Secondary Logo

Journal Logo

Effects of Human Immunodeficiency Virus on Protracted Amenorrhea and Ovarian Dysfunction

Cejtin, Helen E. MD1; Kalinowski, Ann PhD2; Bacchetti, Peter PhD2; Taylor, Robert N. MD, PhD2; Watts, D Heather MD3; Kim, Seijeoung PhD1; Massad, L Stewart MD4; Preston-Martin, Susan PhD5; Anastos, Kathryn MD6; Moxley, Michael MD6; Minkoff, Howard L. MD8

doi: 10.1097/01.AOG.0000245442.29969.5c
Original Research

OBJECTIVE: To characterize ovarian failure and prolonged amenorrhea from other causes in women who are both human immunodeficiency virus (HIV) seropositive and seronegative.

METHODS: This was a cohort study nested in the Women’s Interagency HIV Study, a multicenter U.S. study of HIV infection in women. Prolonged amenorrhea was defined as no vaginal bleeding for at least 1 year. A serum follicle stimulating hormone more than 25 milli–International Units/mL and prolonged amenorrhea were used to define ovarian failure. Logistic regressions, χ2, and t tests were performed to estimate relationships between HIV-infection and cofactors with both ovarian failure and amenorrhea from other causes.

RESULTS: Results were available for 1,431 women (1,139 HIV seropositive and 292 seronegative). More than one half of the HIV positive women with prolonged amenorrhea of at least 1 year did not have ovarian failure. When adjusted for age, HIV seropositive women were about three times more likely than seronegative women to have prolonged amenorrhea without ovarian failure. Body mass index, serum albumin, and parity were all negatively associated with ovarian failure in HIV seropositive women.

CONCLUSION: HIV serostatus is associated with prolonged amenorrhea. It is difficult to ascertain whether the cause of prolonged amenorrhea is ovarian in HIV-infected women without additional testing.


Human immunodeficiency virus–seropositive women often have prolonged amenorrhea, and it can be difficult to determine whether the cause is ovarian failure without further testing.

From the 1John H. Stroger Hospital of Cook County, Chicago, Illinois; 2University of California, San Francisco, San Francisco, California; 3National Institute of Child Health and Human Development, Bethesda, Maryland; 4Southern Illinois University, Springfield, Illinois; 5University of Southern California, Los Angeles, California; 6Lincoln Medical & Mental Health Center, Bronx, New York; 7Georgetown University Hospital, Washington DC; and 8Maimonides Medical Center, Brooklyn, New York.

The Women’s Interagency HIV Study is funded by the National Institute of Allergy and Infectious Diseases, with supplemental funding from the National Cancer Institute, the National Institute of Child Health and Human Development, the National Institute on Drug Abuse, and the National Institute of Craniofacial and Dental Research. U01-AI-35004, U01-AI-31834, U01-AI-34994, U01-AI-34989, U01-HD-32632, U01-AI-34993, U01-AI-42590, M01-RR00079, and M01-RR00083.

The authors thank Mardge Cohen, MD, for her contributions to the original study design.

Corresponding author: Helen E. Cejtin, 1935 W. Farwell, Chicago, IL 60626; e-mail:

A woman’s menopausal status determines her need for contraception and her risk for diseases such as osteoporosis1,2 and coronary artery disease.3 Menopause among American women typically occurs between the ages of 48 and 55, with the median between 50 and 52 years.4 A woman is classified as menopausal when she reports 12 months of amenorrhea,5 and typically no further testing is performed to confirm menopause if she is aged older than 40 years. An elevated serum follicle stimulating hormone (FSH) level may be helpful, however, to distinguish menopause from other causes of amenorrhea, such as anovulation from stress or anorexia.6

Follicle stimulating hormone levels vary from less than or equal to 10 milli–International Units/mL for most of the cycle to a peak of up to 25 milli–International Units/mL at midcycle in reproductive age women. Follicle stimulating hormone begins increasing in the perimenopause and is usually above 25 milli–International Units/mL at the time of the cessation of menses.7

Menopause has not been well described in human immunodeficiency virus (HIV)–infected women, and it may be difficult to ascertain in this group for a number of reasons. The symptoms of menopause include vaginal dryness, sleep disturbances, and most commonly, hot flushes. Although no increase in depression has been reported in menopausal women, they may report mood disturbances as well as memory impairment. Several symptoms associated with ovarian failure are prevalent among older HIV-infected women.8 Hot flushes may be secondary to coexisting infections, such as tuberculosis, and the side effects of medications such as efavirenz9 may be confused with the sleep disturbances that characterize menopause. Depression and HIV-related dementia may be confused with climacteric symptoms as well.

