EB Gold NAMS Presentation 10/06/2016 No financial relationships to - - PDF document

EB Gold NAMS Presentation 10/06/2016 1

Ellen B. Gold, PhD

• Dept. Of Public Health Sciences

University Of California Davis

 No financial relationships to disclose  Vasomotor symptoms (hot flashes/flushes, night sweats)

affect 50-80% of women undergoing the menopause transition

 Urinary incontinence affects about half of midlife women  Sleep disturbances affect 15-40% of midlife women  All of these affect quality of life in the millions of women

annually undergoing the menopause transition and incur in total billions of dollars in health care costs annually

EB Gold NAMS Presentation 10/06/2016 2

 Objectives  Design  Methods  Results

• Baseline characteristics of study sample
• Symptoms: Hot flashes and night sweats, Urinary

incontinence, Sleep, Vaginal dryness

*Funded by NIA, NINR, ORWH, NCCAM; no financial conflicts or other disclosures

 To identify factors that affect the timing and nature (eg,

endocrine changes, symptoms) of the menopause transition and

 To identify the characteristics of the transition that are

related to long-term disease risk indicators, such as changes in cardiovascular risk markers and bone density.

 20-year longitudinal cohort study of the menopause transition

funded by NIA, NINR, ORWH, NCCAM

 7 sites, each recruited minority sample

• Michigan

African Americans

• MGH - Boston

African Americans

• Rush – Chicago

African Americans

• UCD/Kaiser – Oakland

Chinese

• UCLA

Japanese

• New Jersey

Hispanic

• Pittsburgh

African Americans

 2 central labs (Michigan, MRL); Coordinating center - Pittsburgh  Screened 16,025 community-based women for cohort eligibility

(age 42-52 years, menstrual period in last 3 months, intact uterus and >1 ovary, no exogenous hormone use, not pregnant)

 Annual visits; 3302 at baseline, 2502 at visit 13:

• Annual in-person interview, self-administered questionnaire

and food frequency questionnaire (baseline, 05, 09)

• Fasting blood sample drawn on cycle days 2-5 for serum E2,

T, FSH, SHBG, DHEAS, glucose, insulin, clotting factors and lipoprotein levels

EB Gold NAMS Presentation 10/06/2016 3

• Annually measure blood pressure, weight, height and

waist and hip circumference

• At 5 sites measure bone density annually
• In subset of 990, one cycle daily urine (urinary E1C,

PdG, FSH, LH) and daily symptom diary

 Monthly calendar: menstruation, sx, procedures  Visit 4 cognitive tests began; at visit 12 physical function

tests began

 Longitudinal analyses: Repeated measures, multi-variable

analyses to account for multiple observations on the same woman and many confounding variables

Baseline Characteristic African Amer. Cauc. Hisp. Chinese Japanese Median Age (years) 46.1 46.1 46.1 46.6 46.7 % < $50,000 annual hshld income 63.1 39.1 91.0 39.2 25.3 % Smokers 24.6 16.6 16.7 1.6 12.9 % >6 days HF in past 2 wks 14.8 10.2 13.4 6.8 4.3 % Premenop 49.5 52.3 57.9 61.9 62.2 Median BMI 30.2 26.1 28.3 22.4 22.1

 Objectives  Design and Methods

 Results

• Vasomotor symptoms

 Affect the majority of women undergoing menopause transition (Gold

et al. Amer J PH 2006)

 Associated with higher factor VIIc and tPA-ag (Thurston Menop 2011),

CV disease risk (Huang Menop 2009; Roussouw JAMA 2007) and lower bone density (Crandall Menop 2009)

 Frequency and severity of hot flashes related to quality of life and

sleep (Pinkerton Menop 2016)

 A chief menopause-related symptom for which US women seek

medical care (Williams Maturitas 2007; Nicholson Am J Obstet Gynecol 2001), have higher number of outpatient visits (Sarrel Menop 2015)

