Abstract
Objective:
To evaluate if and how race, ethnicity, and socioeconomic status (SES) are associated with the severity of menopause symptoms in a large, diverse sample of women.
Methods:
For this cross-sectional study conducted between March 24, 2019, and January 13, 2023, a total of 68,864 women were enrolled from the Evernow online telehealth platform. Participants underwent a clinical intake survey, which encompassed demographic information, detailed medical questionnaires, and a modified Menopause Rating Scale (MRS). The modified scale was adapted for ease of use online and is available in the supplementary material along with the full intake. Symptom severity was evaluated using a multivariate binomial generalized linear model, accounting for factors such as race, ethnicity, age, BMI, smoking status, bilateral oophorectomy status, and SES. Odds ratios and confidence intervals were calculated based on the linear regression coefficients.
Results:
Of the participants, 67,867 (98.6%) were included in the analysis after excluding outliers and those with unknown oophorectomy status. The majority of respondents identified as White (77.4%), followed by Hispanic (9.0%), Black (6.7%), two or more races/ethnicities (4.4%), Asian (1.2%), Indigenous/First Nations (0.8%), Middle Eastern (0.3%), and South Asian (0.2%). Notably, individuals identifying as Black (hot flashes OR=1.91, 97.5%CI[1.75-2.09], p<.001), Hispanic (skin/hair changes OR=1.58, 97.5%CI[1.45-1.71], p<.001), Indigenous/First Nations (painful sex OR=1.39, 97.5%CI[1.19-2.75], p=.007), Middle Eastern (weight changes OR=2.22, 97.5%CI[1.25-4.37, p=.01), or with two or more races/ethnicities (skin/hair changes OR=1.41, 97.5%CI[1.26-1.58], p<.001) reported higher levels of symptom severity compared to their White counterparts. Conversely, Asian and South Asian participants reported lower symptom severity. Even after incorporating SES into the linear model, racial and ethnic groups with lower SES (Black, Hispanic, Indigenous, and multiple ethnicities) exhibited slight shifts in odds ratios, while maintaining high statistical significance (Black (hot flashes OR=1.87, 97.5%CI[1.72-2.04], p<.001), Hispanic (skin/hair changes OR=1.54, 97.5%CI[1.42-1.68], p<.001), Indigenous/First Nations (painful sex OR=1.74, 97.5%CI[1.17-2.70], p=.009), multiple ethnicities (skin/hair changes OR=1.41, 97.5%CI[1.26-1.58], p<.001)).
Conclusions:
Our study suggests that the relationship between race and ethnicity and the severity of menopause symptoms is not solely explained by differences in SES but is itself an independent factor. Understanding and addressing social, cultural, and economic factors is crucial to reduce disparities in menopausal symptoms.
Keywords: Menopause, Symptoms, Race, Ethnicity, Socioeconomic Status
Summary Sentence:
The relationship between race and ethnicity and menopause is not solely explained by differences in socioeconomic status, but is itself an independent factor.
INTRODUCTION
Menopause is a biological process that occurs in women and gender-diverse individuals as they age, characterized by the cessation of menstrual periods and a decline in hormone production. These hormonal changes are associated with various physical and psychological symptoms, with vasomotor symptoms lasting a median of 7.4 years and having a profound impact on a woman’s quality of life1–4. Race and ethnicity, as well as socioeconomic status, may influence the experience of menopause symptoms.
The Study of Women’s Health Across the Nation (SWAN) is a multiethnic, longitudinal, cohort study of mid-life women. This was the first NIH observational study of its kind and included women from different backgrounds to characterize the menopausal experience of a diverse group of women5. SWAN found that vasomotor symptoms were more prevalent in African-American and Hispanic women, while vaginal dryness was most prevalent in Hispanic women6. Another analysis of the SWAN study found that low-income women reported greater severity and frequency of menopausal symptoms compared to White women and women with higher income; however, these findings should be interpreted with caution, as they can be influenced by factors such as access to healthcare and use of hormone therapy, which varies based on demographic characteristics7,8. Other studies have sought to describe specific symptoms and their relationship to race and ethnicity, but none of them examined all symptoms across the Menopause Rating Scale (MRS)9–13. Further research is needed to explore the relationship between race and ethnicity, socioeconomic status, and menopausal symptoms.
The objective of the current study is to evaluate if and how race, ethnicity, and socioeconomic status are associated with the severity of menopause symptoms in a large, diverse sample of women, expanding on prior research by including granularity in self-reported racial and ethnic groups, and evaluating socioeconomic status as a covariate. The aim is to shed light on the complex role that these factors play in menopausal symptomatology to better inform treatment interventions to improve the quality of life of menopausal women.
METHODS
Survey Development
Evernow is a telehealth platform that provides evidence-based menopause care. Prospective participants must complete a clinical intake survey, which includes demographic information, a medical history questionnaire, and a modified Menopause Rating Scale to determine eligibility and screen for contraindications14. The survey questions are close-ended with answer options provided to participants. Survey question details and answer choices are in the Supplemental Methods.
Participant Eligibility and Recruitment
Participants signed up for care at Evernow, advertised through digital marketing and via word of mouth. Participants usually pay out of pocket for treatment, which varied by the medication recommended by the provider, as well as during the time of the study. Only participants living in the United States were eligible to participate. A total of 68,864 participants were administered a clinical intake survey from March 24, 2019 to January 13, 2023. The survey was administered to prospective participants who self-identified as experiencing one or more symptoms common to perimenopause or menopause. All Evernow prospective participants consented to having their data de-identified and aggregated for the purposes of research. The dataset used is de-identified with no key to re-identify participants available to the researchers participating in the analysis, thus an Institutional Review Board (IRB) review was not required per the University of California San Francisco IRB policies.
