Brief Report: Under-Identification of Symptomatic Menopause in Publicly-Insured Autistic People

Abstract

Menopause is a normal part of aging and in the general population is associated with chronic conditions that impact health, mortality, and well-being. Menopause is experienced differently by autistic individuals, although no studies have investigated this topic in a large sample. The purpose of this study was to investigate rates of, and factors associated with symptomatic menopause among autistic individuals and to identify the prevalence of co-occurring conditions in symptomatic individuals. We included autistic females aged 35–70 years enrolled for 10 + months in 2014–2016 Medicare and/or Medicaid (n = 26,904), excluding those with gender dysphoria. Those with symptomatic menopause were compared to a non-symptomatic reference group on demographic, enrollment characteristics, and co-occurring conditions through logistic regression. Approximately 4% of publicly-insured autistic females aged 46–70 years had symptomatic menopause in their medical records. Intellectual disability was associated with a lower likelihood of symptomatic menopause, and being Medicare-enrolled or dual-enrolled was associated with higher likelihood of having symptomatic menopause recorded. In adjusted models, rates of ADHD, anxiety and depressive disorders, headache/migraine, altered sensory experiences, altered sexual function, and sleep disturbance were significantly higher in the symptomatic menopause sample compared to the reference group. More work to better support autistic women in discussing menopausal symptoms and co-occurring conditions with primary care providers is needed, particularly among those for whom self-report of symptoms are more challenging to ascertain. Factors associated with specific types of health care coverage warrant greater investigation to support better identification.

Menopause is a normal part of aging for individuals female at birth with approximately 90% of females experiencing menopause by age 56 (Shifren et al., 2014). Importantly, age at menopause in the general population is associated with chronic conditions that impact health, mortality, and well-being (El-Khoudary et al., 2020; Shifren et al., 2014) with recent literature linking early menopause (age < 45 years) to cardiovascular disease and fatal congestive heart disease (El-Khoudary et al., 2020) and later menopause with hormone treatment linked to increased breast cancer risk (Collaborative Group on Hormonal Factors in Breast Cancer, 2020). Regardless of age at menopause, the experience of menopause in the general population is accompanied by normal endocrine changes with increased likelihood for other co-occurring chronic conditions requiring healthcare management (Monteleone et al., 2018; Shifren et al., 2014), such as genitourinary changes (Mitchell & Waetjen, 2018), sleep disorders (Tandon et al., 2022), or mental health conditions (Hu et al., 2016).

Few studies have been undertaken to understand menopause in autistic females despite autistic people emphasizing menopause specifically as a priority topic (Moseley et al., 2020a) and the experiences of autistic women and aging generally identified as important gaps via participatory research approaches (Benevides et al., 2020; Moseley et al., 2020a, 2020b). In qualitative research, menopause is reported to be experienced differently by autistic individuals, with new challenges in everyday functioning compared to perceived functioning prior to menopause (Moseley et al., 2020a). Specifically, autistic women report that sensory experiences, executive functioning, sleep, and mood changes impact the experience of menopause symptoms (2020b; Moseley et al., 2020a). Emerging research suggests that mental health challenges, including increased depression and suicide risk, may be higher for autistic women compared to non-autistic women during menopause (Groenman et al., 2022; Moseley et al., 2020b) although cognitive and emotional changes are reported frequently in the general population (Monteleone et al., 2018).

While greater attention to the priorities of autistic individuals across the lifespan has resulted in qualitative findings into menopause experiences of aging autistic women, no studies have investigated this topic quantitatively in a large sample. Understanding the identification and treatment of symptomatic menopause in autistic people is essential for better clinical care and for addressing the knowledge gaps that healthcare providers have expressed related to addressing needs of aging autistic people. The purpose of this study was to investigate rates of symptomatic menopause among autistic females in a publicly-insured U.S. claims sample. We asked the following primary research questions: (1) What is the unadjusted rate of symptomatic menopause among autistic females aged 35–39 (“premature menopause”), 40–45 (“early menopause”), and 46–70 years, and what demographic or enrollment factors differ between those with symptomatic menopause compared to an unmatched control group? And (2) what is the adjusted prevalence of co-occurring mental and physical health conditions among those with symptomatic menopause in the claim record?

Methods

Positionality Statement

The study team includes autistic people, academic and community researchers, and a physician of obstetrics and gynecology who collaboratively contributed to the study.

Ethics

The use of this data for research was approved by the institutional review board of Drexel University, which granted a waiver of informed consent.

