• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

The Menopausal Transition: a 9-Year Prospective Population-Based Study. The Melbourne Women's Midlife Health Project

Posted on: Sunday, 27 February 2005, 03:00 CST

ABSTRACT

Objectives To describe the natural history of the menopause in Australian-born women. To determine the hormonal changes relating to the menopausal transition (MT) and how these affect quality of life, bone mineral density, body composition, cardiovascular disease (CVD) risk and memory.

Design A 9-year prospective, observational study of a population- based sample of 438 Australian-born women aged 45-55 years at baseline. By the 9th year, the retention rate was 88%. Interviews, blood sampling, menstrual calendars, quality of life and physical measures were taken annually, and bone mineral density was measured bi-annually.

Results The late MT coincides with changes in estradiol, follicle stimulating hormone, and free testosterone index, decreases in bone density and mastalgia, and increases in central adiposity, vasomotor symptoms, insomnia and vaginal dryness. Levels of total testosterone and dehydroepiandrosterone sulfate are unchanged by the MT. An increase in CVD risk was associated with increases in weight and free testosterone index and a decrease in estradiol. Depressed mood is increased by symptoms and by stressors occurring in the MT. Sexual functioning significantly deteriorates with the MT and aging, but relational factors have major effects. Menstrual cycles became more variable and longer closer to the final menstrual period.

Conclusions As hormonal changes during the MT directly or indirectly adversely affect quality of life, body composition and CVD risk, maintenance of health parameters in the premenopausal years is crucial for a healthy postmenopause.

Key words: MENOPAUSE, PROSPECTIVE, HEALTH OUTCOMES, TRANSITION, HORMONES, AUSTRALIAN-BORN WOMEN

INTRODUCTION

The Melbourne Women's Midlife Health Project (MWMHP) commenced in 1991. At this time, very little was known about the normal range of experience of the menopause and the factors that may influence women's experience of this event. Also, the role of menopausal changes in longer-term health outcomes, particularly cardiovascular disease and ostcoporosis, was contentious. In the earlier menopausal population studies, there were problems with selection bias due to inadequacies of population sampling and recall bias due to retrospective study design1. Some samples were derived from menopause-clinic populations and are known to give a distorted picture of the menopause1.

As there were few adequate data regarding Australian women's experiences of the mid-life years, the Melbourne study was developed as a cross-sectional survey and a further longitudinal study of premenopausal women through the menopausal transition. The study would provide a representative sample of women drawn from the general population, with the aim of providing an understanding of the menopausal transition and information on health experiences and risk factors for longer-term health outcomes.

The Melbourne study had methodological features similar to studies in North America2,3. These included use of a community- based sample randomly selected from telephone numbers in a computerized database of the general population, a culturally homogeneous population (Australian-born women), an appropriate age band of 45-55 years, and the use of validated and reliable instruments. The Melbourne research team was allowed access to a questionnaire carefully constructed by Kaufert and McKinlay and used in North American population studies. Important differences in the Melbourne study were the use of a validated well-being questionnaire, an inquiry about sexual functioning and annual physical measures and blood collection.

The Melbourne Project was established at the outset as a collaboration between scientists from a range of disciplines and, over the years, 16 co-investigators have contributed to the project. The questions asked and the methods employed (qualitative and quantitative) reflect the integration of a broad range of approaches (public health, biomedical, psychological, and sociological) together with the views expressed by the women themselves. The overall objective of the Project was to describe the natural history of the menopause in Australian-born women. A number of specific objectives were out-lined which included the provision of quantitative information on changes in parameters of women's health during the menopausal transition; the modelling of ways in which psychological, biological, social and lifestyle variables might interact in their effects on health and behavior; the identification of the characteristics of women who may be at increased risk for specific health disorders; and the assessment of factors relating to the use of health-care services (attendance at health professional, use of hormone and other therapies).

The initial age band of the participants in this project was to provide a cross-sectional sample of women who were premenopausal, and could be invited to participate in the longitudinal study, women who were currently experiencing the normal menopausal transition, postmenopausal women and those who had experienced an induced menopause. Meaningful and standard definitions of the stages of the menopause were required for both the cross-sectional and longitudinal phase analyses. At the cross-sectional phase, the category 'premenopause' was assigned to women who had experienced no changes in menstrual frequency or flow in the prior 12 months, the category 'perimenopause' to women who had changes in menstrual frequency or flow in the prior 12 months, and the category natural 'postmenopause' to women who had experienced 12 months of amenorrhea. The term 'surgical menopause' was used to describe women whose menses had terminated due to either a hysterectomy ( oophorectomy) or an endometrial ablation. However, at this time, no standard definition of the menopause had been accepted by the scientific community4. In 1996, the term 'perimenopause' was described by the WHO Scientific Group as 'the period immediately prior to the menopause and the first year after the final menstrual period5.

The longitudinal phase of the Project was a 9-year follow-up of 438 women who, at the time of their initial telephone interview, had menstruated in the last 3 months (had an intact uterus and at least one ovary) and were not taking hormone therapy (HT) or the oral contraceptive pill. Data from this longitudinal phase of the Project were used to determine which aspects of menstrual change best predict time to postmenopause6. The results of this study suggested a two-stage classification scheme for defining the perimenopause. 'Early perimenopause' was defined as the self-reporting of changes in menstrual frequency over the last year, and 'late perimenopause' was defined as the self-reporting of 3-11 months of amenorrhea.

More recently, a workshop was held to address the absence of a relevant staging system for reproductive aging and to discuss the current nomenclature for the menopause. This workshop recommended that the menopausal transition be divided into two stages, early and late, defined by menstrual cycle and endocrine changes. The menopausal transition begins with variation in menstrual cycle length in a woman who has a monotropic follicle stimulating hormone (FSH) rise and ends with the final menstrual period (FMP), which is not recognized until after 12 months of amenorrhea7. The terms early menopausal transition (EMT) and late menopausal transition (LMT) were proposed; this latter phase is similar to our phase of 3-11 months of amenorrhea. The workshop used the term 'late reproductive' rather than premenopausal to characterize the years immediately prior to the menopausal transition. Further verification of these definitions and their association with hormonal changes is in progress. This workshop also recommended that the terms 'perimenopause' and 'climacteric' be used only with patients and in the lay press and not in scientific papers.

METHODS

The study was approved by the Human Research Ethics Committee of the University of Melbourne and the procedures followed were in accordance with the ethical standards of the National Health & Medical Research Council. All subjects provided written informed consent for their participation in the study.

Research design

The MWMHP is a population-based observational study that included both cross-sectional and longitudinal phases. The first stage was the cross-sectional study of 2001 Australian-born women aged 45-55 years who completed a 20-25-min telephone interview. The second stage was a 9-year follow-up of 438 of these women who, at the time of their initial telephone interview, had menstruated in the last 3 months (had an intact uterus and at least one ovary) and were not taking HT or the oral contraceptive pill. The women in the longitudinal study were interviewed and had physical measurements and blood taken annually in their own homes for 8 years of follow- up. In the 9th year of follow-up, only women who were still menstruating and had never taken HT were interviewed (n = 51). From the 5th to the 8th year of follow-up, women who had experienced a surgical menopause were interviewed by telephone.