 Annual direct and indirect costs in hundreds of millions of dollars in US

EB Gold NAMS Presentation 10/06/2016 4

Gold et al, AJPH 2006  Assessed VMS for > 6 days in prior 2 weeks (frequent

VMS) through follow-up visit 13

 Excluded women who reported:

• Hysterectomy or oophorectomy during follow-up (n=330)
• Initiation of HT before first report of frequent VMS (n=583)
• No visits with frequent VMS (n=940)

 Assessed persistence of VMS in women with known FMP

date who reported frequent VMS at >1 visit after the FMP (n=881)

Avis et al, JAMA Int Med 2015

EB Gold NAMS Presentation 10/06/2016 5

Duration of Frequent VMS (years)

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Proportion of Women

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Premenopausal (N=183) Early perimenopausal (N=700) Late perimenopausal (N=266) Postmenopausal (N=291) All Participants (N=1449)

Total Duration of Frequent VMS by Menopausal Status at First VMS Report

Avis et al, JAMA Int Med 2015 Median 7.4 Median 3.4 Avis et al, JAMA Int Med 2015

Total Duration of Frequent VMS by Race/Ethnicity

Duration of Frequent VMS (years)

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Proportion of Women

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 African American (N=479) White (N=652) Chinese (N=95) Hispanic (N=109) Japanese (N=114) All Participants (N=1449)

Median 10.1 Median 4.8

Post FMP Persistence of Frequent VMS by Menopausal Status at First VMS Report

Avis et al, JAMA Int Med 2015

Post-FMP Persistence of Frequent VMS (years)

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Proportion of Women

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 Premenopausal (N=72) Early perimenopausal (N=348) Late perimenopausal (N=191) Postmenopausal (N=268) All Participants (N=881)

Avis et al, JAMA Int Med 2015

Post FMP Persistence of Frequent VMS by Race/Ethnicity

Post-FMP Persistence of Frequent VMS (years)

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Proportion of Women

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 African American (N=310) White (N=380) Chinese (N=65) Hispanic (N=50) Japanese (N=76) All Participants (N=881)

EB Gold NAMS Presentation 10/06/2016 6

Factor Cessation over Total Duration Post FMP Cessation HR (95% CI) Median Years HR (95% CI) Median Years Age at

• nset, years

45-49.9 50-54.9 >55 2.43 (1.30-4.51) 2.71 (1.44-5.09) 3.60 (1.78-7.29) 9.34 7.73 6.42 Ref 1.16 (0.84-1.60) 1.53 (1.05-2.24) 6.76 5.72 4.48 > College 1.25 (1.03-1.52) 7.66 1.40 (1.13-1.74) 4.54 Sx sens. V1 low moderate high 0.77 (0.58-1.00) 0.85 (0.64-1.13) 0.64 (0.46-0.89) 9.57 8.22 10.81 0.76 (0.56-1.04) 0.78 (0.57-1.07) 0.58 (0.39-0.85) 5.48 5.45 8.47 Some perceived stress 0.75 (0.58-0.96) 10.78 0.70 (0.54-0.91) 8.13 Avis et al, JAMA Int Med 2015

aHR=hazard ratio for VMS cessation

 Melbourne Women’s Midlife Health (Col et al 2009): shorter

total duration (5.2 years) for bothersome VMS (mostly white, excluded those with VMS at baseline, VMS at last visit considered no VMS)

 Penn Ovarian Aging: longer total duration, 10.2 years, and

similar post-FMP duration, 4.6 years, (Freeman et al 2014) for moderate to severe hot flashes (VMS in past year, 13-year follow-up, age 35-47 years at entry),

• Also, longer duration with onset of hot flashes early in the

menopause transition or at younger age

Factor Total Cauc. Afr. Amer. Hisp. Chin. Japan . Age (per year) 1.17** 1.16** 1.18** 1.04 1.29** 1.30** BMI (per unit) 1.03** 1.03 1.02 1.03 1.03 1.05 Smoking 1.63** 1.14 1.98** 3.09** 2.97 1.49 Baseline anxiety score 3.10** 2.59** 3.65** 1.64 17.3** 3.36+