Design and Data Source

We used beneficiary data from 2014 to 2016 Medicare and Medicaid claims from all 50 U.S. states, the District of Columbia, and Puerto Rico to conduct this observational retrospective cohort study. In our CMS data purchase, we did not request a non-disabled control sample, therefore no claims were available for a general population sample and we rely on existing literature on general population menopause data for comparison. Demographic and eligibility information were available in personal summary files, and service files included information related to healthcare visits and diagnostic codes. These data sources were selected because in the United States, Medicare and Medicaid insure a large proportion of autistic individuals. Information about Medicaid and Medicare is contained in the Appendix.

Inclusion and Exclusion Criteria

We included autistic adults between the ages of 35–70 years old and of female sex. “Autism spectrum disorder” diagnoses were identified through use of an algorithm through the Chronic Conditions Warehouse (CCW; Centers for Medicare & Medicaid Services, 2022). More information about the CCW is in the Appendix. All enrollees were required to be Medicaid- or Medicare-enrolled for at least 10 months of a claim year. This is because disenrollment for administrative reasons (e.g., due to error) may result in enrollees losing coverage for 1–2 months (Carey et al., 2023; Shea et al., 2022).

Outcome Variable

Menopause is a normal part of aging, although it is quantified in medical claims with ICD-9 or ICD-10 claims when symptoms require medical treatment. We quantified individuals as seeking treatment or care for symptomatic menopause if they had at least one diagnosis code in any inpatient claim or two outpatient claims for menopause, similar to studies of symptomatic menopause in the general population (Moser et al., 2020). Menopause status was quantified as premature menopause (symptomatic menopause claim at age < 40 years), early menopause (age 40–45 years of first symptomatic menopause claim) or expected menopause (age 46–70 years of first symptomatic menopause claim) (U.S. Department of Health and Human Services Office of Women’s Health, n.d.). We included an unmatched reference group of autistic females between 35 and 70 years with no menopause claims on record. We excluded individuals with gender dysphoria because treatment for gender dysphoria influences symptoms or co-occurring conditions that resemble those of menopause. All methods requiring use of diagnosis codes for ICD-9 or ICD-10, including for symptomatic menopause identification or gender dysphoria exclusion, are contained in the Online Appendix.

Co-occurring Conditions

We evaluated the prevalence of 10 conditions that the literature identifies as occurring more frequently in the general population during menopause and which autistic individuals have also identified as problematic: anxiety disorders, heart disease, diabetes, headache or migraine, osteoporosis, altered sexual function, and sleep disturbance (e.g., Monteleone et al., 2018; Shifren et al., 2014), or that autistic research partners identified from lived experience as occurring more frequently during menopause: symptoms of attention-deficit hyperactivity disorder (ADHD), depressive disorders, and altered sensory experiences. For most conditions, we relied on CCW algorithms for identification. For conditions without CCW criteria, we relied on our obstetrics/gynecology physician researcher author for coding recommendations (see Online Appendix for codes).

Covariates

Covariates included demographic (age group 35–39, 40–45, 46–70; race and ethnicity defined as White, Black, Asian/Pacific Islander, Hispanic, Native American/Alaska Native, other, missing; urban/rural status defined as living in a rural county or not; poverty status defined as eligibility for benefits based on poverty; presence of co-occurring intellectual disability, ID) and enrollment characteristics (Medicare-only, Medicaid-only, or both). Intellectual disability was identified using CCW criteria, and due to lack of adaptive behavior variables, we only describe ID broadly (and not with levels of ID).

Data Analysis

For research question 1, we calculated unadjusted rates of symptomatic menopause in each age group and evaluated demographic and enrollment characteristics associated with symptomatic menopause using logistic regression. For research question 2, we conducted ten separate adjusted logistic regressions predicting likelihood of diagnosis for each co-occurring condition in the symptomatic menopause group compared to the reference group adjusting for covariates mentioned above within the three-year study window. We present adjusted odds ratios (aOR) and 95% confidence intervals for each co-occurring condition in the symptomatic menopause group. Full models are available upon request. Statistical significance was set at p < 0.05.

Results

Our total sample included 26,904 female autistic adults aged 35–70 years in 2014–2016 claims for Medicaid, Medicare, or both. Table 1 displays number and unadjusted rates of demographic, enrollment characteristics, and co-occurring conditions in the full sample.

Table 1.