Additionally, amenorrhea may be more common among HIV-seropositive women because of associated wasting10 or anovulation.11 Amenorrhea from such central nervous system (CNS) causes may be misinterpreted as menopause if it continues for at least 1 year. A study with 10 HIV-seropositive women who had amenorrhea for more than 90 days found an elevated serum FSH in only two of the 10 women, indicating that amenorrhea of at least 3 months duration often may be secondary to causes other than menopause.12

The purpose of this study was to characterize prolonged amenorrhea from ovarian failure and other causes in women in the Women’s Interagency HIV Study, to evaluate characteristics associated with each, and to estimate if HIV serostatus is a risk factor for amenorrhea of either cause.

Back to Top | Article Outline


The Women’s Interagency HIV Study is a multicenter prospective cohort study of the natural history of HIV infection in women. There were 2,056 seropositive and 569 seronegative at-risk women who were enrolled in the Women’s Interagency HIV Study between October 1994 and November 1995 from six sites (Bronx and Manhattan, NY; Brooklyn, NY; Chicago, IL; Los Angeles, CA; San Francisco Bay area, CA; and Washington DC). Study methods and cohort characteristics have been described in detail elsewhere,13 and only methods relevant to this analysis are presented. Informed consent was obtained from all participants, and the study was approved by the Human Subjects Protection Committees at the cooperating institutions. The study population for this analysis was restricted to women from the cohort who 1) had at least one study visit in 1997, 2) were no older than 55 years of age at that visit, 3) had not become infected with HIV since entry into the Women’s Interagency HIV Study, 4) did not report a pregnancy in the previous 12 months, 5) did not report using hormonal birth control or hormonal replacement therapy in the previous 6 months, 6) had an intact uterus and one or both ovaries, 7) did not report a nipple discharge, 8) did not report being treated for a thyroid condition, and 9) did not report usage of conventional antipsychotics or risperidone. We restricted the study population because FSH assays were only performed on sera collected at a 1997 study visit, and we wanted to exclude women with other known causes of amenorrhea such as pregnancy, thyroid disease, and medication- or tumor-induced hyperprolactinemia. Because we wanted to study the effects HIV may have on ovarian function, we wanted to restrict the population to women already infected with the HIV virus

Participants returned every 6 months for study visits, which included a detailed interview, a physical examination including weight and gynecologic evaluation, a urine pregnancy test, and the collection of blood for laboratory testing and storage in both local and national repositories. Demographic, behavioral, and medical history data were obtained by standardized survey instruments administered by trained interviewers, and questions relevant to menopause included date of last menstrual period and the presence of any vaginal bleeding in the previous 6 months. Patients were asked if they had ever had a Class C, or acquired immunodeficiency syndrome (AIDS)-defining illness, one of the illnesses, such as cervical cancer or Pneumocystis carinii pneumonia, that would give them a diagnosis of AIDS.14 Historical visit data were examined to ensure consistency of key participant responses for inclusion or exclusion in the study. Additional questions elicited education level, income level, drug use (prescription, over-the-counter, or recreational or illicit), and number of live births. Illicit drug use includes among other drugs use of cocaine, crack, heroin, other opiates, and injection drug use. Opiate use includes the use of any opioid substance, whether illicit or prescribed. In collecting ethnicity data, the “African American” and “White” categories excluded anyone of Hispanic ethnicity. Hispanic women were included with, and made up 90% of the “Other” group. At each study visit, a woman was considered pregnant only if her pregnancy test at that visit was positive, or if she was considered obviously pregnant by both herself and the interviewer. In both HIV seropositive and seronegative women wasting was defined as unintentional weight loss of at least 10% of body weight in the previous 6 months.