*Each factor adjusted for others, menopausal status, education, hx premens sx, sx sensitivity and depressive sx score; ** p<0.01; +p<0.05 Gold et al, Am J Public Health 2006

EB Gold NAMS Presentation 10/06/2016 7

 Longitudinal analyses of women who did not report any

VMS at baseline, n=1546

 Examined incidence of any and frequent VMS (>6 days

in past 2 weeks)

• Concurrent BMI (waist)
• Change in weight (waist) from baseline
• Change in weight (waist) from prior visit

 Determined if relationships varied by menopausal status Gold et al. Menopause 2016 in press

(a) Adjusted: Time‐invariant ‐ baseline age, race/ethnicity, site, V01 symptom sensitivity, baseline CES‐D depression, history of PMS, and education; Time‐varying – smoking, number very upsetting life events Gold et al Menop 2016 in press

p=0.0003 p=0.053 5 10 15 20 25 30 35 40 <25 25-29.9 30+ Percent Incidence Concurrent BMI Pre-/early peri Late peri/post P for interaction by menopause stage <0.0001 2 4 6 8 10 12 14 16 18 <25 25-29.9 30+ Percent Incidence Concurrent BMI Pre & Early Peri Late Peri & Post P for interaction by menopause stage = 0.03

P = 0.10 P = 0.34

EB Gold NAMS Presentation 10/06/2016 8

P for interaction by menopause stage = 0.58

5 10 15 20 25 30 35 40 45 Lost >5% Lost 1- 5% Stable Gained 1- 5% Gained 5- 10% Gained >10% Percent Incidence % Weight Change Since Baseline Pre & Early Peri Late Peri and Post

P = 0.08 P = 0.39

2 4 6 8 10 12 14 16 18 20 Lost >5% Lost 1- 5% Stable Gained 1- 5% Gained 5- 10% Gained >5% Percent Incidence Percent Weight Change Since Baseline Pre & Early Peri Late Peri & Post P for interaction by menopause stage = 0.004

P = 0.004 P = 0.07

 For a given concurrent BMI, incidence of VMS not affected by

weight change; weight changes did not add predictive power; also, adjusting for weight change did not affect relation of concurrent BMI to incident VMS.

 Women who gained the most or the least since baseline had

greater incident frequent VMS in the early stage but lower incidence in the late stage; may be chance associations as no U‐shaped relations for any VMS in early or late stage.

 Results may reflect that most percent weight changes were

relatively small absolute changes and do not rule out that large absolute changes may be needed to change VMS risk.

 Interactions with menopause status: concurrent BMI positively

associated with any VMS in the early stage and negatively associated in the late stage; trends not always monotonic or significant for frequent VMS (small numbers)

 Consistent with Galicchio et al’s (2014) longitudinal study of midlife

women showing change in BMI and weight over 1‐5 years of follow‐ up not related to hot flashes in two racial groups; shorter follow‐up and shorter‐term change examined

 Differed from WHI: postmenopausal women with VMS at baseline

who lost 10+ pounds or 10% or more of their baseline body weight in

• ne year were more likely to “eliminate” VMS (Kroenke et al 2012).

EB Gold NAMS Presentation 10/06/2016 9

 Women’s Health in Lund Area (WHLA) study and Australian

Longitudinal Study on Women’s Health (ALSWH) results differed: VMS reported more by women who gained weight but not significantly less in those who lost weight

 WHLA (Li et al 2003) was cross‐sectional, asked weight gain in last 5

years, did not report time over which VMS reported nor racial/ethnic composition, sample aged 51‐61 years

 ALSWH (Herber‐Gast et al 2013) was longitudinal over 15 years,

asked about VMS in last 12 months in 10,454 women aged 45‐50 years at baseline; race/ethnicity was not reported

 Objectives  Design and Methods

 Results

• Vasomotor symptoms
• Urinary symptoms

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Adjusted* Odds Ratio Stress UI Urge UI Type of Incontinence Cauc Afr Amer Chinese Japanese Hispanic ** ** ** ** *Adj. for social support, BMI, parity; ** significant compared to Cauc Waetjen LE, et al, Amer J Epidemiol 2007 ** **