Sample derivation

A random sample of households in Melbourne was generated from a d\atabase of the telephone numbers from the Melbourne telephone directory using the computerized random number selection of the Roy Morgan Research Centre Pty Ltd. Each number was called up to five times, with calls being placed during the day and evening, both during the week and at weekends. At the time of this recruitment, approximately 1% of Melbourne residences had no telephone and 13% of Melbourne residential telephone numbers were not listed in directories and their characteristics were unknown. The interviews were completed over a 6-week period (11 April-19 May 1991). There were 2938 women identified as eligible to enter the study, after a total of 54 078 telephone numbers had been called8. Of the 2938 women, 105 were unavailable through either being ill or absent for the duration of the study, and 832 women refused to take part in the study. The response rate among those women eligible and available for the study was 70.6% (2001 of 2833).

Preliminary analysis and/or subsequent contact indicated that, contrary to the subject's statement of eligibility, 39 were too young, 26 were too old, 12 refused to give their date of birth and four were not born in Australia. A further seven subjects were excluded from analysis when they proved unsure of whether they had had a hysterectomy and/or bilateral oophorectomy, thus making the determination of their menopausal status impossible. The same reason lead to the exclusion of 16 subjects who were using the oral contraceptive pill, bringing the number of subjects eligible for analysis to 1897.

From the cross-sectional cohort, 779 women met the eligibility criteria for the longitudinal study. Interviews were completed for 438 of these women, giving a response rate of 56%. Of the 341 women who were not interviewed, 24% had refused further contact at the initial telephone interview, and the remainder were uncontactable or declined to participate when approached for the longitudinal study.

Measures and analyses

Fasting morning blood samples were collected annually. Blood was taken between the 4th and 8th days of the menstrual cycle in cycling women. For those not cycling, blood was taken after 3 months or more of amenorrhea. Estradiol was measured using the double antibody radio-immunoassay (RIA) kit purchased from Diagnostic Products Corporation (DPC), Los Angeles, USA. Serum testosterone was measured by double antibody RIA, following sample extraction and polyethylene glycol-enhanced separation of bound from free ligand, using 125 I- iodinated testosterone as tracer. Sex hormone binding globulin (SHBG) and dehydroepiandrosterone sulfate (DHEAS) were measured by an automated chemiluminescent enzyme immunoassay (Diagnostic Products) using the Immulite Automated Analyser. Free testosterone index (FTI) was calculated as the ratio of measured testosterone to measured SHBG x 100. Levels of total cholesterol and high density lipoprotein (HDL) cholesterol and triglycerides were measured according to standard enzymatic methods on routine automated chemistry systems. The level of low density lipoprotein (LDL) cholesterol was calculated using the Friedewald formula adapted for SI. Plasma glucose was measured using the hexokinase method and performed on Dimension clinical chemistry systems (Dade International Inc., Newark, USA). Insulin was measured by RIA using Linco Human Insulin Specific kits (Linco, Missouri, USA).

The following physical measures were recorded annually: systolic and diastolic blood pressures, body height, body weight, body mass index (BMI) (calculated as weight (kg)/height (m^sup 2^)), waist and hip circumference; four skin-fold thicknesses, triceps, biceps, subscapular and supra-iliac, were measured using Harpenden skin- fold callipers. Bone mineral density and total body composition were measured at 2-yearly intervals by dualenergy X-ray absorptiometry using a Hologic QDR-1000W densitometer.

The core questionnaire included variables relating to education, employment, marital status, parity, well-being9, interpersonal stress, lifestyle factors, general somatic problems, menopausal symptoms, attitudes to aging and menopause, health behaviors, health status (self-rated health, chronic conditions, other health problems), family medical history, prescription and non- prescription medication use, health-care use, medical procedures, reproductive and menstrual histories and problematic premenstrual complaints. Menstrual diaries and supplementary questionnaires for information on sexuality, hassles, physical activity, diet, experience of physical and sexual abuse are documented in detail in the specific references.

Data were analyzed both cross-sectionally and longitudinally. Cross-sectional analyses of the baseline sample of 1897 women enabled the description of following variables according to menopausal status and the estimate of associations with other factors: well-being10, self-rated health11, symptoms12,13, treatment- seeking14, hormone therapy use15, hysterectomy16, tubal ligation17, bone mineral density18, biochemical bone markers19, blood pressure and serum lipids20, sexuality21, physical activity22, hormones23, self-rated menopause status24,25 and incontinence26. Cross- sectional analyses can only indicate whether associations exist and are unable to determine the direction of causality. Also, cross- sectional analyses cannot control for premenopausal characteristics nor separate the effects of aging from those of menopause27. The strength of the MWMHP lies in the prospective arm of the study. The results section of this current report is focused on the longitudinal analyses.

The prospective phase of the MWMHP involved the repeated collection of data or measures from the same individuals followed over time. A number of statistical approaches were utilized and are described in detail in the specific papers. Some analyses involved examining the effect of reaching the FMP and, in this case, analysis of covariance was used after dividing the data into premenopausal/ EMT, and LMT/postmenopausal series and adjusting for the premenopausal values. Another technique was repeated measurement multivariate analysis of variance using a number of contrasts to estimate various effects. Random effects time series regression models were also used and this allowed the separate estimation of the effect of differences between, and the effect of changes within, women with regard to the outcome variable. Structural equation modeling was used for examination in detail of a range of factors that may influence the studied end-point, the presence of feedback and of latent or non-measurable variables. The MWMHP investigators have published discussion papers on the various statistical techniques available for analysis of change in longitudinal studies of the menopause28, methodological issues related to menstrual diary data29, the derivation, evaluation, validity and reliability of a short scale to assess female sexual functioning30-33, and measurement of physical activity in mid-life women34.

RESULTS

Population-based sample

Characteristics of sample

The sociodemographic characteristics of the cross-sectional sample are detailed in Table 1. The sample was broadly representative of the population of Australian-born women of equivalent age with regard to demographic characteristics and health behaviors35,36. Of the 1897 eligible participants (median age, 50 years) in the cross-sectional study, 323 (17%) were initially classified as being premenopausal, 531 (28%) were perimenopausal, 357 (19%) were naturally postmenopausal, 399 (21%) were taking hormone therapy and 420 were surgically menopausal (of whom 133 were current HT users).

Table 1 Sociodemographic characteristics of the cross-sectional sample of Australian-born women (n = 1897), aged 45-55 years (median age 50 years)

Comparing the 438 participants with 341 non-participants23, more of the study participants reported that they:

(1) Had better health than most in comparison with women of the same age (48.5% vs. 38.1%, p < 0.005);

(2) Were in paid employment (71.4% vs. 63.0%, p < 0.05);

(3) Had more than 12 years of education (34.3% vs. 24.3%, p < 0.005);

(4) Had had a Papanicolaou smear in the last year (58.3% vs. 50.0%, p < 0.05);

(5) Exercised at least once a week (68.0% vs. 58.1%, p < 0.005); and

(6) Had undergone dilatation and curettage (46.5% vs. 38.4%, p < 0.05).