Stored plasma and sera obtained during a 1997 visit were used for all the analyses. The 1997 visit was chosen for study so that there would be a sizeable number of women not on highly active antiretroviral therapy, and a broader range of HIV RNA and CD4 cell counts in the cohort. Levels of HIV RNA in plasma were measured by an isothermal nucleic acid sequence–based amplification method (Organon Teknika Corp., Durham, NC), with a lower limit of quantification of 80, 200, or 4,000 copies/mL. T-cell subsets were determined locally by immunofluorescence using flow cytometry. Both HIV viral load and flow cytometry were done in laboratories participating in the Division of AIDS (National Institute for Allergy and Infectious Diseases) Quality Assurance Programs.

Assays for FSH were all performed in the same laboratory on the stored sera by Quest Diagnostics Incorporated, using the Advia Centaur analyzer (Bayer Corporation, Norwood, MA). Follicle stimulating hormone was reacted with a liquid-phase polyclonal sheep anti-FSH antibody labeled with acridinium ester. This mixture was then incubated with a mouse monoclonal anti-FSH antibody bound to a solid phase. After removal of liquid phase, bound FSH was measured by chemiluminescence. The assay was standardized against World Health Organization 2nd IS 94/632 reference material.

In this study a woman was classified as having prolonged amenorrhea, defined as no vaginal bleeding in the past 12 months, if she gave a history of no bleeding for the past 6 months on the two sequential semiannual visits before her 1997 visit, or if the date of her last menstrual period was more than a year previous to that visit in the case of a skipped visit. To distinguish menopause from other causes of amenorrhea, menopause was defined as amenorrhea for at least 12 months and a FSH of 25 milli–International Units/mL or more from serum obtained in 1997. A cutoff of 25 milli–International Units/mL for serum FSH was chosen because this is above the peak serum FSH level usually seen in ovulating women.

Human immunodeficiency virus–positive and HIV-negative women were compared for possible differences in demographic characteristics and other indicators using Fisher exact test or likelihood ratio χ2 tests. All statistical tests used were two-tailed, with a statistically significant result defined to be one having a P value of at most 5%. Logistic regression models were used to identify the associations of study outcomes (presence or absence of menopause or prolonged amenorrhea) with proposed risk factors (HIV serostatus, opiate use, smoking history, etc.), while controlling for age. Risk factors that were identified as statistically significant in those logistic regressions were then entered into a further logistic regression controlling for all the factors simultaneously. Those risk factors which still remained statistically significant after being controlled for all the others were retained in the logistic regression model, and those with weaker associations were removed. These analyses were performed for both the entire cohort and for HIV-positive women, to examine which factors were associated with menopause and, among premenopausal women, with amenorrhea. All analyses were performed using the SAS 8 statistical software package (SAS Institute Inc, Cary, NC).

Logistic regression models involving different combinations of variables associated with menopause were analyzed while controlling for age only, and also for age, body mass index (BMI), albumin, and parity, because these were strong predictors. For the presence of amenorrhea, logistic regression was again used to identify risk factors while controlling for age, but later models controlled for albumin and opiate use as well.

For HIV-seropositive women, CD4 lymphocyte counts and viral loads were modeled both as continuous and as categorical variables. We examined the effect of dichotomous detectable compared with nondetectable viral loads and found no effect. Viral loads were modeled to account for detection limits. In our final models, we used the detection limit as though it were the actual value for viral load, modeled log 10 viral load to account for the large spread in values, and again found no effect due to viral load.

Conditions meeting the Centers for Disease Control and Prevention definition of Class C AIDS defining illnesses and antiretroviral therapy use were also examined. Logistic regressions involving different combinations of variables associated with amenorrhea in premenopausal women were also analyzed.

