnaire with acceptable psychometric properties for assessing three dimensions of menopause-related symptoms and demonstrated sensitivity to menopausal status in a population-based cohort.

More than 80% of women experience physical or psychological symptoms in the years approaching menopause, 1 with varying degrees of severity and disruption in their lives. Whether some or all of these symptoms are associated with hormonal fluctuations or rates of change that occur with ovarian aging remains controversial and lacks consistent information from well-designed scientific studies. There is also no clear understanding of whether the symptoms are unidimensional or whether symptoms such as poor sleep and dysphoric mood may be secondary to vasomotor symptoms. 2,3 Information that elucidates relationships between symptoms and hormone dynamics in the late reproductive years is an important link in women's reproductive health care.

We are conducting a 10-year longitudinal study of ovarian aging that includes study of common menopausal symptoms such as hot flashes, poor sleep, and depressed mood and their associations with hormonal changes. To assess these common menopausal symptoms, we adapted a brief menopause symptom questionnaire from the Kupperman Index 4 to be included in the structured interviews administered to the cohort. We required a short, clinically relevant symptom list but found no brief instrument with known psychometric properties when we initiated our study. We adapted the Kupperman Index because it was brief (and the most widely used measure of menopausal symptoms in clinical studies), although its symptom selection and wording were outmoded and it lacked statistical justification for either summing or weighting item scores or using factor scales. 5

We have now examined psychometric properties of our 12-item menopause symptom list (MSL) in the cohort of African American and Caucasian women ages 38 to 52 years (N = 350). We report here the internal consistency, item correlations, and factor structure of the measure. The factor analysis results were applied to longitudinal assessments of the cohort over a 3-year period to determine the sensitivity of the MSL to menopausal status. The menopausal status of the participants was determined from menstrual cycle dates and daily symptom records reported by the participants. Convergent validity of the MSL with other standard measures of mood, stress, health, and quality of life was also determined.

From the Departments of 1Obstetrics/Gynecology and 2Psychiatry, the 3Center for Clinical Epidemiology and Biostatistics, and the 4Center for Research in Reproduction and Women's Health, University of Pennsylvania, Philadelphia, PA.

Received March 24, 2002; revised and accepted October 9, 2002.

This study was supported by grant RO1-AG-12745 from the National Institutes of Health.

Address reprint requests to Ellen W. Freeman, PhD, Department of Ob/Gyn, 3701 Market Street, Suite 820, Philadelphia, PA 19104-5509, USA. E-mail:

©2003The North American Menopause Society