Midlife women: symptoms associated with menopausal transition and early postmenopause and quality of life
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Menopause: The Journal of The North American Menopause Society Vol. 20, No. 1, pp. 000/000 DOI: 10.1097/gme.0b013e31825a2a91 * 2012 by The North American Menopause Society Midlife women: symptoms associated with menopausal transition and early postmenopause and quality of life Catherine A. Greenblum, PhD, FNP-BC, ARNP,1 Meredeth A. Rowe, PhD, RN, FAAN,1 Donna Felber Neff, PhD, RN,2 and Jesse S. Greenblum, MD, MS, FACOG3 Abstract Objective: The objectives of this study were to examine the effects of symptoms associated with the menopausal transition and early postmenopause on quality of life and to determine if there is a clustering of symptoms that has a larger effect on quality of life than individual symptoms. Methods: This study used data from a cross-sectional study on women aged 45 to 60 years. Descriptive statistics and regression analyses were used to analyze the data. Results: More than 60% of the participants reported three or more symptoms. The symptom clusters that had the highest impact on quality of life were sleep disturbances and vaginal dryness, which accounted for 9.7% of the variance in quality-of-life scores. A parsimonious model of individual symptoms, including sleep disturbances, fatigue, and anxiety, accounted for 16.7% of the variance in quality of life. This group of symptoms, not represented by a cluster, had the highest impact on quality of life. Conclusions: The symptoms found to most significantly affect quality of life are sleep disturbances, fatigue, and anxiety, suggesting that appropriate management of sleep disorders and anxiety may be beneficial to women undergoing the transition to postmenopause. Unanticipated clusters of symptoms point toward a unique symptom experience influenced by factors other than a decline in ovarian function. In this study, symptoms commonly associated with the menopausal transition and early postmenopause negatively affect quality of life; however, the results indicate that quality of life in midlife women is affected by these symptoms only to a small extent. Key Words: Menopause Y Menopausal transition Y Early menopause Y Menopausal symptoms Y Symptom experience Y Symptom cluster Y Quality of life Y Sleep disturbances. T he transition through menopause is a life event that can through menopause and early postmenopause because of the profoundly affect quality of life. More than 80% of large number of symptoms that may co-occur. women report physical and psychological symptoms Symptoms experienced with the menopausal transition and that commonly accompany menopause, with varying degrees early postmenopause are varied and span both physical and of severity and life disruption.1 Few empirical studies, how- psychological domains. Anovulatory cycles and ovarian failure ever, have examined the interrelated nature of symptoms may be accompanied by a multiplicity of physical symptoms. associated with the menopausal transition and early post- Vasomotor symptoms, including hot flashes and night sweats, menopause and the effects of those symptom groups on sleep disturbances, vaginal dryness, urinary incontinence, and quality of life. In some chronic diseases, symptoms may have weight gain, are common physical conditions experienced by greater impact when they co-occur in distinct clusters2,3; this midlife women in the transition through menopause and early impact is referred to as Bsymptom experience.[ It is impor- postmenopause.4,5 Psychological symptoms frequently associ- tant to understand symptom experience during the transition ated with menopause include fatigue, irritability, and anxiety.4,6 Some symptoms associated with changing hormone levels are Received January 8, 2012; revised and accepted April 12, 2012. directly linked with estrogen depletion. Hot flashes, night sweats, From the 1University of South Florida College of Nursing, Tampa, FL; and vaginal atrophy resulting in vaginal dryness are correlated 2 University of Florida College of Nursing, Gainesville, FL; and 3Uni- versity of Florida College of Medicine, Gainesville, FL. with changing levels of sex hormones.7 Other symptoms, such Funding/support: This study was supported by the Thomas H. Maren as sleep disturbances, fatigue, anxiety, and weight gain, al- Fellowship, University of Florida. though common to the experience of menopause, are multi- Financial disclosure/conflicts of interest: None reported. factorial in cause and occur in nonpostmenopausal women as Jesse S. Greenblum is a courtesy faculty at the University of Florida well. Studies find that most women experience at least one or College of Medicine. more of these symptoms as they transition through the post- Reprints are not available from the authors. menopausal stage of life.4,8 Despite a majority of women Address correspondence to: Catherine A. Greenblum, PhD, FNP-BC, experiencing multiple symptoms, the literature still presents a ARNP, University of South Florida College of Nursing, 12901 Bruce B. Downs Boulevard, MDC 22, Tampa, FL 33612-4766. E-mail: cgreenbl@ gap on whether clusters of symptoms consistently occur and health.usf.edu what effect symptom clusters have on quality of life. Menopause, Vol. 20, No. 1, 2012 1 Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.