Participants did not differ in age, BMI, marital status, parity, symptoms, well-being, interpersonal stress, number of surgical procedures, proportion of current smokers, recent alcohol consumption, house-hold composition, proportion who had breast checks in the last year, proportion who had a tubal ligation, use of medications, treatment for chronic conditions, and suffering from premenstrual complaints.

After 9 years of follow-up, 35 women (8%) had not reached their FMP, 224 women (51%) had experienced the natural menopausal transition and were postmenopausal, 91 women (21%) had used HT prior to their FMP, 37 women (8%) had experienced a surgical menopause and 51 (12%) had dropped out of the study. The women who dropped out were significantly different with respect to two variables only - they were less likely to be married or live with a partner (65% vs. 81%, p < 0.01) or to exercise at least once per week (53% vs. 70%, p < 0.05). At years 3, 6 and 11 of follow-up, the percentages of women who had experienced the natural menopausal transition were 13%, 36% and 52% respectively37. Ten (5%) of the 224 women who had experienced the natural menopause transition reported an unexplained 'menstrual cycle' bleed more than 12 months after their last menstrual period38.

Characteristics of surgical menopause and hormone therapy users

In the cross-sectional cohort, women who had undergone hysterectomy were compared with women who had had a natural menopause16. Hysterectomized women were significantly more likely to use HT, to be of lower \educational level, to report a history of troublesome premenstrual complaints, more dilatation and curettage procedures and non-gynecological operations, and use of prescription medications.

During follow-up, 37 (8%) women from the longitudinal cohort experienced a surgical menopause, 16 had experienced removal of their uterus, 14 had their uterus and both ovaries removed, and seven had had an endometrial ablation. At baseline, comparing these women with the women who did not experience a hysterectomy, they were more likely to rate their health as worse than women of the same age (21% versus 5%, p < 0.005) and less likely to be currently smoking (5% versus 20%, p < 0.05). After 8 years of follow-up, these surgically menopausal women were more symptomatic (an average of 6.3 symptoms reported vs. 4.7 symptoms, p < 0.05) and were more likely to have taken HT during follow-up (65% vs. 49%, p < 0.005). Women who experienced a surgical menopause during follow-up were also more likely to suffer bladder control problems26.

Twenty-one percent of the women in the cross-sectional sample were using HT. Users differed significantly from non-users with respect to a number of variables15. The HT users were significantly more likely to have experienced hot flushes, have a lower body mass index, drink more alcohol, have had a recent Papanicolaou smear, and had their breasts examined by a health practitioner. Women with low positive affect were more likely to be current users of HT, as were women reporting a greater interest in sex within the previous 12 months. Analysis of the long-itudinal cohort found that, at the 6th year of follow-up, and excluding women who were still premcnopausal, 42% of the women had used HT at some stage (ever-users)39. Baseline factors that were significantly associated with ever-use of HT included having a breast check with a health professional, having had a tubal ligation and a history of premenstrual complaints, and taking non-prescription medications39. Ever-users of HT did not appear to differ significantly from non-users with respect to lifestyle, sociodemographic and health status variables. By the end of the 9th year of follow-up, 181 women had used HT at some time, in other words 43% of the eligible longitudinal cohort. The differences between ever-users and non-users after 8 years of follow-up were similar to those found at the 6th follow-up year.

Hormonal changes

Four years before the final menses, the geometric mean (95% confidence interval) of FSH was 10.5 (7.8-14.1) IU/l (Table 2). During the 18 months on either side of the FMP was when most of the changes in FSH were observed. Mean FSH levels rose approximately five-fold over 3 years, from 17.5 (13.5-22.5) IU/l at 18 months before the FMP to 48.4 (39.0-60.1) IU/l at the time of the FMP, and to 100.5 (91.1-110.8) IU/l at 18 months after the FMP. Mean estradiol levels were characterized by large variability in the premenopausal/EMT years; mean levels were 287 (142-581) pmol/l at 4 years prior to FMP and fell by approximately 60% to 113 (83-155) pmol/l at the time of the FMP and to 35 (29-41) pmol/l at 18 months after the FMP. Between 2 and 5 years after the FMP, mean estradiol levels reached the sensitivity level for the assay (20 pmol/l)40.

Table 2 Geometric means (95% confidence level) of follicle stimulating hormone (FSH), estradiol, inhibin, sex hormone binding globulin (SHBG), testosterone, free testosterone index and dehydroepiandrosterone sulfate (DHEAS) across the menopausal transition. Pre-final menstrual period (FMP) levels of FSH, estradiol, SHBG, testosterone, free testosterone index and DHEAS are 4 years before the FMP; inhibin levels relate to 2.5 years before the FMP40,41

Levels of inhibin A (INHA) and inhibin B (INHB) fell during the menopausal transition. Mean levels of INHA were 25.0 (16.9-37.0) ng/ l 2.5 years before FMP and fell by approximately 50% to 13.5 (11.7- 15.4) ng/l at the time of FMP, with a minor fall thereafter. The proportion of INHA values undetectable was 68% at 6 months and 94% at 4 years after FMP. Mean levels of INHB were 48.9 (33.1-72.4) ng/ l at 2.5 years before FMP, fell by about 30% to 35.2 (29.5-42.0) ng/ l at the time of the FMP, and fell by about 40% to 28.5 (26.5-30.8) ng/l at 6 months after the FMP. The proportion of INHB undetectable was 83% at 6 months after the FMP, and 100% by 3.5 years after the FMP40.

Mean SHBG levels decreased during the menopausal transition, the time of maximum change being estimated at 2 years before the FMP. SHBG levels were negatively associated with BMI (p < 0.0001) and positively associated with estradiol levels (p< 0.0001). SHBG levels decreased over the transition by approximately 43% and about one- third of the decline in these levels was explained by adjusting for estradiol and BMI. The mean level of testosterone did not vary across the menopausal transition and was also not related to age in this cohort. The mean value of the free testosterone index increased with time relative to the FMP by 80% over a 6-year period, from 1.5 at 4 years before the FMP to 2.7 at 2 years after the FMP. The time of maximum change was estimated to be 2.2 years before the FMP. The free testosterone index was also positively associated with BMI, with an average increase of 4% for each unit increase in BMI (p < 0.001). The mean level of DHEAS was not related to the FMP, but decreased with increasing age and BMI (both p < 0.05)41.