Survey Analysis
Survey results were de-identified before the analysis. For symptom severity, a rating of 1 or 2 was classified as “not severe”, and a rating of 3 or 4 was classified as “severe”. Participants who did not experience a given symptom were removed from severity analysis for that symptom. Participants identifying as East Asian, Asian American, or Southeast Asian were categorized as Asian. Those identifying as African-American or African were categorized as Black. Any participant who identified as more than one race or ethnicity were assigned to the Two or More Ethnicities category. Age was separated into those younger than 52 and those 52 or older for demographics and used as a continuous variable in generalized linear models. Smoking status combined those who actively used tobacco via cigarettes, vapes, or other. BMI was stratified by CDC BMI indices of underweight (<18.5), healthy weight (18.5-24.9), overweight (25.0-29.5, obesity (30.0-39.9), and severe obesity (BMI 40.0-59.9) for demographics, and used as a continuous variable in generalized linear models. Neighborhood affluence and disadvantage scores were used as continuous variables. Outliers for age (>70) and BMI (<15 and ≥60) were excluded from analysis. Additionally, those who reported a history of a hysterectomy but oophorectomy status was unknown were excluded.
The study used the National Neighborhood Data Archive (NaNDA) dataset to incorporate socioeconomic status into the analysis15. Affluence and disadvantage scores were used as proxies for socioeconomic status, and were reflective of the time period from 2013-2017. Affluence score reflects the proportion of households with income greater than $75K, proportion of age 16+ employed in professional or managerial occupations, and proportion of adults with a Bachelor’s Degree or higher. Disadvantage score incorporates the proportion of female headed families with children, households with public assistance income or food stamps, families with income below the federal poverty level, and population age 16+ who are unemployed. ZIP codes of survey participants were crosswalked to ZIP Code Tabulation Areas (ZCTAs) using UDS Mapper’s 2017 crosswalk data16, and affluence and disadvantage scores were assigned to survey participants.
Prevalence of severe symptoms for each variable was calculated and tested for multicollinearity before inclusion in the model. Multivariate binomial generalized linear models were created with symptom severity as a dependent variable, and race and ethnicity, affluence score and disadvantage score as the independent variables. Additional models used affluence and disadvantage score as dependent variables and race and ethnicity as an independent variable. Potential confounders that were controlled for in adjusted models included age, BMI, smoking status, bilateral oophorectomy status, affluence, disadvantage scores, and all hysterectomies for unusual periods. White was used as a reference for race and ethnicity. Odds ratios, confidence intervals, and p-values were calculated from the linear regression coefficients. All analyses were performed in R17.
RESULTS
In this study, 68,864 participants completed an online survey. Outliers for BMI, age and participants with unknown oophorectomy status were excluded, leaving 67,867 participants for analysis. The majority identified as White (77.4%), followed by Hispanic (9.0%), Black (6.7%), two or more races/ethnicities (4.4%), Asian (1.2%), Indigenous/First Nations (0.8%), Middle Eastern (0.3%), and South Asian (0.2%) (Table 1). About 45% of participants were under the age of 52, and the majority were overweight or obese (37.9% and 36.5%, respectively). Thirteen percent of participants were smokers. Of the 29.7% of participants who had hysterectomies, 15.8% did not have their ovaries removed, 2.3% reported unilateral oophorectomies, and 11.6% reported bilateral oophorectomies.
TABLE 1.
Demographics
| Characteristic | No. (%) (N=67867) |
|---|---|
| Race/Ethnicity | |
|
| |
| Asian | 790 (1.2) |
| Black | 4579 (6.7) |
| Hispanic/Latinx | 6116 (9.0) |
| Indigenous/First Nations | 531 (0.8) |
| Middle Eastern | 206 (0.3) |
| South Asian | 161 (0.2) |
| Two or More Ethnicities | 2970 (4.4) |
| White | 52514 (77.4) |
|
Age
| |
| <52 | 30568 (45.0) |
| ≥52 | 37299 (55.0) |
|
BMI
| |
| <18.5 | 424 (0.6) |
| 18.5-24.9 | 15137 (22.3) |
| 25-29.9 | 23227 (34.2) |
| 30-39.9 | 24267 (35.8) |
| ≥40 | 4812 (7.1) |
|
Smoking
| |
| Non-Smoker | 59011 (87.0) |
| Smoker | 8856 (13.0) |
|
Hysterectomy
| |
| None | 47734 (70.3) |
| Uterus | 10702 (15.8) |
| Unilateral Oophorectomy | 1582 (2.3) |
| Bilateral Oophorectomy | 7849 (11.6) |
Across diverse racial and ethnic groups, the prevalence of severe menopausal symptoms varied widely and is found in Table 2. Notably, Black individuals consistently reported higher prevalence rates for several symptoms, including hot flashes (80.7%), night sweats (80.0%), sleep disturbances (88.9%), brain fog (76.4%), painful sex (84.6%), and unusual periods (80.8%). Asian participants, on the other hand, tended to experience lower prevalence rates for these symptoms, ranging from 58.7% for increased facial hair/acne to 78.8% for mood swings. Indigenous/First Nations individuals consistently demonstrated higher prevalence rates, particularly for sleep disturbances (87.3%), skin/hair changes (83.4%), and joint/muscular discomfort (84.1%). South Asian participants generally reported lower prevalence rates across symptoms, with a notable low of 47% for heart discomfort.
Table 2.