Back to Top | Article Outline


There were 1,431 (1,139 HIV-seropositive and 292 seronegative) women from our cohort who were eligible for this study. Of 1,970 women with at least one study visit in 1997, exclusions from this study were as follows: five were excluded for seroconversion to HIV-positive while in the study, 113 were excluded because they were aged older than 55 years, 156 for removal of either their uterus or adnexa, 81 for pregnancy in the previous 12 months, 154 for use of hormones in the previous 6 months, 16 for nipple discharge, 12 for reported treatment of thyroid conditions, and 2 for reported use of hyperprolactinemia-inducing medications. The population was predominantly African American and low income, with significant use of tobacco and illicit substances (Table 1). In 1997, at the time blood was collected for serum FSH level, 10.7% of seropositive women reported having had Centers for Disease Control and Prevention Class C symptoms, 47.1% had at some time been on highly active antiretroviral therapy, and the mean CD4 cell count was 386 cells/μL. The occurrence of prolonged amenorrhea was associated with HIV infection. A total of 136 women reported no menstruation during the past 12 months; 120 (10.5%) of 1,139 HIV-infected women and 16 (5.5%) of 292 HIV-uninfected women reported protracted amenorrhea (odds ratio [OR] 1.93, 95 % confidence interval [CI] 1.09–3.44, P=.025 adjusted for age). Amenorrhea without ovarian failure was relatively common. Of the 136 women who reported amenorrhea, 67 also had elevated FSH levels; 11 (68.8%) of HIV-uninfected and 56 (46.7%) of HIV-infected women with amenorrhea had FSH levels consistent with menopause as the cause. Human immunodeficiency virus infection had no statistically proven effect on the occurrence of FSH elevation among women with amenorrhea when adjusted for age, but the 95% CI is large, and includes a fairly large effect, (OR 2.34, CI .63–8.75, P=.21).

Table 1

Table 1

Table 1

Table 1

Among the 69 women with prolonged amenorrhea who were not in menopause, five were HIV-seronegative and 64 HIV-seropositive. Human immunodeficiency virus serostatus was significantly associated with prolonged amenorrhea from causes other than menopause (OR 3.16, 95% CI 1.26–7.95, P=.015) when adjusted for age

The median age in the menopausal group of women was 47 years, with a range of 35–55 years. In the premenopausal group of women the median age was 37 years, with a range of 19–53 years. Because age is so closely associated with menopause, all analyses were adjusted for age. We found that parity (OR .80, 95% CI .68–.95, P=.010), BMI (OR .93, 95% CI .88–.98, P=.005), and albumin (OR .52, 95% CI .28–.98, P=.010) were all significantly inversely associated with menopause when controlling simultaneously for age and each other, and therefore all subsequent analyses adjusted for these covariates.

Neither income, education, tobacco, nor use of illicit drugs or opiates was substantially associated with menopause in either the seropositive or seronegative women or the entire cohort (Table 2). Neither current tobacco use nor a history of tobacco use was associated with menopause. Positive HIV serostatus and nonwhite race both seemed to have substantial positive associations with menopause, but these did not reach statistical significance.

Table 2

Table 2

Among HIV-seropositive women, there seemed to be no substantial association between CD4 cell count, viral load, history of a Class C illness, wasting, or use of antiretroviral therapy and menopause when adjusted for age, BMI, albumin, and parity (Table 2).

Among HIV-infected women with normal FSH levels (Table 3), amenorrhea was independently associated with age, opiate use, low albumin, a history of Class C illness, low income, and “White” compared with “Hispanic/other” race (OR 2.17, 95%CI 1.27–3.73, P=.005 per 10 year increase in age, OR 2.93, 95% CI 1.52–5.63, P=.001 for opiates, OR 2.19, 95% CI 1.30–3.69, P=.006 for unit decrease in albumin, OR 3.10, 95% CI 1.54–6.25, P=.002 for history of Class C illness, OR 3.45, 95% CI 1.47–8.10, P=.004 for low income, and OR .222, 95% CI .079–.63, P=.006 for Hispanic/other ethnicity).

Table 3

Table 3

Back to Top | Article Outline


We found a prevalence of prolonged amenorrhea that was greater than expected among seropositive women in our cohort, and it was often secondary to causes other than ovarian failure, most likely from a central cause. Hypothalamic dysfunction resulting in decreased testosterone levels in HIV-seropositive men have been similarly reported in the literature.15–17 Some studies found an increase in amenorrhea among seropositive women even in the absence of wasting or AIDS,18 whereas still others19,20 showed no association between HIV serostatus and amenorrhea. A previous study including some women from our cohort21 estimated the odds ratio of HIV serostatus being associated with prolonged cycles (more than 90 days) to be 1.32 (95% CI 0.68–2.6). This study was different from ours, however, in that it included only women aged younger than 45 years who were menstruating, whereas our study included as well women aged 45 to 55 years, whether reporting menses or amenorrhea. It is possible that HIV infection may affect menstrual function only in perimenopausal women with declining estrogen levels, which would not have been detected by the former study design.