Menstrual changes leading up to the FMP

Retrospective self-report at interview of changes in menstrual frequency was shown to have low sensitivity to measures based on prospectively kept menstrual diaries42. Data from menstrual diaries29 were used to investigate menstrual patterns and time to final menstrual period. In the premenopausal phase, cycle length rarely varied outside the 21-35-day range and did not show a rising or falling trend. In the phase prior to the FMP, sequences generally became increasingly variable in length, with a steady trend toward a mean cycle length above 35 days. An indicator of the approach of menopause is a rise in 'running range' of cycle lengths to 42 days43 (the 'running range' is the difference, in days, between the shortest and longest cycles in a recorded sequence of menstrual cycles). The time remaining to the FMP can be estimated with an accurate knowledge of the length of a woman's menstrual cycles, FSH levels, and the woman's own perception of where she is in the menopausal transition44. For women over 45 years of age, the time remaining in the menopausal transition from the day on which 6 or more weeks separate the longest and shortest recent cycles is related to self-rating of menopausal status and to serum FSH levels. The median number of months remaining ranges from 11 for those with FSH of greater than 20 IU/l and who see themselves as in transition to 21 months for those with lower FSH and who notice little evidence of being in transition44.

Changes in health outcomes

Quality of life

Symptoms During mid-life, the women in our cohort reported that they were bothered by many symptoms affecting most body systems. The symptoms that appear to be specifically related to the hormonal changes of the menopausal transition are vasomotor symptoms, insomnia, vaginal dryness, and breast tenderness. The reporting of vasomotor symptoms and vaginal dryness increased as women progressed from early to late menopausal transition, whereas breast soreness/ tenderness decreased during this period. Insomnia also increased during the transition and was found to reflect bothersome hot flushes and psychosocial factors45.

The frequency of reporting bothersome hot flushes commences to rise about 2 years before the FMP and reaches a maximum about 2 years after the FMP. There is a gradual decrease in frequency, reaching premenopausal levels after a further 6 years. Approximately 75% of our cohort reported suffering bothersome hot flushes at some time during the transition. From a list of 22 general somatic and menopausal symptoms, women reported an annual mean of four to five bothersome symptoms. The number of women reporting five or more symptoms increased from the early to late menopausal transition45. Eighty-six percent of our cohort consulted a doctor for menopausal problems during the menopausal transition, with an annual mean of 31%. Those who are more symptomatic, with mood or vasomotor symptoms, consulted doctors more often and were more likely to use hormone therapy46.

After 3 years of follow-up, data were analyzed to evaluate whether a history of menstrually related problems (premenstrual complaints) was a significant predictive marker of a more symptomatic menopausal transition experience. Significant relationships were found between a prior history of both physical and psychological premenstrual complaints and a more symptomatic transition characterized by dysphoria, skeletal, digestive and respiratory symptoms47.

Mood Mood was assessed annually using a validated measure of psychological well-being - the Affectometer 2^sup 9^. The Affectometer provides three measures of psychological well-being: a measure of positive affect, a measure of negative affect, and an overall measure of well-being, which was the difference between the positive and negative affect scales. Analysis of data from the first 4 years of the longitudinal study found that well-being was decreased in the first 2 years after the final menstrual period but began to improve after this time48. This analysis used pooled data from the 4 years of follow-up.

Subsequent longitudinal analysis of 8 follow-up years of well- being data, by analysis of covariance on summary statistics, found that, as women passed from early to later in the menopausal transition, negative mood declined significantly, positive mood did not change and well-being improved significantly49. Well-being was also significantly affected by changes in marital status\, work satisfaction, daily hassles and life events.

Another analysis of 6 follow-up years focused on negative mood scores. Repeated measures multivariate analysis of covariance found that negative mood scores decreased significantly over time and were not related to the natural menopausal transition, hormone levels, age or education. The magnitude of negative mood was significantly predicted by baseline reporting of premenstrual complaints, negative attitudes to aging and menopause, and parity of one. During follow- up, the magnitude of negative mood was significantly adversely affected by prior experience of negative mood, experience of bothersome symptoms, poor self-rated health, negative feelings for partner, no partner, current smoking, low exercise, daily hassles, and high stress. Negative mood was reduced by decreasing symptoms, improving health, positive feelings for partner, gaining a partner, and reducing stress. The menopausal transition had an indirect effect in amplifying the effect of reducing paid work, poor health and daily hassles50.

Positive mood scores were analyzed using data from 8 follow-up years. Positive mood remained relatively stable over time and was not related to the natural menopausal transition, age or education. In the early phase of the menopausal transition, positive mood was adversely influenced by baseline interpersonal stress and negative attitudes to aging. In the late menopausal transition and postmenopausally, the factors significantly affecting positive mood were positive mood scores in the premenopause, changes in dysphoric symptoms12, major life events, daily hassles, marital status and work satisfaction51. One factor that was reported, in a separate analysis, as having an affect on mood in mid-life women was the departure of children from or their re-entry to the home. For the majority of women, the departure of the last child from the household leads to positive changes in women's mood state and a reduced number of daily hassles. Return of offspring may have an adverse effect on the sexual relating of parents52.

Sexuality Women's sexuality during the menopausal transition was investigated. Using the Personal Experiences Questionnaire31, six sexual function domains were identified: feelings for partner, sexual responsivity, frequency of sexual activities, libido, partner problems, dyspareunia/vaginal dryness. A total sex score was obtained by summing the scores for sexual responsivity, sexual frequency and libido. A total score of 7 or less was considered to indicate sexual dysfunction. Using structural equation modeling, psychosocial factors, symptoms, and the menopausal transition were all found to affect women's sexual functioning during the mid-life years53.

From early to late menopausal transition, the percentage of women with scores indicating sexual dysfunction increased from 42% to 88%. In the early menopausal transition, women with low total sex scores had lower estradiol levels but similar androgen levels to those with higher scores. Decreasing sexual domain scores correlated with decreasing estradiol levels but not with androgen levels54.

To determine whether changes in women's sexual functioning during mid-life were due to aging or menopause, women who experienced the menopausal transition were compared with two control groups (A and B) of similarly aged women. Control group A consisted of women who remained premcnopausal or who were in the early menopausal transition, and control group B included women who remained postmenopausal over 5 years. Sexual responsivity was adversely affected by both aging and the menopausal transition. The other domains of sexual functioning were significantly adversely affected by the menopausal transition. The relationship with the partner and the woman's perception of his ability to perform sexually were also adversely affected by the menopausal transition55.

Further modeling of sexual functioning found that libido could be classified with responsivity and this new domain was called sexual response. It was also determined that the sexual behavior of a mid- life woman was predominantly determined by her previous behavior, change in partner status and feelings toward her partner. These factors explain more than 50% of the variability on both sexual frequency and response33. Using structural equation modeling, estradiol was found to significantly affect sexual response and dyspareunia.

Self-rated health and life satisfaction These two variables are important components of quality of life. Women were asked annually to rate their present health compared with that of other women about the same age as worse than, the same as, or better than most. Our prospective study found a small decline in self-rated health with age but no significant effect of the menopausal transition. An increase in BMI and a decrease in 'feelings for partner' predicted a change in self-rated health from a baseline status of 'better than most'. Having an operation or procedure in the last year and an increase in the number of symptoms predicted a decline in self- rated health from the baseline status, 'same as others'56.