Percentages of Participants within each Race and Ethnicity Category Experiencing Symptom Presence
| Symptoms | No. (% Present) | |||||||
|---|---|---|---|---|---|---|---|---|
| White | Asian | Black | Hispanic/Latinx | Indigenous/First Nations | Middle Eastern | South Asian | Two or More Ethnicities | |
| Hot Flashes | 34320/52514 (65.4) | 485/790 (61.4) | 3604/4579 (78.7) | 4183/6116 (68.4) | 373/531 (70.2) | 126/206 (61.2) | 86/161 (53.4) | 1993/2970 (67.1) |
| Night Sweats | 34260/52512 (65.2) | 430/790 (54.4) | 3437/4579 (75.1) | 3964/6116 (64.8) | 336/531 (63.3) | 118/206 (57.3) | 92/161 (57.1) | 1960/2970 (66.0) |
| Sleep Disturbances | 41909/52514 (79.8) | 591/790 (74.8) | 3696/4579 (80.7) | 4918/6116 (80.4) | 441/531 (83.1) | 115/206 (75.2) | 129/161 (80.1) | 2393/2970 (80.6) |
| Brain Fog | 41384/52514 (78.8) | 556/790 (70.4) | 3333/4579 (72.8) | 4594/6116 (75.1) | 445/531 (83.8) | 149/206 (72.3) | 114/161 (70.8) | 2366/2970 (79.7) |
| Painful Sex | 19673/52514 (37.5) | 322/790 (40.8) | 1737/4579 (37.9) | 2407/6116 (39.4) | 211/531 (39.7) | 79/206 (38.3) | 57/161 (35.4) | 1113/2970 (37.5) |
| Vaginal Dryness | 29960/52514 (57.0) | 480/790 (60.8) | 2849/4579 (62.2) | 3766/6116 (61.6) | 343/531 (64.6) | 126/206 (61.2) | 87/161 (54.0) | 1661/2970 (55.9) |
| Urgent Bathroom Trips | 15337/52514 (29.2) | 198/790 (25.1) | 1289/4579 (28.2) | 2020/6116 (33.0) | 198/531 (37.3) | 55/206 (26.7) | 40/161 (24.8) | 954/2970 (32.1) |
| Mood Swings | 31459/52514 (59.9) | 442/790 (55.9) | 2589/4579 (58.7) | 3788/6116 (61.9) | 353/531 (66.5) | 113/206 (54.9) | 90/161 (55.9) | 1809/2970 (60.9) |
| Unusual Period | 9375/52514 (17.9) | 183/790 (23.2) | 682/4579 (14.9) | 1230/6116 (20.1) | 99/531 (18.6) | 45/206 (21.8) | 39/161 (24.2) | 655/2970 (22.1) |
| Skin/Hair Changes | 32119/52514 (61.2) | 474/790 (60.0) | 2698/4579 (58.9) | 4117/6116 (67.3) | 366/531 (68.9) | 124/206 (60.2) | 107/161 (66.5) | 2016/2970 (67.9) |
| Increased Facial Hair/Acne | 18452/52514 (35.2) | 143/790 (18.2) | 1602/4579 (35.0) | 1859/6116 (30.4) | 189/531 (35.6) | 53/206 (25.7) | 43/161 (26.7) | 1054/2970 (35.5) |
| Fatigue/Low Energy | 43571/52514 (83.0) | 569/790 (72.0) | 3465/4579 (75.7) | 4973/6116 (81.3) | 473/531 (89.0) | 143/206 (69.4) | 115/161 (71.4) | 2482/2970 (83.6) |
| Weight Changes | 42440/52514 (80.8) | 566/790 (71.6) | 3414/4579 (74.6) | 4867/6116 (79.6) | 437/531 (82.3) | 153/206 (74.3) | 123/161 (76.4) | 2393/2970 (80.6) |
| Anxiety/Depression | 33024/52514 (62.9) | 410/790 (51.9) | 2619/4579 (57.2) | 3810/6116 (62.3) | 364/531 (68.5) | 121/206 (58.7) | 86/161 (53.4) | 1871/2970 (63.0) |
| Heart Discomfort | 16838/52514 (32.1) | 237/790 (30.0) | 1420/4579 (32.3) | 1973/6116 (32.3) | 198/531 (37.3) | 67/206 (32.5) | 51/161 (31.7) | 1039/2970 (35.0) |
| Joint/Muscular Discomfort | 34279/52514 (65.3) | 485/790 (61.4) | 2984/4579 (65.2) | 4211/6116 (68.9) | 364/531 (68.5) | 126/206 (61.2) | 94/161 (58.4) | 2050/2970 (69.0) |
For our analysis we determined the odds ratios of severe symptoms (score 3 or 4), for each racial and ethnic group, using White as the reference. Within this analysis participants identifying as Black, Hispanic, Indigenous/First Nations, Middle Eastern, and two or more races/ethnicities, reported greater symptom severity as compared to those identifying as White for many symptoms (Figure 1, Table 3, and Supplemental Table 1). For almost all symptoms, participants identifying as Asian or South Asian were found to have similar symptom severity than White participants. The symptom severity profile, however, varied across races and ethnicities. Those identifying as Black had the highest odds ratios for severe vasomotor symptoms (hot flashes (OR=1.91, 97.5% CI[1.75-2.09]), night sweats (OR=1.87, 97.5% CI[1.72-2.04]), sleep disturbances (OR= 1.67, 97.5% CI[1.50-1.86]), Table 3) and increased facial hair/acne (OR=1.47, 97.5% CI[1.30-1.65]). When BMI is accounted for, weight changes were least severe for Black women (OR=0.90, 97.5% CI[0.80-1.01]). All other symptoms for those identifying as Black had significant higher odds ratios than White participants. Indigenous/First Nations participants experienced the most severe anxiety/depression (OR=1.70, 97.5% CI[1.30-2.24]), painful sex (OR=1.77, 97.5% CI[1.19-2.75]), and joint/muscular discomfort (OR=1.72, 97.5% CI[1.28-2.34]). Middle Eastern and Indigenous/First Nations participants had the highest severity for brain fog (OR=1.76, 97.5% CI[1.18-2.70] and OR=1.48, 97.5% CI[1.17-1.88]) and fatigue/low energy (OR=1.91, 97.5% CI[1.12-3.55] and OR=1.64, 97.5%CI[1.20-2.31]). Hispanic identifying women had more severe symptoms than White women, with all symptoms having statistically significant (p<0.01) odds ratios greater than 1.15. Those who identified as two or more races/ethnicities had slightly more severe symptoms than those identifying as White, except for painful sex, night sweats, vaginal dryness, increased facial hair/acne and weight changes which were statistically similar to the severity of White participants. Of note, Asian women experienced significantly worse heart discomfort than White participants (OR=1.43, 97.5% CI[1.10-1.86]) and South Asian respondents had the highest severity for weight changes of all races and ethnicities (OR=2.96, 97.5% CI[1.53-6.62]) although statistically comparable to Middle Eastern (OR=2.22, 97.5% CI[1.25-4.37]). When calculating the prevalence of symptoms by either presence or absence, we found that elevated symptom severity did not always correspond with greater symptom frequency (Table 2).
