Institutional members access full text with Ovid®

Health-Related Quality of Life in a Multiethnic Sample of Middle-Aged Women: Study of Womens Health Across the Nation (SWAN)

Avis, Nancy E. PhD*; Ory, Marcia PhD†; Matthews, Karen A. PhD‡; Schocken, Miriam PhD§; Bromberger, Joyce PhD∥; Colvin, Alicia MPHpar;

Original Articles

Background. Relatively little is known about the association between menopause and health-related quality of life (HRQL) across ethnic groups.

Objectives. To examine the association between HRQL and early perimenopause and ethnicity, adjusting for health, lifestyle, psychosocial, and sociodemographic factors.

Research Design. Questionnaires were administered to pre- and early perimenopausal women.

Subjects. We studied a cohort of 3302 black, Chinese, Hispanic, Japanese, and white women aged 42 to 52 years from the multisite Study of Women’s Health Across the Nation (SWAN).

Measures. We measured HRQL, menstrual regularity, and a variety of covariates. HRQL was assessed with 5 subscales from the Short Form-36; impaired functioning was defined as being in the 25% most impaired on a subscale.

Results. In unadjusted, but not adjusted, analyses, significantly more early perimenopausal women, as compared with premenopausal women, were classified as having impaired functioning on each of the 5 subscales. For 4 of the subscales, the effect of menopausal status was explained by menopause-related symptoms. There were significant ethnic group differences across all 5 subscales in unadjusted analyses. Ethnicity was no longer significant for the Vitality or Role–Emotional subscales when adjusted for health variables or for the Role–Physical subscale when analyses were adjusted for socioeconomic status, health, lifestyle, or social circumstances. Ethnicity remained significant for the Bodily Pain and Social Functioning subscales, even in adjusted analyses.

Conclusions. Early perimenopause is not associated with impaired functioning when adjusted for symptoms. Significant ethnic differences in HRQL exist. Some, but not all, differences can be explained by differences in health, lifestyle, and social circumstances.

Health-related quality of life (HRQL) has emerged as an important outcome in assessing disease progression and evaluating the effectiveness of clinical trials. HRQL generally denotes aspects of life most likely to be affected by changes in health status 1 and is generally viewed as multidimensional and consisting of the following domains: physical health and functioning, emotional functioning, role limitations, and social functioning. The majority of research in this area has focused on understanding the impact of disease on HRQL and is conducted on specific population groups generally characterized by physical diseases and disabilities. 2 Less is known about HRQL in nonclinical populations and about HRQL across different ethnic groups. 3 The conceptual framework developed by Wilson and Cleary 4 provides a useful way to view HRQL. According to this model, HRQL can be influenced by biologic and physiological variables, symptoms, and functional status. Characteristics of the individual such as personality and psychologic status and characteristics of the environment such as stress, economic, and social supports can influence symptom and functional status as well as HRQL directly.

Interest in the association between menopause and HRQL has recently increased. Menopause and its associated symptoms are often thought to have a negative impact on quality of life. 5,6 However, most studies of menopause have viewed quality of life rather narrowly by focusing largely on negative symptoms of menopause (ie, equating symptoms with quality of life) and/or have been patient- or clinic-based. 5,7,8 Nonclinic samples of women have not generally demonstrated a significant impact of menopause on well-being or mood, 9–12 although some studies report an increase in negative symptomatology during perimenopause. 13–16

A few studies have looked at broader quality-of-life outcomes among menopausal women. 6,17–20 However, except for Ledésert et al., 6 these studies focus only on overall life satisfaction or well-being and not specific HRQL domains. Using the Nottingham Health Profile (NHP), Ledésert 6 found that postmenopausal women scored significantly worse than premenopausal women on social isolation, pain, sleep, and energy domains of the NHP. However, after adjusting for age and symptoms, pre- and postmenopausal women did not significantly differ.

The Medical Outcomes Short-Form 36 (SF-36) is a commonly used measure to assess health status or HRQL. Although the SF-36 has been translated into many languages and used in numerous countries, little data exist comparing domain scores across ethnic groups. 3 Those studies that do make such comparisons tend to find better HRQL among Asians, 21,22 even controlling for sociodemographic factors. Blacks and whites do not differ after adjusting for sociodemographic factors. 23 However, these studies are limited by small numbers in some ethnic groups and control for only a small number of variables.

Studies of varying age and nonpatient groups have examined cross-sectional predictors of HRQL domains. Particularly strong predictors of HRQL include socioeconomic status, 18 depressive symptomatology, 24,25 negative health behaviors, 26,27 psychologic predispositions, 4 and social circumstances such as adverse psychosocial work conditions 28 and social support and living arrangements. 29–31 Because these variables can differ across ethnic groups, 32–35 they could well explain ethnic differences in HRQL.

This article examines HRQL domains of the SF-36 among women participating in the multiethnic Study of Women’s Health Across the Nation (SWAN). All women were either pre- or early perimenopausal. Because SWAN recruited women from the general population, we expected the majority of women to be healthy and SF-36 scores to be highly skewed toward good functioning. 36 We therefore focused on women who had impaired functioning, defined as being at the most impaired quartile of the SWAN distribution for each SF-36 domain. 36

The research objectives of the present study address the following questions: 1) Is early perimenopause associated with impaired HRQL? 2) Is ethnicity associated with impaired HRQL? 3) If ethnicity is associated with impaired HRQL, can this be explained by socioeconomic status indicators, disease burden, symptoms, lifestyle, or social circumstances? 4) If perimenopause is associated with impaired functioning, can this be explained by menopause-related symptoms? Based on the Wilson and Cleary model suggesting that symptoms are an important mediator between a physiological condition or state (ie, menopause) and HRQL, we hypothesize that early peri- menopause is not associated with impaired HRQL when analyses adjust for menopause-related symptoms.

From *Wake Forest University School of Medicine, Department of Public Health Sciences, Winston-Salem, North Carolina.

From †Texas A&M University System, School of Rural Public Health, College Station, Texas.

From the ‡University of Pittsburgh Medical Center, Department of Psychiatry, Pittsburgh, Pennsylvania.

From the §University of California at Los Angeles School of Medicine, Division of Geriatrics, Los Angeles, California.

From the par;University of Pittsburgh Medical Center, Graduate School of Public Health, Pittsburgh, Pennsylvania.

The Study of Women’s health Across the Nation (SWAN) was funded by the National Institute on Aging, the National Institute of Nursing Research, and the Office of Research on Women’s Health of the National Institutes of Health. Supplemental funding from National Institute of Mental Health, the National Institute on Child Health and Human Development, the National Center on Complementary and Alternative Medicine, and the Office of AIDS Research is also gratefully acknowledged.

Address correspondence and reprint requests to: Nancy E. Avis, PhD, Wake Forest University School of Medicine, Department of Public Health Sciences, Section on Social Sciences and Health Policy, Piedmont Plaza II, 2nd floor, Winston-Salem, NC 27157-1063. E-mail:

© 2003 Lippincott Williams & Wilkins, Inc.