Two self-reported measures of life satisfaction, Life satisfaction Index-Z scale and Satisfaction with life scale, were measured in year 6 only. Explanatory variables used in this analysis were baseline sociodemographic measures, attitudes towards menopause and aging documented in years 2 and 5, respectively; all variables were measured in year 6. Women overwhelmingly endorsed positive responses to life satisfaction questions. Life satisfaction was unrelated to menopausal status, hormone levels or hormone therapy. Life satisfaction was closely related to mood, predicted by earlier attitudes to aging and menopause, and positively associated with feelings for partner and exercise and negatively associated with daily hassles, interpersonal stress, dysphoric symptoms and current smoking57.

Body composition

Bone mineral density Changes in bone mineral density (BMD), weight, amount and distribution of body fat during the menopausal transition were described and variables associated with these changes determined.

Low premenopausal BMD, a decrease in BMD, and an increase in bone fragility that occur as a result of both aging and the menopause are major determinants of subsequent risk for osteoporotic fracture. In addition, low BMI, low calcium intake, low physical activity, and smoking can affect BMD58. In our cohort, a change in menopausal status from premenopausal/EMT to late menopausal transition was associated with an annual change in BMD of -0.9% at the lumbar spine and -0.7% at the femoral neck. The change to postmenopausal status was associated with a change of -2.5% at the lumbar spine and -1.7% at the femoral neck. All these changes were significantly greater than those in women who remained premenopausal or in the EMT59. The Australian-born cohort had multiple risk factors for osteoporosis, but only anthropometric values (BMI, waist, hip, trunk skin-fold measures) were associated with BMD at baseline. Significant changes during the menopausal transition in anthropometric variables and calcium intake were in the direction that could decrease the risk of osteoporosis but were not found to have an effect on bone loss59,60.

Estradiol was the only endogenous hormone to have a significant effect on BMD during the menopausal transition. An estradiol level of about 240 pmol/l was required for the preservation of BMD during the transition61.

Weight Over 5 years of follow-up, the mean weight gain of the cohort was 2.1 kg. Baseline age was negatively and significantly associated with weight gain. Overall weight gain was not related to change in menopausal status or to any lifestyle factors measured. However, suprailiac skin-fold measurements, waist circumference and waist/hip ratio all increased during the menopausal transition. Continuous hormone therapy users showed no gain in mean weight, suprailiac skin-fold or waist measurements over the follow-up period62. By the 8th year of follow-up, 58% of the cohort had a BMI of 25 or greater and were classified as being overweight.

Central abdominal fat A sample of the cohort (n = 102) had their whole body composition measured using dual-energy X-ray absorptiometry (DXA) in the 6th year of follow-up. Women in the early menopausal transition had significantly lower percentage body fat than women in the later stages of the menopausal transition. The percentage fat in the central abdominal area was estimated from the DXA scans and multiple regression analysis found it was positively associated with current weight, increase in weight since baseline, baseline free testosterone index and with the increase in free testosterone since baseline63.

Cardiovascular health

Longitudinal analysis of change in individual coronary heart disease (CHD) risk factors was completed after 5 years of follow- up. This included HDL cholesterol, LDL cholesterol, triglyceride levels and diastolic blood pressure. These variables were analyzed in relation to the FMP and to other known risk factors for CHD - smoking, alcohol intake, exercise and BMI. HDL cholesterol was the only factor showing a change related to the menopause, decreasing after the menopause, with the greatest rate of decrease occurring toward the end of the first year after the FMP. Changes in the other lipids, blood pressure and BMI were related to increasing age or to simultaneous changes in one of the other risk factors. All changes related to age were deleterious and the effects of decreasing smoking, increasing exercise and moderate alcohol intake were beneficial64. Increases in the level of physical activity were shown in a separate analysis to be positively associated with increases in HDL cholesterol65.

Diabetes is a strong risk factor for CHD and impaired fasting glycemia (IFG) is associated with the risk of developing diabe\tes. Over a 6-year follow-up period, 43 women (16%) of the cohort recorded IFG and an additional six women had been previously diagnosed with diabetes. The women who developed IFG had significantly higher levels of body fat and dyslipidemia, premenopausally compared with women who did not develop IFG. The menopause per se did not have an effect on the development of IFG, but weight gain during this period was associated with an increase in insulin levels66.

Subsequent analysis of CHD risk incorporated a scoring scheme (PROCAM) to estimate the global risk of CHD and data from 8 years of follow-up were used. This analysis found that higher than average BMI and free testosterone level, lower than average estradiol level and an increase in BMI and a decrease in estradiol level during the study period were associated with an increase in CHD risk67. Hormone therapy use did not have a significant effect on percentage risk of an acute coronary event67.

Memory

A study using one memory test was carried out during the 8th year of follow-up. Episodic verbal memory was assessed with a supraspan 10-word list recall task (CERAD). Memory did not vary with any variables that measured estrogen exposures, including menopausal status, time from the FMP, use of hormone therapy or serum estradiol concentration68. Memory was significantly and positively correlated with education and alcohol intake and negatively correlated with negative mood68. A better result in this word-list recall memory test was found to be significantly associated with higher serum concentrations of LDL cholesterol as well as recent (over the prior 3 years) increases in triglycerides and LDL cholesterol69.

Other variables

Joint aches and stiffness In the cross-sectional analysis, over 50% of the cohort reported being bothered by aches or stiff joints. This was the most prevalent symptom reported20. After 8 years of follow-up, this was still the most common complaint reported by the cohort70. Further analysis of these data is in progress.

Physical, sexual and emotional violence A 'violence questionnaire', incorporating a modified Conflict Tactics Scale and questions on sexual abuse experienced during childhood and adult life, was completed by the participants in the 6th year of follow- up. Ninety-two percent of the women completed the questionnaire. Overall, 28.5% of the women had experienced some form of domestic violence (physical, sexual or emotional) during their lifetime; 5.5% had experienced severe physical abuse in the past year at the hands of a partner and 11.8% had experienced rape or attempted rape between the age of 16 and the time of the survey. Regarding abuse in childhood, 8.9% of women had experienced physical abuse, 42.3% had experienced non-contact sexual abuse, and 35.7% contact sexual abuse71.

The association between childhood abuse (sexual and physical) and mid-aged women's sexual functioning was investigated. Women who had experienced penetrative childhood sexual abuse were significantly more likely to have fewer children (median 2) than women who had not experienced penetrative childhood sexual abuse (median 3). Feelings for partner were significantly affected by childhood sexual abuse. Those who experienced penetrative childhood sexual abuse had, on average, significantly shorter current relationships. Women who had experienced both childhood sexual and physical abuse reported a lower frequency of current sexual activities72.

Attitudes to aging Attitudes to aging are part of the psychological phenomena surrounding the mid-life of women. Two questionnaires were compared: the Reactions to Aging Questionnaire and the Worries about Aging version. The former may be more informative in abilities to assess emotions, opinions, and beliefs associated with aging in mid-aged women, and thus serve as a better predictor of health-related factors73. The Worries about Aging Scale was utilized in years 1 and 5 of follow-up in the MWMHP. Experiencing the menopausal transition was not related to any significant changes in worries about aging.