FIG. 1.
Adjusted Odds Ratios for Symptom Severity by Race and Ethnicity. Adjusted for age, BMI, smoking, and oophorectomy status. For unusual periods adjusted for all hysterectomy. Reference = White. Symptom Scores: 1-2 = Not Severe, 3-4 = Severe. Arrows indicate 95% CI extends beyond range.
TABLE 3.
Adjusted Odds Ratios for Symptom Severity by Race and Ethnicitya
| Symptom | Odds Ratio (97.5% CI) P value | ||||||
|---|---|---|---|---|---|---|---|
| Asian | Black | Hispanic/Latinx | Indigenous/First Nations | Middle Eastern | South Asian | Two or More Ethnicities | |
| Hot Flashes | 0.93 (0.77-1.13) P=.48 |
1.91 (1.75-2.09)
P<.001 |
1.27 (1.19-1.37)
P<.001 |
1.27 (1.01-1.61) P=.04 |
1.13 (0.78-1.67) P=.54 |
1.35 (0.85-2.20) P=.22 |
1.25 (1.13-1.38)
P<.001 |
| Night Sweats | 0.86 (0.70-1.05) P=.13 |
1.87 (1.72-2.04)
P<.001 |
1.21 (1.12-1.30)
P<.001 |
1.37 (1.07-1.77)
P=.01 |
1.39 (0.93-2.14) P=.12 |
1.27 (0.82-2.04) P=.30 |
1.09 (0.99-1.20) P=.09 |
| Sleep Disturbances | 0.89 (0.73-1.09) P=.26 |
1.67 (1.50-1.86)
P<.001 |
1.29 (1.18-1.40)
P<.001 |
1.44 (1.10-1.93)
P=.01 |
0.80 (0.55-1.19) P=.26 |
0.88 (0.59-1.36) P=.55 |
1.21 (1.09-1.36)
P=.001 |
| Brain Fog | 0.98 (0.82-1.18) P=.86 |
1.19 (1.10-1.30)
P<.001 |
1.15 (1.07-1.24)
P<.001 |
1.48 (1.17-1.88)
P=.001 |
1.76 (1.18-2.70)
P=.007 |
0.99 (0.67-1.50) P=.96 |
1.18 (1.07-1.30)
P=.001 |
| Painful Sex | 1.05 (0.80-1.39) P=.73 |
1.38 (1.20-1.59)
P<.001 |
1.39 (1.24-1.56)
P<.001 |
1.77 (1.19-2.75)
P=.007 |
1.36 (0.76-2.65) P=.34 |
1.15 (0.63-2.30) P=.66 |
0.96 (0.83-1.12) P=.58 |
| Vaginal Dryness | 1.06 (0.88-1.28) P=.56 |
1.34 (1.23-1.47)
P<.001 |
1.41 (1.31-1.53)
P<.001 |
1.41 (1.11-1.81)
P=.006 |
1.52 (1.02-2.31) P=.04 |
1.04 (0.67-1.64) P=.86 |
1.11 (1.00-1.24) P=.05 |
| Urgent Bathroom Trips | 1.02 (0.76-1.38) P=.89 |
1.43 (1.25-1.63)
P<.001 |
1.43 (1.29-1.60)
P<.001 |
1.33 (0.97-1.85) P=.08 |
0.99 (0.57-1.78) P=.98 |
1.57 (0.80-3.29) P=.21 |
1.23 (1.07-1.42)
P=.005 |
| Mood Swings | 1.05 (0.86-1.29) P=.65 |
1.53 (1.39-1.68)
P<.001 |
1.47 (1.35-1.59)
P<.001 |
1.35 (1.05-1.74)
P=.02 |
1.29 (0.86-2.01) P=.23 |
0.88 (0.57-1.37) P=.56 |
1.21 (1.09-1.35)
P<.001 |
| Unusual Period | 0.89 (0.65-1.22) P=.45 |
1.65 (1.35-2.01)
P<.001 |
1.47 (1.28-1.70)
P<.001 |
1.30 (0.83-2.11) P=.27 |
0.90 (0.49-1.76) P=.75 |
1.11 (0.57-2.33) P=.78 |
1.35 (1.12-1.63)
P=.002 |
| Skin/Hair Changes | 1.25 (1.00-1.55) P=.05 |
1.65 (1.48-1.83)
P<.001 |
1.58 (1.45-1.71)
P<.001 |
1.33 (1.03-1.73) P=.03 |
1.59 (1.02-2.60) P=.05 |
1.26 (0.82-2.02) P=.31 |
1.41 (1.26-1.58)
P<.001 |
| Increased Facial Hair/Acne | 0.83 (0.60-1.17) P=.29 |
1.47 (1.30-1.65)
P<.001 |
1.16 (1.05-1.29)
P=.004 |
1.40 (1.01-1.96) P=.04 |
1.35 (0.76-2.49) P=.32 |
1.27 (0.68-2.49) P=.46 |
1.17 (1.02-1.34) P=.03 |
| Fatigue/Low Energy | 0.85 (0.69-1.06) P=.13 |
1.16 (1.04-1.29)
P=.01 |
1.33 (1.21-1.46)
P<.001 |
1.64 (1.20-2.31)
P=.003 |
1.95 (1.14-3.63)
P=.02 |
0.97 (0.61-1.62) P=.90 |
1.17 (1.04-1.33)
P=.01 |
| Weight Changes | 1.10 (0.87-1.40) P=43 |
0.90 (0.80-1.01) P=.07 |
1.31 (1.18-1.45)
P<.001 |
1.73 (1.20-2.60)
P=.005 |
2.22 (1.25-4.37)
P=.01 |
2.96 (1.53-6.62)
P=.003 |
1.04 (0.91-1.18) P=.56 |
| Anxiety/Depression | 1.02 (0.82-1.27) P=.87 |
1.32 (1.20-1.45)
P<.001 |
1.32 (1.22-1.43)
P<.001 |
1.70 (1.30-2.24)
P<.001 |
1.19 (0.80-1.82) P=.41 |
0.88 (0.56-1.40) P=.56 |
1.13 (1.02-1.26)
P=.02 |
| Heart Discomfort |
1.43 (1.10-1.86)
P=.007 |
1.59 (1.42-1.78)
P<.001 |
1.40 (1.27-1.54)
P<.001 |
1.54 (1.16-2.06)
P=.003 |
1.38 (0.85-2.26) P=.19 |
0.99 (0.56-1.71) P=.96 |
1.20 (1.06-1.36)
P=.005 |
| Joint/Muscular Discomfort | 0.88 (0.73-1.08) P=.22 |
1.44 (1.30-1.60)
P<.001 |
1.37 (1.26-1.49)
P<.001 |
1.72 (1.28-2.34)
P<.001 |
0.92 (0.62-1.39) P=.67 |
1.18 (0.75-1.94) P=.49 |
1.23 (1.10-1.38)
P<.001 |
Severity: 1-2 = Not Severe, 3-4= Severe. Adjusted for Age, BMI, Smoking, Bilateral Oophorectomy. (For Unusual Period adjusted for Hysterectomy instead of bilateral oophorectomy). Significant (p<.05) values bolded.