GREENBLUM ET AL Quality of life is a broad, multidimensional concept that 45 to 60 years. Data were gathered using questionnaires that lacks a precise definition in the medical literature.9 The World participants completed at the study site. Health Organization has defined quality of life as Bindividuals’ perception of their position in life in the context of the culture Study population and sample and value systems in which they live and in relation to their A convenience sample of 150 community-dwelling women goals, expectations, standards, and concerns.[10 Quality of life participated in the parent study; 112 of those women reported tends to decline in midlife women, and there is a need to at least one menopausal symptom, and this subgroup became determine what role, if any, symptoms commonly associated the sample for the study reported here. Original participants with the transition to menopause and early postmenopause play were recruited via a flyer posted in the waiting room at the in this phenomenon.11,12 Quality of life is an important out- study site, an obstetric-gynecological clinic in northeast come measure of health care, and understanding the impact of Florida. Interested women, patients in the clinic, and accom- menopause on quality of life is a critically important part of the panying family members or friends self-selected to take the self- care of symptomatic postmenopausal women.13 administered questionnaire packets anonymously. Packets were Symptom experience includes perception of, evaluation of, returned to a locked box in the study site waiting room. No and response to a symptom. Symptom evaluation occurs when protected health data or identifying information was collected individuals make a judgment about the severity, cause, treat- on the survey forms. ability, and effects of a particular symptom on their life.14. Subscales for phys- The objectives of this study were to examine the effects of ical and psychological symptoms were created. Cronbach’s > symptoms commonly associated with the transition through for the psychological symptom subscale (fatigue, irritability, menopause and early postmenopause on quality of life and to and anxiety) was 0.597, indicating adequate reliability.18 For determine if there is a clustering of symptoms that has a larger the physical symptom subscale (sleep disturbances, weight gain, effect on quality of life than individual symptoms. The spe- urinary incontinence, vaginal dryness, and vasomotor symp- cific aims of the study were as follows: toms), Cronbach’s > was 0.275Van indication of the diverse nature of the physical symptoms of menopause that is reflected & To determine what symptoms commonly associated with in the lack of consistency among the subscale constructs. the transition through menopause and early postmeno- Quality of life was measured using the Utian Quality of pause tend to co-occur together. Life scale, a 23-item questionnaire developed to measure & To determine the impact of each symptom cluster on quality of life specifically during the climacteric years of life.9 quality of life. Utian et al9 incorporated the constructs of occupational quality & To develop a model of symptoms commonly associated of life, health quality of life, emotional quality of life, and with the transition through menopause and early post- sexual quality of life to form a total quality-of-life score for menopause that predicts the highest negative influence on this population. Scored on a five-point Likert-type scale from quality of life. 1 (not true of me) to 5 (very true of me), the women’s scores are a total summed score, with higher total scores associated with higher quality of life. The Utian Quality of Life scale is METHODS reported in the literature as psychometrically sound.9 Study design This study used data from a cross-sectional study that Demographic questionnaire focused on symptoms commonly associated with the meno- Data on age, race, current marital status, educational level, pausal transition and early postmenopause in women aged and household income were gathered with a demographic 2 Menopause, Vol. 20, No. 1, 2012 * 2012 The North American Menopause Society Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.