DISCUSSION

The MWMHP has documented changes over the menopausal transition in a number of health parameters and in a wide range of determinants, including psychosocial, lifestyle and biological factors, and has contributed to the development of rigorous methods and novel analytic techniques for cohort studies of the menopause. This observational study provides both a needed description of the natural history of a range of health outcomes and behaviors of Australian-born women, as well as identifying causal links that can then be tested further with molecular biological approaches and clinical trials. One limitation of the project was that the results only relate to Australian-born women; however, the study is collaborating with other international cohort studies so that global comparisons can be made74. Also, other international investigators studying the menopause made use of our methodology75-77.

The cross-sectional phase of the project did discover some significant associations but is limited by being unable to take into account the large effect of intrinsic factors (measured as baseline levels in the prospective study) or to determine causal pathways. Longitudinal studies do allow the determination of health outcomes attributable to menopause and reduced ovarian function or to aging. Longitudinal studies of the menopause can be limited by small numbers experiencing the menopausal transition due to short follow- up time, poor retention rates or use of exogenous hormones prior to the FMP. During the 9 follow-up years of the MWMHP, 51% of the women experienced a natural menopausal transition and 12% dropped out, leaving sufficient numbers to examine the effects of the natural menopause transition. Age at baseline is another limitation to consider when studying the transition. The minimum age of the women in the MWMHP was 45 years, and women who had experienced an earlier menopause were excluded from the cohort. Our average age of menopause was 53 years, slightly higher than in other studies, and this probably reflects our baseline age range.

Another limitation of our study is the healthy bias of the sample. The women recruited into the longitudinal study from the population-based sample were more likely to exercise more than once a week. The dropout rate was very low, 12% after 8 years of follow- up, and the women who were retained were more likely to exercise and to be married or living with a partner. However, there was a wide range of lifestyle and sociodemographic characteristics and the range is representative of an Australian-born population. Over 8 years of follow-up, there was a decrease in the number of women who smoked (19% to 14%), little change in alcohol consumption (about 25% of women drank eight or more glasses of alcohol per week) and an increase in participation in physical activities for fitness or recreational purposes. In the 8th year of follow-up, 68% of women exercised more than once a week for fitness or recreational purposes, compared with 56% at baseline (unpublished findings).

Dietary factors were investigated throughout the study, in particular calcium intake in relation to bone health and the intake of phytoestrogens in relation to health variables. Approximately 60% of the cohort had a calcium intake below that recommended for optimum bone health18. The finding of no significant effect of calcium intake or physical activity on bone changes during the menopausal transition may be due to inadequate levels of these factors or that there is an overriding effect of changes in estradiol levels. Fourteen percent of the cohort consumed more than 40 mg/day of the phytoestrogen, isoflavone78. The high phytoestrogen intake was only one component of a healthier lifestyle when these consumers were compared with women with a low phytoestrogen intake and so it was not possible to assign particular health benefits to the one dietary component.

The design of the MWMHP enabled the hormone changes over the menopausal transition to be demonstrated, although a limitation to this investigation is that we only took one blood sample from each woman per year. Reproductive aging is marked by increasing circulating FSH levels and falling estradiol and inhibin levels. There is a high degree of individual variability and modest tracking of individual levels such that the interpretation of isolated hormonal measurements on women during the menopausal transition cannot be used reliably to define their reproductive status. Using data from menstrual diaries, as well as hormone levels and women's self-rating, is a more reliable method of predicting the timing of the final menstrual period.

This study was able to report changes in the dimeric inhibins in relation to the menopause. Mean levels of both INHA and INHB fell before the final menses, the major falls in INHA being observed in the last 18 months before final menses and in INHB over a somewhat longer period. INHA is primarily a secretory product of the dominant follicle and the corpus luteum and INHB of the small antral follicles in the recruited cohort. INHB may be regarded as indicating the size of that recruited cohort. Further data79 on the dimeric inhibins have found that INHB is the most important regulator of early follicular-phase serum FSH levels in normal women and that rising levels of FSH with age are a result of lower levels of INHB. A decline in INHB allows an increased level of FSH to maintain a stimulus to the dominant follicle in order to maintain the overall secretion of estradiol and INHA.

In the MWMHP cohort, the menopause per se and the accompanying decline in estradiol levels were shown to be directly associated with: an increase in vasomotor symptoms, vaginal dryness and dyspareunia; a decrease in breast tenderness; a decline in bone mineral density of the spine and femoral neck regions; the increase in \risk of coronary heart disease calculated using the PROCAM score; an increase in central abdominal fat and a decline in sexual functioning. In the latter three health outcomes, other factors also contributed significantly to the results. Higher than average BMI and an increase in BMI had a strong influence on CHD risk and abdominal fat, whereas aging and partner factors affected sexual responsivity. Lipid levels that contribute to the PROCAM score for CHD risk are also significantly affected by increasing age.

Mood, self-rated health and life satisfaction were not directly related to the menopausal transition. Psychosocial, lifestyle and prior attitudes had a significant effect on these outcomes. The menopausal transition did have an indirect effect on mood by amplifying the effect of psychosocial variables. Other health outcomes on which the menopause and hormone levels did not have a significant effect included incontinence, impaired fasting glucose levels and memory.

The longitudinal cohort of the MWMHP is now in the 12th year of follow-up. Areas of investigation include further cognitive function testing, HT and brain function, prevalence and determinants of osteoarthritis, role of inflammatory markers in bone mineral density, cardiovascular disease and brain function, and determinants of breast density.

In conclusion, the Project's longitudinal study, with adequate measures of health outcomes and data on a range of hormonal, biological, psychosocial and lifestyle variables, has provided important information on changes during the menopausal transition. The project has identified those health changes significantly associated with declining estradiol. Premenopausal women should be advised of these hormone-related changes and counseled on lifestyle changes that might prevent adverse health outcomes in postmenopause, particularly changes that would avoid the increase in body weight that generally occurs in the mid-life years. Promoting positive attitudes to aging and the menopause and healthy lifestyles should be used both as community interventions and as part of individual approaches by health practitioners with mid-aged women.

ACKNOWLEDGEMENTS

We would like to thank all the study participants who gave so generously of their time in order to make this project a success. We would like to thank the following co-investigators and researchers who were involved in this study: Adele Green, John Hopper, Peter Ebeling, John Wark, Mark Wahlqvist, Madeleine Ball, Carol Morse, Alastair MacLennan, Anthony Smith, Julia Shelley, John Randolph, Maggi Ryan, David Purdie, Emma Dudley, Corry Garamszegi, Heather Amiconi, Margaret Bolton and Joan Rudder. We would also like to thank Mr Nick Balazs and the staff of the Department of Biochemistry at the Monash Medical Centre for the hormone and lipid assays. The authors would like to thank Margaret Clark and Nicola Blaxill for help with the manuscript.

Conflict of interest J.R.G., H.G.B. and J.R.T. nil. Professor Lorraine Dennerstein has been a consultant to Eli Lilly Ltd., Procter and Gamble, and Pfizer drug companies. Professor Lehert is a consultant for Merck and Co.