Incidence Of Any, Stress, And Urge Urinary Incontinence Per 100 Woman‐years By Menopausal Status, Baseline ‐ Visit 06 , SWAN

Waetjen et al. Obstet Gynecol 2009

EB Gold NAMS Presentation 10/06/2016 10

Waetjen, et al. Obstet Gynecol 2009 Hazard Ratio

a b

• aAdj. for anxiety change, BMI, education, basics, race/ethnicity, parity, weight change;
• bAdj. for BMI, race/ethnicity, weight changes;
• cAdj. for anxiety change, education, basics, race/ethnicity, HBP, change in life stressors

* * * * * *95% CI Excludes1.0

c

 At baseline 46.7% had UI, 15.3% several days/week  Over first 6 years of SWAN, 14.7% worsened, 32.4%

improved, 52.9% reported no change in frequency

 Change in menopausal status was not associated with

worsening

 Each pound of weight gain significantly increased odds of

worsening and reduced odds of improving

 Odds of improving did not vary by race/ethnicity Waetjen LE, et al, ObGyn 2008  Objectives  Design and Methods

 Results

• Vasomotor symptoms
• Urinary symptoms
• Difficulty sleeping

Kravitz et al, Sleep 2008

EB Gold NAMS Presentation 10/06/2016 11

Adjusted* Odds Ratio *Adj. for age, race/ethnicity, site, depr. & anxiety sx, get up to urinate, pain, life events, psychotropic & sleep meds; **sig. different from premenopause Kravitz et al, Sleep 2008 ** ** ** ** ** ** ** ** ** **  For all 3 types of sleep difficulties, a significant dose-response

relationship was observed with number of days of reporting VMS, similar to Penn Ovarian Aging study (Pien et al 2008; Freeman et al 2015).

 Decreases in annual E2 levels were significantly associated

with trouble falling asleep and waking up several times.

 Increases in annual FSH levels were significantly associated

with waking up several times.

 Significantly more Caucasian women reported waking up

several times and awaking early than other racial/ethnic groups.

Kravitz et al, Sleep 2008  Objectives  Design and Methods

 Results

• Vasomotor symptoms
• Urinary symptoms
• Difficulty sleeping
• Vaginal dryness

Adjusted Odds Ratio

a Adjusted for age, race/ethnicity, education, employment, marital

status, parity, BMI, smoking, physical activity Gold et al Am J Epidemiol 2000 Menopausal Status

EB Gold NAMS Presentation 10/06/2016 12

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 Race/ethnicity

• Afr. Am*

Hispanic* White Chinese Japanese Adjusted Odds Ratio

aAdjusted for age, menopause status, education, parity employment,

marital status, parity, BMI, smoking, physical activity Gold et al Am J Epidemiol 2000  Virtually all symptoms differed by race/ethnicity, adjusted for

differences in socioeconomic and lifestyle factors, suggests symptom risks differ by race/ethnicity

 Vasomotor sx increased most in association with progression

through the menopause transition, followed by sleep difficulties

 Changes in UI largely not related to the menopause transition  BMI related to many outcomes, except vaginal dryness:

• Concurrent BMI, not weight change, most related to VMS
• Weight control should be addressed in a culturally

appropriate manner

The Study of Women‘s Health Across the Nation (SWAN) has grant support from the National Institutes of Health (NIH), DHHS, through the National Institute on Aging (NIA), the National Institute of Nursing Research (NINR), and the NIH Office of Research on Women‘s Health (ORWH) (Grants U01NR004061; U01AG012505, U01AG012535, U01AG012531, U01AG012539, U01AG012546, U01AG012553,U01AG012554, U01AG012495). The content of this presentation is solely the responsibility of the authors and does not necessarily reflect the official views of the NIA, NINR, ORWH, or the NIH.

We wish to express our appreciation too for the women who have participated in SWAN for over 15 years and given of their time and information for the benefit of future generations of midlife women!