MENOPAUSAL SYMPTOMS AND QUALITY OF LIFE TABLE 1. Demographic data for Florida, Nassau County, FL cluster on quality-of-life scores and to identify which cluster and the study population had the greatest negative impact on quality of life. Nassau Study For research question 3, all symptoms were entered into a Characteristics Floridaa County, FLb population linear regression model using a backward stepwise entry Age, mean (range), y 52 (45-60) technique to determine the impact of the group of symptoms (n = 112) on quality-of-life scores and to develop a parsimonious model Marital status, % Not currently married 27 (n = 30) of symptoms impacting quality of life. Currently married 73 (n = 82) Race, % RESULTS White non-Hispanic 79.8 89.3 81.3 (n = 91) Descriptive statistics Hispanic 21 2.8 4.5 (n = 5) African American 15.9 8.3 10.7 (n = 12) Participants generally were white non-Hispanic, had a mean Asian 2.3 0.8 3.6 (n = 4) age of 52 years, were currently married, had some college Educational attainment, % education or higher, and had a median household income in the High school graduate/GED 84.9 81 92.8 (n = 104) Bachelor’s degree or higher 25.6 18.9 33.9 (n = 38) highest income category of $50,000 or more. The demographics Median household income, $ 47,802 59,000 Q50,000 of the study population approximated the demographic data a The 2010 US Census Bureau data are for both sexes; study population data for the state of Florida, with the exception of educational attain- are for women only. ment and an underrepresentation of Hispanic women in the b The 2009 US Census Bureau data are for both sexes; study population data are for women only. study sample. These results are summarized in Table 1. The most commonly experienced symptom was hot flashes, questionnaire. Race was self-identified from a list, and both with 73.2% of women currently experiencing that symptom. educational status and income were grouped into categories. In decreasing order, the remaining frequencies were as fol- lows: fatigue (58.0%), sleep disturbances (56.3%), anxiety Procedures (53.6%), irritability (51.8%), weight gain (51.8%), vaginal The primary investigator was involved in the clinic but was dryness (48.2%), and urinary incontinence (32.1%). More not involved in the care of any of the study participants. than 60% of the participants reported three or more symptoms, Women in the waiting roomVboth patients in the clinic and with a mode of four symptoms (Table 2). Total quality-of-life those accompanying themVwere free to take a questionnaire, scores ranged from 42 to 92, with a mean score of 72.18 T and the primary investigator had no knowledge of who par- 10.49 and with higher scores indicating higher quality of life. ticipated in the study. This study was reviewed and granted On examination of bivariate relationships, although vaso- exempt status by the University of Florida Institutional motor hot flashes were the most commonly reported symp- Review Board before data collection. tom, correlations with other symptoms were all below 0.01, and none reached statistical significance. The strongest rela- Statistical methods tionship was observed between the psychological variables of The primary objectives of this study were to evaluate the irritability, anxiety, and fatigue. Correlations among the effects of symptoms commonly associated with the meno- physical symptoms were very low (Table 3). The strongest pausal transition and early postmenopause on quality of life correlation was observed between sleep disturbances and and to determine whether a clustering of common symptoms quality of life, with a significant negative correlation with or a model of single symptoms had the greatest negative quality of life (r = j0.33, P G 0.001). impact on quality of life in women aged 45 to 60 years. IBM Question 1: What symptoms commonly associated with the Statistical Package for the Social Sciences version 19 was menopausal transition and early postmenopause tended to used for data analysis. > was set to 0.05. co-occur together? For research question 1, principal components analysis by Oblimin rotation was used to determine the symptoms that A principal components analysis using a forced three-factor tended to occur together. Items were considered to load on a solution explained 56.47% of the variance. The scree plot given factor if the coefficient rounded to at least 0.50. Because showed no distinct elbow at any point. All items had a loading previous literature indicated a three-factor model,1 the first TABLE 2. Frequency of the number of reported symptoms iteration was forced at a three-factor model. A priori, it was decided that if there was less than 50% of explained variance Number of Valid Cumulative symptoms reported Frequency percentage percentage or a number of items with a coefficient less than 0.50, then both a two-factor model and a four-factor model would be 1 6 5.4 5.4 2 17 15.2 20.5 explored. There were no missing data for symptoms used in 3 18 16.1 36.6 the factor analysis; any symptom not endorsed was assumed 4 25 22.3 58.9 not to be present for that participant. 5 17 15.2 74.1 6 12 10.7 84.8 For research question 2, three one-step linear regression 7 11 9.8 94.6 models were run using the symptom clusters obtained from 8 6 5.4 100.0 research question 1 to determine the impact of each symptom Total 112 100.0 Menopause, Vol. 20, No. 1, 2012 3 Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.