As race and ethnicity are social constructs, we investigated the impact of socioeconomic status (SES) on symptom severity. We used the affluence and disadvantage scores for the ZCTA corresponding to each respondent as a proxy for their SES. Participants in our survey had lower disadvantage and higher affluence scores compared to the general population. The mean affluence score for survey participants was 0.394 and the United States population was 0.326 (Supplemental Table 2). For disadvantage scores, survey participants had a mean of 0.096, while the general population was 0.099. We stratified affluence and disadvantage scores into four categories, with 1 being the lowest and 4 the highest. Across all symptoms, higher affluence score corresponded to lower symptom severity, with varying magnitudes of impact for each symptom (Supplemental Fig. 1 and Supplemental Table 3). The results when analyzing symptom severity by disadvantage score were similar (Supplemental Fig. 2 and Supplemental Table 4). Of note, our analysis showed that those identifying as Black, Hispanic, and Indigenous/First Nations were more likely to experience lower affluence (Supplemental Fig. 3 and Supplemental Table 5). Those identifying as two or more races/ethnicities, Black, or Hispanic/Latinx had elevated prevalence of high disadvantage. Higher affluence scores were found for those identifying as Asian, South Asian, and Middle Eastern as compared to White.
To account for the impact of socioeconomic status in driving the disparities in symptom severity between racial and ethnic groups, after testing for multicollinearity we included affluence scores as a confounding variable in our linear model, and present this analysis separately in Figure 2 (and Table 4). For the races and ethnicities that had lower socioeconomic status (Black, Hispanic, Indigenous, and two or more ethnicities), most symptoms had lowered odds ratios when accounting the affluence score. Overall, these shifts in odds ratios were small. The largest OR decreases were by 0.09, for night sweats in Black (OR=1.78, 97.5%CI[1.63-1.95]) and Hispanic individuals (OR=1.18, 97.5%CI[1.10-1.27]), fatigue/low energy for Indigenous/First Nations women (OR=1.55, 97.5% CI[1.13-2.18]). For those identifying as two or more ethnicities, most odds ratios were minimally affected, with unusual period having the greatest decrease of 0.02 (OR=1.33, 97.5%CI[1.11-1.61]). The statistical significance of odds ratios was minimally impacted, with only fatigue/low energy for Black women becoming non-significant. Interestingly, adjusting for affluence score increased the odds ratios for all symptoms among those with higher socioeconomic status: Asian, Middle Eastern, and South Asian. For those identifying as Asian, most symptoms remained non-significant with the exception of heart discomfort (OR=1.49, 97.5% CI[1.15-1.93]) remaining significant (p=0.003) and skin/hair changes (OR=1.29, 97.5% CI[1.04-1.60]) becoming significant (p=0.02). For those identifying as Middle Eastern, brain fog (OR=1.89, 97.5% CI[1.27-2.91]), fatigue/low energy (OR=2.20, 97.5% CI[1.29-4.09]), and weight changes (OR=2.26, 97.5% CI[1.27-4.46]) all remained statistically significant, while vaginal dryness (OR=1.62, 97.5% CI[1.09-2.47]) and skin/hair changes (OR=1.70, 97.5% CI[1.09-2.78]), became significant. For South Asian women, weight change (OR=3.04, 97.5% CI[1.57-6.80]) was still statistically significant and all other symptoms remained insignificant. Similar results were observed using disadvantage score in the model in place of affluence score (Supplemental Table 6).
FIG. 2.
Adjusted Odds Ratios for Symptom Severity by Race and Ethnicity Including Affluence. Adjusted for Affluence score, age, BMI, smoking, and oophorectomy status. For unusual periods adjusted for all hysterectomy. Reference = White. Symptom Scores: 1-2 = Not Severe, 3-4 = Severe. Arrows indicate 95% CI extends beyond range.
TABLE 4.