Among HIV seropositive women, amenorrhea not secondary to ovarian causes was associated with a number of factors besides age. These were opiate use, low serum albumin, a history of Class C illness, low income, and non-Hispanic race. Opiate use has been associated with amenorrhea in the literature,22 presumably through a central action. Although among HIV-seropositive women traditional markers of severe disease such as a high viral load and low CD4 lymphocyte count were not significantly associated with amenorrhea, both a history of Class C illness and a low albumin were strongly associated with it. Low albumin levels reflect a poor nutritional status before the occurrence of weight loss, and it is one of the strongest predictors of death in HIV-positive women, independent of CD4 lymphocyte count or serum HIV-1 RNA levels.23,24 As a reflection of protein-energy malnutrition, low albumin levels may reflect a state of conservation of energy and result in the shutdown of ovarian function. Income is related to both general health and nutrition, and income, employment, and education have all been previously associated with menopause.25–27 The inverse association between Hispanic race and amenorrhea may be a reflection of the better nutrition seen among this ethnic group when compared with African-American and white populations.28,29 An association between African-American race and age at menopause, but not amenorrhea, is suggested in some studies, with African Americans experiencing menopause 6–12 months earlier than their white counterparts, at a median age of 49.3 years.30,31

Amenorrhea with FSH evidence of ovarian failure was related to a number of factors. It was not significantly associated with HIV infection when adjusted for age, but the CI for this comparison was very large, and this study was not adequately powered to identify what might be a clinically significant effect. Parity, body mass index, and albumin were all associated with ovarian failure, even after controlling for age. Parity has previously been associated with age at menopause, possibly because nulliparous women may have incessant ovulation, resulting in early depletion of their ovarian follicles.32–34 The basis for the association with body mass index is less clear. It is well-known that adipose tissue contributes to higher levels of circulating estrogens,35 which may augment reproductive functioning in women near the end of their reproductive lives through less depletion of follicles.36 The association of low BMI with menopause has been reported in some studies,37–39 but not in others.25,26,40,41 Still other studies found an association between menopause and being on a weight reduction diet, without an association with BMI per se.30

In our cohort tobacco use was not associated with menopause. The association between tobacco use and menopause is well established in the literature, with smokers going through menopause 1.5 years earlier.40,42,43 Although an association between smoking and menopause did not reach statistical significance in this study, almost three quarters of the cohort have used tobacco in the past or are current users, making an effect difficult to detect.

It is of great importance that a woman understand whether she is menopausal. If she is, not only can she stop using contraception, but she may begin increased efforts at maintaining bone density and decreasing her modifiable risk factors for coronary heart disease. If she resumes vaginal bleeding, a menopausal woman needs a thorough evaluation of her endometrial cavity. Among HIV-seropositive women in our study group with prolonged amenorrhea, less than one half had ovarian failure as determined by an elevated serum FSH, whereas most of the seronegative women with amenorrhea demonstrated an ovarian cause. Even in the seropositive women at least 45 years of age with prolonged amenorrhea, only 74.6% were menopausal, less than the 90% rate of menopause expected in the general population of women in that age group with amenorrhea for at least 12 months.44 On the contrary, 100% of the seronegative women with amenorrhea in that age group in our study were in menopause. Our results suggest that HIV infection is clearly associated with amenorrhea, and that it may be positively associated with ovarian failure in women with amenorrhea. Our study emphasizes that ovarian failure cannot be accurately ascertained in HIV-seropositive women without laboratory confirmation. After excluding other possible causes of amenorrhea, such as pregnancy, thyroid disease, or hyperprolactinemia, serum FSH testing should be performed before it is assumed that such a woman is menopausal.