GREENBLUM ET AL TABLE 3. Pearson’s correlations of menopausal symptoms 1 2 3 4 5 6 7 8 9 a 1. Sleep disturbances r 0.089 0.036 0.086 0.262 j0.086 0.106 j0.059 j0.328a P 0.351 0.710 0.370 0.005 0.370 0.266 0.540 0.000 2. Fatigue r j0.065 0.338a 0.260a 0.096 0.198b 0.302a 0.174 P 0.496 0.000 0.006 0.312 0.037 0.001 0.067 3. Vasomotor hot flashes r j0.059 0.003 0.099 0.028 0.102 0.039 P 0.536 0.976 0.297 0.771 0.283 0.681 4. Irritability r 0.392a 0.323a 0.167 0.034 0.001 P 0.000 0.001 0.079 0.718 0.991 5. Anxiety r 0.074 0.104 j0.003 j0.226b P 0.437 0.275 0.975 0.017 6. Vaginal dryness r 0.101 0.001 0.103 P 0.289 0.989 0.279 7. Urinary incontinence r 0.396a j0.008 P 0.000 0.932 8. Weight gain r 0.028 P 0.766 9. Quality of life r a Correlation is significant at the 0.01 level (two-tailed). b Correlation is significant at the 0.05 level (two-tailed). factor rounded to 0.5 or greater on at least one factor, meeting menopause that predicts the highest negative influence on the predetermined criterion for an acceptable solution. Con- quality of life? sequently, a three-factor model was accepted (Table 4). Factor Symptoms were entered into a linear regression model 1 is best described as grouping psychological symptoms, with using the backward method of selection. The resulting parsi- anxiety, irritability, and fatigue loading most strongly on this monious model included only those menopausal symptoms factor. Factor 2 is a pair of physiologically related symptoms with a significant effect (P e 0.05) on quality-of-life scores. of weight gain and urinary stress incontinence. Factor 3 The symptoms remaining in the model at the last step were groups vaginal dryness and sleep disturbances. sleep disturbances, fatigue, and anxiety, accounting for 16.7% Question 2: What is the impact of each symptom cluster on of the variance in quality of life (Table 6). The A values of quality of life? sleep disturbances (j0.294), fatigue (j0.256), and anxiety (j0.215) indicate a negative effect on quality of life. This Demographic variables were not included in the final model group of symptoms, not represented by a cluster, had the because there was no significant relationship between any highest impact on quality of life; each cluster individually had demographic variable and model variables in both bivariate less effect than the parsimonious symptom model containing analyses and when placed in step 1 of the overall regression sleep disturbances, fatigue, and anxiety. The r2 for the parsi- model (F = 1.58, P 9 0.05). Accordingly, three one-step models monious three-symptom model almost equaled the total r2 for were run based on the clusters found from the principal com- the eight symptoms in the clusters (Table 7). ponents analysis. The symptom cluster that had the highest impact on quality of life was factor 3, sleep disturbances and vaginal dryness, which accounted for 9.7% of the variance in DISCUSSION quality-of-life scores (Table 5). In this sample of women experiencing at least one symptom Question 3: Can we develop a model of symptoms commonly associated with the menopausal transition and early post- associated with the menopausal transition and early post- menopause, 95% claimed to be currently experiencing more than one symptom. Symptoms clustered together in either psychological or physical domains, with the exception of an TABLE 4. Factor loadings Component TABLE 5. Regression models for each factor symptom cluster 1 2 3 R2 A Sleep disturbances 0.425 0.009 0.618 Model 1 0.086 Fatigue 0.496 0.453 0.002 Anxiety 0.240 Anxiety 0.763 0.024 0.152 Irritability 0.039 Irritability 0.779 0.013 0.355 Fatigue j0.303 Vaginal dryness 0.362 0.069 0.800 Model 2a 0.017 Urinary incontinence 0.141 0.733 0.065 Urinary incontinence j0.023 Weight gain 0.107 0.878 0.043 Weight gain 0.038 Vasomotor hot flashes 0.125 0.175 0.161 Model 3 0.097 Variance explained, % 25.2 16.8 14.5 Sleep disturbances j0.321 Cronbach’s > including factors 0.597 0.275 0.187 Vaginal dryness 0.076 a Factor scales were included in the boldfaced items for each factor. Model 2 was not statistically significant at P = 0.934. 4 Menopause, Vol. 20, No. 1, 2012 * 2012 The North American Menopause Society Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.