Adjusted Odds Ratios for Symptom Severity by Race and Ethnicity including Affluence Scorea
| Odds Ratio (97.5% CI) P value | |||||||
|---|---|---|---|---|---|---|---|
| Symptoms | Asian | Black | Hispanic/Latinx | Indigenous/First Nations | Middle Eastern | South Asian | Two or More Ethnicities |
| Hot Flashes | 0.97 (0.81-1.18) P=.78 |
1.87 (1.72-2.04)
P<.001 |
1.25 (1.16-1.34)
P<.001 |
1.23 (0.98-1.56) P=.08 |
1.23 (0.85-1.82) P=.29 |
1.50 (0.94-2.45) P=.09 |
1.24 (1.12-1.38)
P<.001 |
| Night Sweats | 0.86 (0.71-1.06) P=.14 |
1.78 (1.63-1.95)
P<.001 |
1.18 (1.10-1.27)
P<.001 |
1.35 (1.05-1.74)
P=.02 |
1.41 (0.94-2.17) P=.11 |
1.30 (0.83-2.08) P=.26 |
1.07 (0.97-1.19) P=.17 |
| Sleep Disturbances | 0.92 (0.75-1.13) P=.40 |
1.65 (1.48-1.83)
P<.001 |
1.26 (1.16-1.37)
P<.001 |
1.40 (1.06-1.87)
P=.02 |
0.86 (0.60-1.29) P=.45 |
0.97 (0.65-1.51) P=.90 |
1.21 (1.08-1.36)
P<.001 |
| Brain Fog | 1.02 (0.85-1.22) P=.86 |
1.16 (1.07-1.27)
P<.001 |
1.13 (1.05-1.21)
P<.001 |
1.43 (1.14-1.82)
P=.003 |
1.89 (1.27-2.91)
P=.003 |
1.10 (0.74-1.67) P=.64 |
1.17 (1.07-1.29) P=.001 |
| Painful Sex | 1.07 (0.82-1.42) P=.63 |
1.36 (1.18-1.56)
P<.001 |
1.37 (1.22-1.54)
P<.001 |
1.74 (1.17-2.70)
P=.009 |
1.41 (0.79-2.76) P=.28 |
1.22 (0.66-2.43) P=.54 |
0.95 (0.82-1.11) P=.53 |
| Vaginal Dryness | 1.09 (0.90-1.32) P=.38 |
1.32 (1.21-1.44)
P<.001 |
1.39 (1.29-1.50)
P<.001 |
1.38 (1.09-1.77)
P=.01 |
1.62 (1.09-2.47)
P=.02 |
1.13 (0.73-1.79) P=.58 |
1.11 (1.00-1.24) P=.05 |
| Urgent Bathroom Trips | 1.05 (0.79-1.42) P=.73 |
1.40 (1.22-1.60)
P<.001 |
1.41 (1.27-1.57)
P<.001 |
1.30 (0.95-1.81) P=.12 |
1.07 (0.62-1.92) P=.81 |
1.73 (0.88-3.63) P=.13 |
1.23 (1.06-1.42)
P=.006 |
| Mood Swings | 1.09 (0.89-1.34) P=.45 |
1.49 (1.35-1.64)
P<.001 |
1.44 (1.33-1.56)
P<.001 |
1.31 (1.02-1.69)
P=.03 |
1.42 (0.94-2.20) P=.13 |
1.00 (0.65-1.56) P=.86 |
1.21 (1.09-1.35)
P<.001 |
| Unusual Period | 0.90 (0.66-1.24) P=.52 |
1.60 (1.32-1.96)
P<.001 |
1.43 (1.24-1.65)
P<.001 |
1.25 (0.80-2.04) P=.34 |
0.95 (0.51-1.85) P=.87 |
1.21 (0.62-2.54) P=.60 |
1.33 (1.11-1.61)
P=.003 |
| Skin/Hair Changes |
1.29 (1.04-1.60)
P=.02 |
1.61 (1.45-1.80)
P<.001 |
1.54 (1.42-1.68)
P<.001 |
1.28 (1.00-1.68) P=.06 |
1.70 (1.09-2.78)
P=.02 |
1.40 (0.90-2.24) P=.15 |
1.41 (1.26-1.58)
P<.001 |
| Increased Facial Hair/Acne | 0.87 (0.62-1.22) P=.42 |
1.44 (1.28-1.63)
P<.001 |
1.15 (1.04-1.28)
P=.008 |
1.36 (0.99-1.91) P=.06 |
1.43 (0.81-2.66) P=.23 |
1.42 (0.76-2.78) P=.28 |
1.16 (1.01-1.33) P=.03 |
| Fatigue/Low Energy | 0.90 (0.73-1.12) P=.35 |
1.12 (1.00-1.25) P=.05 |
1.30 (1.18-1.42)
P<.001 |
1.55 (1.13-2.18)
P=.009 |
2.20 (1.29-4.09)
P=.007 |
1.16 (0.73-1.94) P=.55 |
1.17 (1.03-1.32)
P=.01 |
| Weight Changes | 1.11 (0.88-1.41) P=.39 |
0.89 (0.79-1.01) P=.06 |
1.31 (1.18-1.45)
P<.001 |
1.72 (1.19-2.58)
P=.006 |
2.26 (1.27-4.46)
P=.01 |
3.04 (1.57-6.80)
P=.003 |
1.04 (0.92-1.19) P=.54 |
| Anxiety/Depression | 1.06 (0.86-1.32) P=.57 |
1.29 (1.17-1.42)
P<.001 |
1.30 (1.19-1.40)
P<.001 |
1.63 (1.25-2.16)
P<.001 |
1.30 (0.87-1.99) P=.21 |
0.99 (0.64-1.58) P=.97 |
1.13 (1.01-1.26)
P=.03 |
| Heart Discomfort |
1.49 (1.15-1.93)
P=.003 |
1.56 (1.39-1.74)
P<.001 |
1.37 (1.25-1.51)
P<.001 |
1.50 (1.12-2.00)
P=.006 |
1.50 (0.93-2.46) P=.10 |
1.08 (0.62-1.88) P=.79 |
1.20 (1.06-1.36)
P=.005 |
| Joint/Muscular Discomfort | 0.92 (0.76-1.13) P=.43 |
1.41 (1.27-1.57)
P<.001 |
1.34 (1.23-1.45) P<.001 |
1.65 (1.23-2.25)
P=.001 |
1.02 (0.69-1.55) P=.90 |
1.39 (0.88-2.28) P=.18 |
1.23 (1.10-1.38)
P<.001 |
Severity: 1-2 = Not Severe, 3-4 = Severe. Adjusted for Age, BMI, Smoking, Bilateral Oophorectomy, and Affluence Score. (For Unusual Period adjusted for Hysterectomy instead of bilateral oophorectomy). Significant (p<.05) values bolded.
DISCUSSION
Our study is to our knowledge the largest and most comprehensive to date to demonstrate that the association between race and ethnicity and the severity of menopause symptoms is not solely explained by differences in socioeconomic status but is itself an independent factor. After controlling for several proxies of socioeconomic status, which themselves have some limitations, self-identified Black women, Hispanic, Indigenous, or multi-racial or multi-ethnic groups have a higher prevalence of worse menopause symptom severity. Our findings add to a growing body of evidence highlighting the strong connection between race and ethnicity, along with a lower socioeconomic status, and the severity of menopause symptoms5.