Back to Top | Article Outline


1. Cummings SR. Epidemiology of osteoporosis in women. In: Korenman SG, editor. The menopause: proceedings of the Serono Symposium on the Menopause, March 1989, Napa, CA. Norwell (MA): Serono Symposium;1990; p. 1–15.
2. Kritz-Silverstein D, Barrett-Connor E. Early menopause, number of reproductive years, and bone mineral density in postmenopausal women. Am J Public Health 1993;83:983–8.
3. Colditz GA, Willett WC, Stampfer MJ, Rosner B, Speizer FE, Hennekens CH. Menopause and the risk of coronary heart disease in women. N Engl J Med 1987;316:1105–10.
4. Speroff L, Fritz MA. Clinical gynecologic endocrinology and infertility. 7th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2005.
5. Research on the menopause in the 1990s. Report of a WHO scientific group. Geneva (Switzerland): World Health Organization, 1996.
6. Cejtin HE. Gynecologic issues in the HIV-infected woman. Obstet Gynecol Clin N Am 2003;30:711–29.
7. Byyny RL, Speroff L. The perimenopause transition: Health maintenance for perimenopausal women. In: Byyny RL, Speroff L, editors. A clinical guide for the care of older women: primary and preventive care. 2nd ed. Baltimore (MD): Williams & Wilkins; 1996. p. 143–60.
8. Clark RA, Cohn SE, Jarek C, Craven KS, Lyons C, Jacobson M, et al. Perimenopausal symptomatology among HIV-infected women at least 40 years of age. J Acquir Immune Defic Syndr 2000;23:99–100.
9. Guidelines for the use of antiretroviral agents in HIV-infected adults and adolescents. Department of Health and Human Services and Henry J. Kaiser Family Foundation [published erratum appears in MMWR Morb Mortal Wkly Rep 1998;47:619]. MMWR Recomm Rep 1998;47(RR-5):43–82.
10. Grinspoon S, Corcoran C, Miller K, Biller BM, Askari H, Wang E, et al. Body composition and endocrine function in women with acquired immunodeficiency syndrome wasting [published erratum appears in J Clin Endocrinol Metab 1997;82:3360.]. J Clin Endocrinol Metab 1997;82:1332–7.
11. Greenblatt RM, Ameli N, Grant RM, Bacchetti P, Taylor RN. Impact of the ovulatory cycle on virologic and immunologic markers in HIV-infected women. J Infect Dis 2000;18:82–90.
12. Clark RA, Mulligan K, Stamenovic E, Chang B, Watts H, Andersen J, et al. Frequency of anovulation and early menopause among women enrolled in selected adult AIDS clinical trials group studies. J Infect Dis 2001;184:1325–7.
13. Barkan SE, Melnick SL, Preston-Martin S, Weber K, Kalish LA, Miotti P, et al. The Women’s Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology 1998;9:117–25.
14. Centers for Disease Control and Prevention. 1993 revised classification for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. MMWR Recomm Rep 1992;41(RR-17):1–19.
15. Raffi F, Brisseau JM, Planchon B, Remi JP, Barrier JH, Grolleau JY. Endocrine function in 98 HIV-infected patients: a prospective study. AIDS 1991;5:729–33.
16. Dobs AS, Few WL 3rd, Blackman MR, Harman SM, Hoover DR, Graham NM. Serum hormones in men with human immunodeficiency–associated wasting. J Clin Endocrinol Metab 1996;81:4108–12.
17. Grunfeld C, Feingold KR. Metabolic disturbances and wasting in the acquired immunodeficiency syndrome. N Engl J Med 1992;327:329–37.
18. Chirgwin KD, Feldman J, Muneyyirci-Delale O, Landesman S, Minkoff H. Menstrual function in human immunodeficiency virus–infected women without acquired immunodeficiency syndrome. J Acquir Immune Defic Syndr Hum Retrovirol 1996;12:489–94.
19. Ellerbrock TV, Wright TC, Bush TJ, Dole P, Brudney K, Chiasson MA. Characteristics of menstruation in women infected with human immunodeficiency virus. Obstet Gynecol 1996;87:1030–4.
20. Shah PN, Smith JR, Wells C, Barton SE, Kitchen VS, Steer PJ. Menstrual symptoms in women infected by the human immunodeficiency virus. Obstet Gynecol 1994;83:397–400.
21. Harlow SD, Schuman P, Cohen M, Ohmit SE, Cu-Uvin S, Lin X, et al. Effect of HIV infection on menstrual cycle length. J Acquir Immune Defic Syndr 2000;24:68–75.