MENOPAUSAL SYMPTOMS AND QUALITY OF LIFE TABLE 6. Regression model of menopausal symptoms using a through menopause and early postmenopause and the effects backward method of selection of these symptoms on quality of life in midlife women.1,8,11,24 Standardized A Significance The findings of this study were consistent with the results of Sleep j0.294 0.001 previous research studies, noting the clustering of symptoms in Fatigue j0.256 0.005 three groupings, symptom clusters including symptoms with Anxiety j0.215 0.022 different underlying factors, and the finding that symptoms associated with the transition to menopause and early post- anomalous group that contained sleep disturbances and vagi- menopause are individualized.1,25 This study highlights the nal dryness. The first factor consisting of psychological need for an assessment of sleep and anxiety symptoms in symptoms (anxiety, irritability, and fatigue) in agreement with midlife women; however, further research is needed to con- published literature.1 Factor 2 consisted of weight gain and firm these findings. urinary stress incontinence. This pair of symptoms is physio- logically linked. Weight gain is a risk factor for urinary stress incontinence, and these two symptoms have been linked in a CONCLUSIONS previous study.6 In the third factor, vaginal dryness and sleep In the past, it has been common practice to treat any and all disturbances unexpectedly loaded together. Although both symptoms associated with the transition through menopause symptoms have been linked to declining estrogen levels seen and early postmenopause with hormone supplementation. in menopause,19,20 sleep disturbances are commonly caused With current research highlighting the risks of hormone ther- by a multiplicity of factors, including chronic diseases, sleep apy, this practice has changed, and innovative approaches to apnea, pain, and vasomotor symptoms. Hot flashes did not treating symptoms and improving quality of life in midlife load on any factor, indicating that they did not tend to co- women are needed. The symptoms found to most significantly occur consistently with any one of these clusters (Table 4). affect quality of life in this study were sleep disturbances, This finding echoes the results reported by Freeman et al,1 fatigue, and anxiety, suggesting that nonhormonal therapy to where hot flashes loaded only with aches (a symptom not treat sleep disorders and anxietyVrather than treatment of examined in this study) in a factor analysis. diminishing sex hormone levelsVshould be considered as first- The unanticipated clusters of symptoms point away from line therapy. If menopausal symptoms were truly biologically a singular biological alteration causing symptoms (ie, estro- based on changing ovarian steroid levels, a Bsyndrome[ would gen depletion) and toward a unique symptom experience be seen across all cultures in most women, but this is not sup- for postmenopausal women influenced by factors besides ported by the literature. In this study population, the symptoms declining estrogen levels. Indeed, the symptoms that most did not tend to cluster in a biological manner in relation to a affected quality of life in this study (sleep disturbances, fa- decrease in ovarian function. Estrogen-based symptoms, such tigue, and anxiety) crossed symptom clusters, have multiple as hot flashes and urogenital atrophy, were not significantly causes, and are not completely associated with hormonal correlated. This suggests the conclusion that these symptoms changes. are not wholly biologically mediated by ovarian steroids and Menopausal symptoms negatively affected quality of life in that there is some other dynamics that is not well understood. this study; however, the total effects were relatively low even There are many challenges in research on symptoms asso- when significant, indicating that quality of life in this age ciated with the transition to menopause and early menopause group is only affected by symptoms associated with meno- and symptom clusters, most prominently a lack of under- pause to a small extent. Even in the combined model, the standing of the relationships among the various symptoms. menopausal symptoms only accounted for less than 10% of The current literature reports a complex array of factors the variance in quality of life. Previous work based on the affecting quality of life in midlife women. This research study Seattle Midlife Women’s Health Study linked factors such as supports the diversity of menopause experience and its effect career, financial, and relationship issues to well-being in on quality of life. Women should be counseled individu- midlife women, and the roles of these factors in quality of life ally, and healthcare providers should consider each woman’s deserve further investigation.21