Several studies have shown there are racial and ethnic differences in the prevalence and severity of symptoms associated with the menopausal transition, with Black and Hispanic women reporting worse symptoms5. Prior literature has suggested women from lower socioeconomic backgrounds are more likely to experience worse menopausal symptoms than those from higher socioeconomic backgrounds6. To explore this further, our study aimed to determine whether adjusting for SES would eliminate racial and ethnic disparities in menopausal symptomatology. Even after controlling for proxies of SES, Black and Hispanic women still reported worse menopausal symptoms than non-Hispanic White women. This suggests factors outside of SES may be contributing to disparities in symptomatology.
Limited data exists on the menopause experiences of Indigenous women, but a systematic review revealed substantial variability in knowledge and symptom experience18. Indigenous women often experience an earlier onset of menopause attributed to factors such as poor nutrition and a harsher lifestyle. Our study suggests Indigenous women experience more severe menopause symptoms than White women, mirroring the symptom severity of Black and Hispanic women. Given limited access to care and increased discrimination, these findings highlight the need for further research and healthcare interventions to address the disparities in menopause symptom severity experienced by Indigenous women.
The menopause experiences of South Asian women living in the United States have received limited attention in the literature. A study conducted in Australia, which has a large South Asian population, found that women of Indian descent had a lower prevalence of classical menopausal symptoms compared to Whites, but reported higher levels of physical and psychological symptoms than the typical vasomotor symptoms19. Our study sheds new light on the symptom severity of South Asian women living in the United States, suggesting they have a similar level of symptomatology to White women, except for weight changes. South Asian women rated weight changes three times more severe than White women. The reason for this difference remains unclear. A plausible hypothesis is an increased rate of metabolic dysfunction in these women, as seen in those with polycystic ovary syndrome20. Further research into this relationship is necessary to gain a deeper understanding of the menopause experiences of South Asian women.
The severity of menopausal symptoms varies among women and can be influenced by genetics, lifestyle, and health factors21. Cultural, social, and economic factors affecting access to healthcare and ability to manage symptoms can also contribute to disparities. For example, a study by Finley et al. suggested that women of moderate to high income were three times more likely than those of low income to use hormone therapy22. It is also well documented in other areas of healthcare that Black and Hispanic women face significant barriers to accessing care and are more likely to discontinue treatment, barriers that can’t necessarily be controlled for with proxies for SES23. Differences in cultural attitudes, beliefs, and practices regarding menopause and aging may also contribute to disparities in symptom experience among women from diverse communities. Hormonal profiles may also differ among women of different ethnicities24.
This study highlights that while there is a correlation between race and ethnicity and the severity of menopausal symptoms, further research is needed to better understand the complexities of this issue, including cultural, lifestyle, genetic, or other factors. The authors caution against attributing differences solely to race and ethnicity, and highlights the need for a more nuanced understanding of this important health issue.
This is to our knowledge the largest study to date investigating menopause symptom severity by race or ethnicity when controlling for income and includes a robust sample size to examine symptomatology among different groups. Other strengths include controlling for important covariates, such as oophorectomy and hysterectomy status, which are contributors to menopausal symptoms, and incorporating a proxy for SES into the model as a covariate.
Limitations include the fact that all participants were English-speaking women who sought care through the Evernow platform, so the results may not represent all individuals across different racial and ethnic groups nor should be taken to represent prevalence rates of symptomatology. Participants must have had at least one MRS symptom to be included in this study, which may bias the findings and generalizability of the findings. Participants self-reported their history and symptoms which can introduce bias. Although we report it, we did not examine symptom prevalence in depth as the focus of this study was on symptom severity. Furthermore, the use of ZIP code data to control for SES may not accurately reflect an individual’s true SES and other measures of SES that were not accounted for could also be determinants of symptom severity, such as structural racism or geographic issues. Finally, it is important to note that this study only suggests associations and does not allow for the evaluation of causal relationships, necessitating further research to fully understand the findings.
CONCLUSIONS
This study suggests that the severity of menopause symptoms is associated with race and ethnicity, regardless of SES, and highlights the importance of considering race and ethnicity as independent factors in menopause experience. Importantly, certain ethnic groups may be at increased risk for adverse cardiovascular events due to frequent and persistent VMS and may require risk reduction strategies25. The study emphasizes the need to address social, cultural, and economic factors that contribute to disparities in menopausal symptoms and improve access to healthcare for all women with symptoms during this life stage.
Supplementary Material
Sources of Financial Support:
No specific funding was received for conducting or participating in this study. As disclosed, some authors are medical advisors or employees of Evernow and receive separate compensation for these roles.
Footnotes
Conflict of interest/financial disclosures: At the time of this study, Dr. Leah Millheiser was Chief Medical Officer of Evernow. Dr. Eduardo Hariton is a paid medical advisor of Evernow. Alicia Jackson, PhD is Founder and CEO of Evernow. Jennifer Garrison, PhD is Chief Science Advisor of Evernow. Evernow is an online telehealth platform providing menopause treatment.
Previous Presentations: Earlier versions of this work were presented at the North American Menopause Society meeting in October of 2022 in Atlanta, GA and at the Reproductive Aging Conference in May of 2023 in Chicago, IL.