Source of funding Between 1991 and 2000, from the Victoria Health Promotion Foundation, the Public Health Research and Development Committee of the Australian National Health and Medical, the University of Melbourne, Australian Dairy Corporation, Percy Baxter Trust, H & L Hecht Trust, Estate of the late Daniel Scott, Ian Potter Foundation, Smorgan Family Trust, Leigh & Marjorie Bronwen Murray Trust, Helen M Schutt Trust, the Australasian Menopause Society, Eli Lilly Ltd and Pharmacia. The writing of this monograph was supported by a grant from the ANZ Trustees. Organon Australia Pty Ltd provided a grant to Prince Henry's Medical Research Institute for the blood measurements. The project has a current grant (2002-03) from Wyeth America Pty Ltd.

References

1. McKinlay SM, McKinlay JB. Selected studies of the menopause - a methodological critique. J Biol Sci 1973;5:533-55

2. Kaufert P. Women and their health in the middle years: a Manitoba project. Soc Sci Med 1984;18:279-81

3. Avis NE, McKinlay SM. A longitudinal analysis of women's attitudes toward the menopause: results from the Massachusetts Women's Health Study. Maturitas 1991;13:65-79

4. McKinlay SM, Brambilla DJ, Prosner JG. The normal menopausal transition. Am J Hum Biol 1992;4:37-46

5. WHO Scientific Group. Research on the Menopause in the 1990s, WHO Technical Services Report 866. Geneva: World Health Organization, 1996

6. Dudley E, Hopper J, Taffe J, Guthrie JR, Burger H, Dennerstein L. Using longitudinal data to define the perimenopause by menstrual cycle characteristics. Climacteric 1998;1:18-25

7. Soules MR, Sherman S, Parrott E, et al. Executive summary: stages of reproductive aging workshop (STRAW). Fertil Steril 2001;76:874-8

8. Smith A, Dennerstein L, Morse C, Hopper J, Green A. Costs and benefits of the use of commercial market research approaches in large scale surveys. Letter to the Editor. Med J Aust 1992;157:504

9. Kammann R, Flett R. Affectometer 2: a scale to measure current level of general happiness. Aust J Psychol 1983;35:259-65

10. Dennerstein L, Smith A, Morse C. Psychological well-being, mid-life and the menopause. Maturitas 1994;20:1-11

11. Smith AMA, Shelley JM, Dennerstein E. Self-rated health: biological continuum or social discontinuity? Soc Sci Med 1994;39:77- 83

12. Dennerstein L, Smith AMA, Morse CA, et al. Menopausal symptomatology in Australian women. Med J Aust 1993;159:232-6

13. Guthrie J, Dennerstein E, Burger H, Hopper J. Hot flushes, menstrual status and hormone levels in a population-based sample of midlife women. Obstet Gynecol 1996;88:437-42

14. Morse C, Smith A, Dennerstein L, Green A, Hopper J, Burger H. The treatment-seeking woman at menopause. Maturitas 1994;18,3:161- 73

15. Shelley J, Smith A, Dudley E, Dennerstein L. Use of hormone replacement therapy by Melbourne women. Aust J Pub Health 1995;19:387-92

16. Dennerstein L, Shelley J, Smith AMA, Ryan M. Hysterectomy experience among mid-aged Australian women. Med J Aust 1994;161:311- 3

17. Taffe J, Green A, Dudley E, Dennerstein L. Tubal ligation, hysterectomy and D & C: evidence from the Melbourne Women's Midlife Health Project. Gynecol Obstet Invest 2000;50:117-22

18. Guthrie J, Ebeling P, Hopper J, Wark J, Dennerstein L, Burger H. Bone mineral density and hormone levels in menopausal Australian born women. Gynecol Endocrinol 1996;10:199-205

19. Ebeling P, Atley L, Guthrie J, et al. Bone turnover markers and bone density across the menopausal transition. J Clin Endocrinol Metab 1996;81:3366-71

20. Shelley J, Green A, Smith A, et al. Relationship of endogenous sex hormones to lipids and blood pressure in mid-aged women. Ann Epidemiol 1998;8:39-45

21. Dennerstein L, Smith AMA, Morse GA, Burger H. Sexuality and the menopause. J Psychosom Obstet Gynaecol 1994;15:59-66

22. Guthrie JR, Smith AMA, Dennerstein L, Morse C. Physical activity and the menopause experience: a cross-sectional study. Maturitas 1995;20:71-80

23. Burger H, Dudley E, Hopper J, et al. The endocrinology of the menopausal transition: a cross-sectional study of a population based sample. J Clin Endocrinol Metab 1995;80:3537-45

24. Garamszegi C, Dennerstein, Dudley E, Guthrie J, Ryan M, Burger HG. Menopausal status: subjectively and objectively defined. J Psychosom Obstet Gynecol 1998;19:165-73

25. Taffe J, Garamszegi C, Dudley E, Dennerstein L. Determinants of self rated menopause status. Maturitas 1997;27:223-9

26. Sherburn M, Guthrie JR, Dudley E, O'Connell H, Dennerstein L. Is incontinence associated with the menopause? Obstet Gynecol 2001;98:628-33

27. Dennerstein L, Helmes E. The menopausal transition and quality of life: methodologic issues. Quality Life Res 2000;9:721- 31

28. Lehert P, Dennerstein L. Statistical techniques for the analysis of change in longitudinal studies of the menopause. Acta Obstet Gynecol 2002;81:581-7

29. Taffe J, Dennerstein L. Menstrual diary data and the menopausal transition: methodological issues. Acta Obstet Gynecol 2002;81:588-94

30. Dennerstein L, Dudley E, Hopper J, Burger H. Sexuality, hormones and the menopausal transition. Maturitas 1997;26:83-93

31. Dennerstein L, Lehert P, Dudley E. Short scale to measure female sexuality; adapted from McCoy Female Sexuality questionnaire. J Sex Marital Ther 2001;27:339-51

32. Dennerstein L, Anderson-Hunt M, Dudley E. Evaluation of a short scale to assess female sexual functioning. J Sex Marital Ther 2002;28:389-97

33. Dennerstein L, Lehert P. Modelling mid-aged women's sexual functioning: a prospective, population-based study. J Sex Marital Ther 2004;30:173-83

34. Guthrie JR. Physical activity: measurement in mid-life women. Acta Obstet Gynecol 2002;81:595-602

35. Risk Factor Prevalence Study Management Committee. Risk factor prevalence study: survey no. 3, 1989. Canberra: National Heart Foundation of Australia and Australian Institute of Health, 1990

36. Health WIZ National Social Health Database, Data set P11: census 1991. Canberra: Department of Human Services and Health, 1994

37. Guthrie JR, Dennerstein L. Studying the menopausal transition: methodological considerations, Climacteric 2003;6:354-6

38. Guthrie JR, Dennerstein L, Burger H. How reliably does 12- month amenorrhea define final menstrual period? Data from a longitudinal study. Fetter to the Editor. Climacteric 2002;5:92

39. Guthrie JR, Garamszegi C, Dudley E, Dennerstein L, Green A, Burger H. Hormone therapy use in Australian born women: a longitudinal study. Med J Aust 1999;171:358-61