22. Harlow SD, Cohen M, Ohmit SE, Schuman P, Cu-Uvin S, Lin X, et al. Substance use and psychotherapeutic medications: a likely contributor to menstrual disorders in women who are seropositive for human immunodeficiency virus. Am J Obstet Gynecol 2003;188:881–6.
23. Feldman JG, Burns DN, Gange SJ, Bacchetti P, Cohen M, Anastos K, et al. Serum albumin as a predictor of survival in HIV-infected women in the Women’s Interagency HIV Study. AIDS 2000;14:863–70.
24. Feldman JG, Gange SJ, Bacchetti P, Cohen M, Young M, Squires KE, et al. Serum albumin is a powerful predictor of survival among HIV-1-infected women. J Acquir Immune Defic Syndr 2003;33:66–73.
25. Gold EB, Bromberger J, Crawford S, Samuels S, Greendale GA, Harlow SD, et al. Factors associated with age at natural menopause in a multiethnic sample of midlife women. Am J Epidemiology 2001;153:865–74.
26. Luoto R, Kaprio J, Uutela A. Age at natural menopause and sociodemographic status in Finland. Am J Epidemiol 1994;139:64–76.
27. Torgerson DJ, Avenell A, Russell IT, Reid DM. Factors associated with onset of menopause in women aged 45-49. Maturitas 1994;19:83–92.
28. Thompson FE, Midthune D, Subar AF, McNeel T, Berrigan D, Kipnis V. Dietary intake estimates in the National Health Interview Survey, 2000: methodology, results, and interpretation. J Am Diet Assoc 2005;105:352–63.
29. Sharma S, Murphy SP, Wilkens LR, Shen L, Hankin JH, Monroe KR, et al. Adherence to the food guide pyramid recommendations among African Americans and Latinos: results from the Multiethnic Cohort. J Am Diet Assoc 2004;104:1873–7.
30. Bromberger JT, Matthews KA, Kuller LH, Wing RR, Meilahn EN, Plantinga P. Prospective study of the determinants of age at menopause. Am J Epidemiol 1997;145:124–33.
31. MacMahon B, Worcester J. Age at menopause. United States—1960–1962. Vital Health Stat 11 1966;11:1–20.
32. Cramer DW, Xu H, Harlow BL. Does “incessant” ovulation increase risk for early menopause? Am J Obstet Gynecol 1995;172:568–73.
33. Whelan EA, Sandler DP, McConnaughey DR, Weinberg CR. Menstrual and reproductive characteristics and age at natural menopause. Am J Epidemiol 1990;131:625–32.
34. Soberon J, Calderon JJ, Goldzieher JW. Relation of parity to age at menopause. Am J Obstet Gynecol 1966;96:96–100.
35. MacDonald PC, Edman CD, Hemsell DL, Porter JC, Siiteri PK. Effect of obesity on conversion of plasma androstenedione to estrone in postmenopausal women with and without endometrial cancer. Am J Obstet Gynecol 1978;130:448–55.
36. Garner PR. The impact of obesity on reproductive function. Semin Reprod Endocrinol 1990;8:32–43.
37. Palmer JR, Rosenberg L, Wise LA, Horton NJ, Adams-Campbell LL. Onset of natural menopause in African American women. Am J Public Health 2003;93:299–306.
38. Willett W, Stampfer MJ, Bain C, Lipnick R, Speizer FE, Rosner B, et al. Cigarette smoking, relative weight, and menopause. Am J Epidemiol 1983;117:651–8.
39. Sherman BM, Wallace RB, Bean JA, Chang Y, Schlabaugh L. The relationship of menopausal hot flushes to medical and reproductive experience. J Gerontol 1981;36:306–9.
40. Brambilla DJ, McKinlay SM. A prospective study of factors affecting age at menopause. J Clin Epidemiol 1989;42:1031–9.
41. Hardy R, Kuh D, Wadsworth M. Smoking, body mass index, socioeconomic status and the menopausal transition in a British national cohort. Int J Epidemiol 2000;29:845–51.
42. Lindquist O, Bengtsson C. Menopausal age in relation to smoking. Acta Med Scand 1979;205:73–7.
43. Mattison DR, Thorgierssen SS. Smoking and industrial pollution, and their effects on menopause and ovarian cancer. Lancet 1978;1:187–8.
44. Wallace RB, Sherman BM, Bean JA, Treloar AE, Schlabaugh L. Probability of menopause with increasing duration of amenorrhea in middle-aged women. Am J Obstet Gynecol 1979;135:1021–4.
© 2006 The American College of Obstetricians and Gynecologists