References:
- 1.Mitchell ES, Woods NF. Symptom experiences of midlife women: observations from the Seattle Midlife Women’s Health Study. Maturitas. 1996;25(1):1–10. doi: 10.1016/0378-5122(96)01047-x [DOI] [PubMed] [Google Scholar]
- 2.Santoro N, Epperson CN, Mathews SB. Menopausal Symptoms and Their Management. Endocrinol Metab Clin North Am. 2015;44(3):497–515. doi: 10.1016/j.ecl.2015.05.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Avis NE, Crawford SL, Greendale G, et al. Duration of Menopausal Vasomotor Symptoms Over the Menopause Transition. JAMA Internal Medicine. 2015;175(4):531–539. doi: 10.1001/jamainternmed.2014.8063 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Avis NE, Ory M, Matthews KA, Schocken M, Bromberger J, Colvin A. Health-related quality of life in a multiethnic sample of middle-aged women: Study of Women’s Health Across the Nation (SWAN). Med Care. 2003;41(11):1262–1276. doi: 10.1097/01.MLR.0000093479.39115.AF [DOI] [PubMed] [Google Scholar]
- 5.Green R, Santoro N. Menopausal symptoms and ethnicity: the Study of Women’s Health Across the Nation. Womens Health (Lond). 2009;5(2):127–133. doi: 10.2217/17455057.5.2.127 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gold EB, Block G, Crawford S, et al. Lifestyle and demographic factors in relation to vasomotor symptoms: baseline results from the Study of Women’s Health Across the Nation. Am J Epidemiol. 2004;159(12):1189–1199. doi: 10.1093/aje/kwh168 [DOI] [PubMed] [Google Scholar]
- 7.Pershad A, Morris JM, Pace D, Khanna P. Racial disparities in menopausal hormone therapy acceptance: a pilot study. Menopause. 2022;29(11):1263. doi: 10.1097/GME.0000000000002061 [DOI] [PubMed] [Google Scholar]
- 8.Thurston RC, Joffe H. Vasomotor symptoms and menopause: findings from the Study of Women’s Health across the Nation. Obstet Gynecol Clin North Am. 2011;38(3):489–501. doi: 10.1016/j.ogc.2011.05.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bromberger JT, Schott LL, Avis NE, et al. Psychosocial and health-related risk factors for depressive symptom trajectories among midlife women over 15 years: Study of Women’s Health Across the Nation (SWAN). Psychol Med. 2019;49(2):250–259. doi: 10.1017/S0033291718000703 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dusendang JR, Reeves AN, Karvonen-Gutierrez CA, Herman WH, Ylitalo KR, Harlow SD. The association between perceived discrimination in midlife and peripheral neuropathy in a population-based cohort of women: the Study of Women’s Health Across the Nation. Annals of Epidemiology. 2019;37:10–16. doi: 10.1016/j.annepidem.2019.07.012 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Swanson LM, Hood MM, Hall MH, et al. Associations between sleep and cognitive performance in a racially/ethnically diverse cohort: the Study of Women’s Health Across the Nation. Sleep. 2021;44(2):zsaa182. doi: 10.1093/sleep/zsaa182 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Harlow SD, Burnett-Bowie SAM, Greendale GA, et al. Disparities in Reproductive Aging and Midlife Health between Black and White women: The Study of Women’s Health Across the Nation (SWAN). Womens Midlife Health. 2022;8(1):3. doi: 10.1186/s40695-022-00073-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gold EB, Colvin A, Avis N, et al. Longitudinal analysis of the association between vasomotor symptoms and race/ethnicity across the menopausal transition: study of women’s health across the nation. Am J Public Health. 2006;96(7):1226–1235. doi: 10.2105/AJPH.2005.066936 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Heinemann K, Ruebig A, Potthoff P, et al. The Menopause Rating Scale (MRS) scale: A methodological review. Health Qual Life Outcomes. 2004;2:45. doi: 10.1186/1477-7525-2-45 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Melendez R, Clarke P, Khan A, Gomez-Lopez I, Li M, Chenoweth M. National Neighborhood Data Archive (NaNDA): Socioeconomic Status and Demographic Characteristics of ZIP Code Tabulation Areas, United States, 2008-2017. 2020. doi: 10.3886/E120462V1 [DOI] [Google Scholar]
- 16.UDS Mapper Zip Code to ZCTA Crosswalk. Available at: https://udsmapper.org/zip-code-to-zcta-crosswalk/. Accessed December 13, 2022.
- 17.R Core Team. R: A language and environment for statistical computing. 2021. Available at: https://www.R-project.org/
- 18.Chadha N, Chadha V, Ross S, Sydora BC. Experience of menopause in aboriginal women: a systematic review. Climacteric. 2016;19(1):17–26. doi: 10.3109/13697137.2015.1119112 [DOI] [PubMed] [Google Scholar]
- 19.Hafiz I, Liu J, Eden J. A quantitative analysis of the menopause experience of Indian women living in Sydney. Aust N Z J Obstet Gynaecol. 2007;47(4):329–334. doi: 10.1111/j.1479-828X.2007.00746.x [DOI] [PubMed] [Google Scholar]
- 20.Chahal N, Quinn M, Jaswa EA, Kao CN, Cedars MI, Huddleston HG. Comparison of metabolic syndrome elements in White and Asian women with polycystic ovary syndrome: results of a regional, American cross-sectional study. F S Rep. 2020;1(3):305–313. doi: 10.1016/j.xfre.2020.09.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Richard-Davis G, Wellons M. Racial and ethnic differences in the physiology and clinical symptoms of menopause. Semin Reprod Med. 2013;31(5):380–386. doi: 10.1055/s-0033-1348897 [DOI] [PubMed] [Google Scholar]
- 22.Finley C, Gregg EW, Solomon LJ, Gay E. Disparities in Hormone Replacement Therapy Use by Socioeconomic Status in a Primary Care Population. Journal of Community Health. 2001;26(1):39–50. doi: 10.1023/A:1026537114638 [DOI] [PubMed] [Google Scholar]
- 23.Roberts MC, Wheeler SB, Reeder-Hayes K. Racial/Ethnic and socioeconomic disparities in endocrine therapy adherence in breast cancer: a systematic review. Am J Public Health. 2015;105 Suppl 3(Suppl 3):e4–e15. doi: 10.2105/AJPH.2014.302490 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Kotlyar AM, Seifer DB. Ethnicity/Race and Age-Specific Variations of Serum AMH in Women—A Review. Front Endocrinol (Lausanne). 2021;11:593216. doi: 10.3389/fendo.2020.593216 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Thurston RC, Aslanidou Vlachos HE, Derby CA, et al. Menopausal Vasomotor Symptoms and Risk of Incident Cardiovascular Disease Events in SWAN. Journal of the American Heart Association. 2021;10(3):e017416. doi: 10.1161/JAHA.120.017416 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.