GREENBLUM ET AL Acknowledgments: We thank Jason Beckstead, PhD, University of and psychosocial and demographic factors. Qual Life Res 2004;13: South Florida College of Nursing, for consulting on statistical anal- 933-946. yses in both the original dissertation manuscript and this manuscript. 12. Ham O. Predictors of health-related quality of life among low-income midlife women. West J Nurs Res 2011;33:63-78. 13. Col N, Haskins A, Ewan-Whyte C. Measuring the impact of menopausal symptoms on quality of life: methodological considerations. Menopause REFERENCES 2009;16:843-845. 14. Dodd M, Janson S, Facione N, et al. Advancing the science of symptom 1. Freeman EW, Sammel MD, Liu L, Martin P. Psychometric properties of management. J Adv Nurs 2001;33:668-676. a menopausal symptom list. Menopause 2003;10:258-265. 15. Mechanic D. Sociological dimensions of illness behavior. Soc Sci Med 2. Barsevick AM. The elusive concept of the symptom cluster. Oncol Nurs 1995;41:1207-1216. Forum 2007;34:971-980. 16. Mechanic D. The concept of illness behavior. J Chronic Dis 1962;15: 3. Kirkova J, Walsh D, Aktas A, Davis M. Cancer symptom clusters: old 189-194. concept but new data. Am J Hosp Palliat Care 2010;27:282-288. 17. Kupperman HS, Wetchler BB, Blatt M. Contemporary therapy of the 4. Lewis V. Undertreatment of menopausal symptoms and novel options for menopausal syndrome. JAMA 1959;171:1627-1637. comprehensive management. Cur Med Res Opin 2009;25:2689-2698. 18. Field A. Discovering Statistics Using SPSS. 2nd ed. Thousand Oaks, CA: 5. Thurston R, Joffe H. Vasomotor symptoms and menopause: findings Sage Publications, 2005. from the Study of Women’s Health Across the Nation. Obstet Gynecol 19. Greenblum C, Greenblum J, Neff D. Vaginal estrogen use in menopausal Clin North Am 2011;38:489-501. women: is it safe? Am J Nurs Pract 2009;13:26-34. 6. Ford K, Sowers M, Crutchfield M, Wilson A, Jannausch M. A longi- 20. Murphy PJ, Campbell SS. Sex hormones, sleep, and core body temper- tudinal study of the predictors of prevalence and severity of symptoms ature in older postmenopausal women. Sleep 2007;12:1788-1794. commonly associated with menopause. Menopause 2005;12:308-317. 21. Woods NF, Mitchell ES. Is the menopause transition stressful? Meno- 7. The American College of Obstetricians and Gynecologists. Genitourinary pause Manag 2010;19:25-27. tract changes. Obstet Gynecol 2004;104(suppl 4):56S-61S. 22. Woods NF, Mitchell ES, Percival D, Smith-DiJulio K. Is the menopause 8. Avis NE, Stellato R, Crawford S, et al. Is there a menopausal syndrome? transition stressful? Observations of perceived stress from the Seattle Menopausal status and symptoms across racial/ethnic groups. Soc Sci Midlife Women’s Health Study. Sleep 2010;33:539-540. Med 2001;52:345-356. 23. Woods NF, Smith-DiJulio K, Percival D, Tao E, Mariella A, Mitchell ES. 9. Utian W, Janata J, Kingsberg S, Schluchter M, Hamilton J. The Utian Depressed mood during the menopausal transition and early postmeno- Quality of Life (UQOL) scale: development and validation of an instru- pause: observations from the Seattle Midlife Women’s Health Study. ment to quantify quality of life through and beyond menopause. Meno- Menopause 2008;15:223-232. pause 2002;9:402-410. 24. Smith-DiJulio K, Woods NF, Mitchell ES. Well-being during the meno- 10. Division of Mental Health and Prevention of Substance Abuse World pause transition and early postmenopause. Womens Health Issues 2008;18: Health Organization. WHOQOL Measuring Quality of Life. Geneva, 310-318. Switzerland: World Health Organization, 1997. 25. Cray L, Woods NF, Mitchell E. Symptom clusters during the late meno- 11. Avis NE, Assmann SF, Kravitz HM. Quality of life in diverse groups pausal transition stage: observations from the Seattle Midlife Women’s of midlife women: assessing the influence of menopause, health status Health Study. Menopause 2010;17:972-977. 6 Menopause, Vol. 20, No. 1, 2012 * 2012 The North American Menopause Society Copyright © 2012 The North American Menopause Society. Unauthorized reproduction of this article is prohibited.
You can also read