40. Burger H, Dudley E, Hopper J, et al. Prospectively measured levels of serum FSH, estradiol and the dimeric inhibins during the menopausal transition in a population-based cohort of women. J Clin Endocrinol Metab 1999;84:4025-30

41. Burger HG, Dudley EC, Cui J, Dennerstein L, Hopper JL. A prospective longitudinal study of serum testosterone\, dehydroepiandrosterone sulfate, and sex hormone-binding globulin levels through the menopause transition. J Clin Endocrinol Metab 2000;85;2832-8

42. Taffe J, Dennerstein L. Retrospective self-report compared with menstrual diary data prospectively kept during the menopausal transition, Climacteric 2000;3:183-91

43. Taffe J, Dennerstein L. Menstrual patterns leading to the final menstrual period. Menopause 2002;9:32-40

44. Taffe J, Dennerstein L. Time to the final menstrual period. Fertil Steril 2002;78:397-403

45. Dennerstein L, Dudley EC, Hopper JL, Guthrie JR, Burger HG, A prospective population-based study of menopausal symptoms. Obstet Gynecol 2000;96;351-8

46. Guthrie JR, Dennerstein L, Donnelly V. Health care seeking for menopausal problems. Climacteric 2003;6:112-7

47. Morse C, Dudley E, Guthrie JR, Dennerstein L. Relationships between premenstrual complaints and perimenopausal experiences. J Psychosom Obstet Cynccol 1998;19:182-91

48. Dennerstein L, Dudley E, Burger HG. Wellbeing and the menopausal transition. J Psychosom Obstet Gynecol 1997;18:95-101

49. Dennerstein L, Lehert P, Guthrie J. The effects of the menopausal transition and biopsychosocial factors on well-being. Arch Women's Menial Health 2002;5:15-22

50. Dennerstein L, Lehert P, Burger H, Dudley E. Mood and the menopausal transition. J Nerv Ment Dis 1999;187:685-91

51. Dennerstein L, Lehert P, Dudley E, Guthrie JR. Factors contributing to positive mood during the menopausal transition. J Nerv Ment Dis 2001;189:84-9

52. Denneretein L, Dudley E, Guthrie J. Empty nest or revolving door? A prospective study of women's quality of life in midlife during the phase of children leaving and re-entering the home. Psychol Med 2002;32:545-50

53. Dennerstein L, Lehert P, Burger H, Dudley E. Factors affecting sexual functioning of women in the midlife years. Climacteric 1999;2:254-62

54. Dennerstein L, Randolph J, Taffe J, Dudley E, Burger H, Hormones, mood, sexuality and the menopausal transition. Fertil Steril 2002;77(Suppl 4):S42-8

55. Dennerstein L, Dudley E, Burger H. Are changes in sexual functioning during midlife due to aging or menopause? Fertil Steril 2001;76:456-60

56. Dennerstein L, Dudley EC, Guthrie JR. Predictors of declining self-rated health during the transition to menopause. J Psychosom Res 2003;54:147-53

57. Dennerstein L, Dudley E, Guthrie JR, Barrett-Connor E. Life satisfaction, symptoms and the menopausal transition. Medscape Women's Health 2000;5. http://www.medscape.com/ viewarticle/408924

58. Guthrie JR, Ebeling P, Barrett-Connor E, et al. A prospective study of bone loss in perimenopausal Australian born women. Osteoporos Int 1998;8:282-90

59. Guthrie JR, Dennerstein L, Wark JD. Risk factors for osteoporosis: a review. Medscape 'Women's Health 2000;5. http:// www.medscape.com/viewarticle/408926

60. Guthrie JR, Ebeling PR, Dennerstein L, Wark JD. Risk factors for osteoporosis: prevalence, change and association with bone density. Medscape Women's Health 2000;5. http://www.medscape.com/ viewarticle/408930

61. Guthrie JR, Lehert P, Dennerstein L, Burger HG, Ebeling PR, Wark JD. The relative effect of endogenous estradiol and androgens on menopausal bone loss: a longitudinal study. Osteoporos Int 2004;15:881-6

62. Guthrie J, Dennerstein L, Dudley E. Weight gain and the menopause: a 5 year prospective study. Climacteric 1999;2:205-11

63. Guthrie JR, Dennerstein L, Taffe JR, et al. Central abdominal fat and endogenous hormones during the menopausal transition. Fertil Steril 2003;79:1335-40

64. Do KA, Green A, Guthrie JR, Dudley EC, Burger HG, Dennerstein L. A longitudinal study of risk factors for coronary heart disease across the menopausal transition. Am J Epidemiol 2000;151:584-93

65. Guthrie J, Dudley E, Hopper J, Dennerstein L. Changes in physical activity and health outcomes in a population-based cohort of midlife Australian-born women. Aust NZ J Public Health 1997;21:682-7

66. Guthrie JR, Ball M, Dudley EC, et al. Impaired fasting glycaemia in middle-aged women: a prospective study. Int J Obesity 2001;25:646-51

67. Guthrie JR, Taffe JR, Lehert P, Burger HG, Dennerstein L. Association between hormonal changes at menopause and the risk of a coronary event: a longitudinal study. Menopause 2004;11:315-22

68. Henderson VW, Guthrie JR, Dudley EC, Burger HG, Dennerstein L. Estrogen exposures and memory at midlife: a population-based study of women. Neurology 2003;60:1369-71

69. Henderson VW, Guthrie JR, Dennerstein L. Serum lipids and memory in a population-based cohort of middle-aged women. J Neurol Neurosurg Psychiatry 2003;74:1530-4

70. Szoeke C, Dennerstein L, Cicuttini F, Guthrie JR. Ostcoarthritis, menopause and aging. Climacteric 2002;5(Suppl):P-24- 03

71. Mazza D, Dennerstein L, Garamszegi C, Dudley E. The physical, sexual and emotional violence history of middle-aged women: a community-based prevalence study. Med J Aust 2001;175:199-201

72. Dennerstein L, Guthrie JR, Alford S. Childhood abuse and its association with mid-aged women's sexual functioning. J Sex Marital Ther 2004;4:225-34

73. Netz Y, Guthrie JR, Garamszegi C, Dennerstein L. Attitudes of middle-aged women to ageing - contribution of the Reactions to Aging questionnaire. Climacteric 2001;4:306-13

74. Gold EB, Sternfeld B, Kelsey JL, et al. Relation of demographic and lifestyle factors to symptoms in a multiracial/ ethnic population of women 40-55 years of age. Am J Epidemiol 2000;152:1463-73

75. Punyahotra S, Dennerstein L. Menopausal experiences of Thai women. 1. Symptoms and their correlates. Maturitas 1997;26:1-7

76. Punyahotra S, Dennerstein L. Menopausal experiences of Thai women. 2. The cultural context. Maturitas 1997;26:9-14

77. Zhao G, Wang L, Yan R, Dennerstein L. Menopausal symptoms: ex

Source: Climacteric

More News in this Category