Demographic and enrollment characteristics and co-occurring conditions among female autistic adults aged 35–70 with and without symptomatic menopause

	Symptomatic menopause		Referencea		Full Sample		General menopausal population literature
	N = 797		N = 26,107		N = 26,904
	N	Col%	N	Col%	N	Col%
Age group (years)
35–39	41	5.14	6,020	23.06	6,061	22.53
40–45	96	12.05	5,196	19.90	5,292	19.67
46–70	660	82.81	14,891	57.04	15,551	57.80
Race/ethnicity
White	638	80.05	18,929	72.51	19,567	72.73
Black	91	11.42	4,122	15.79	4,213	15.66
Asian/Pacific Islander	*	*	*	*	530	1.97
Hispanic	33	4.14	1,471	5.63	1,504	5.59
Native Indian/Alaska Native	*	*	*	*	159	0.59
Other	*	*	*	*	122	0.45
Missing	13	1.63	796	3.05	809	3.01
Urbanicity
Urban	594	74.53	18,286	70.04	18,880	70.18
Suburban	135	16.94	5,517	21.13	5,652	21.01
Rural	24	3.01	920	3.52	944	3.51
Missing	44	5.52	1,384	5.30	1,428	5.31
Enrollment
Dual eligible	455	57.09	10,212	39.12	11,707	43.51
Medicaid only	155	19.45	11,552	44.25	4,530	16.84
Medicare only	187	23.46	4,343	16.64	10,667	39.65
Poverty	102	11.80	2,909	11.14	3,011	11.19%
Intellectual disability	526	66.00	17,965	68.81	18,491	68.73
Co-occurring conditions
Anxiety	444	55.71	10,957	41.97	11,401	42.38	3.6% (Hu et al., 2016); 6.5% (Moser et al., 2020)
Depression	384	48.18	9,304	35.64	9,688	36.01	4.6% (Hu et al., 2016); 0.6% (Moser et al., 2020); “Most women do not become clinically depressed” (Shifren et al., 2014)
ADHD	216	27.10	5,233	20.04	5,449	20.25
Diabetes	218	27.35	5,768	22.09	5,986	22.25	3.7% (Moser et al., 2020)
Osteoporosis	116	14.55	2,212	8.47	2,328	8.65	2.4% (Moser et al., 2020)
Heart disease	404	50.69	10,735	41.12	11,139	41.40	3.6% (Moser et al., 2020)
Acute myocardial infarction	*	*	*	*	99	0.37
Heart failure and non-ischemic heart disease	25	3.14	814	3.12	839	3.12
Hypertension	340	42.66	9,195	35.22	9,535	35.44	15.4% (Moser et al., 2020)
Ischemic heart disease	77	9.66	1,865	7.14	1,942	7.22
Stroke	53	6.65	1,139	4.36	1,192	4.43
Peripheral vascular disease	82	10.29	1,934	7.41	2,016	7.49
Headache/migraine	190	23.84	3,542	13.57	3,732	13.87	10–20% (Monteleone et al., 2018)
Sensory experiences	173	21.71	3,245	12.43	3,418	12.70
Altered sexual function	33	4.14	306	1.17	339	1.26	42% (Monteleone et al., 2018); “highly prevalent” (Shifren et al., 2014)
Sleep disturbance	698	87.58	18,762	71.87	19,460	72.33	2.8% sleep disturbance (Hu et al., 2016); 2.0% insomnia (Moser et al., 2020); 40–60% sleep disruption (Monteleone et al., 2018); 35–60% (Tandon et al., 2022)

Author’s analysis of 2014–2016 Medicare and Medicaid claims

^aReference group were autistic females with no symptomatic menopause claim on record

^*Suppressed values per CMS data use agreement

Research Question 1

Among those aged 35–39 years (n = 6061), 0.68% were coded as experiencing premature menopause symptoms; of the 40–45 age group (n = 5292), 1.81% were coded as experiencing early menopause symptoms; and of the 46–70 age group (n = 15,551), 4.24% were coded as experiencing menopause symptoms. This was confirmed in adjusted analyses: older individuals aged 46–70 years old were more likely to have symptomatic menopause claims on record compared to those aged 35–39 years (aOR = 5.33, 95% confidence interval [CI]: 3.85–7.36) as were those aged 40–45 years (aOR = 2.46, 95% CI 1.70–3.56). Individuals with co-occurring ID were less likely to have a symptomatic menopause claim (aOR = 0.72, 95% CI 0.60–0.87). Medicare-only individuals had higher odds of having a symptomatic menopause claim on record (aOR 1.89, 95% CI 1.49–2.38) as did dual-enrollees in both Medicare and Medicaid (aOR = 2.39, 95% CI 1.95–2.93). All other demographic co-variates (race/ethnicity, enrollment on the basis of poverty, and rural status) were non-significant in the regression model.

Research Question 2

Autistic females experiencing symptomatic menopause at any age were significantly more likely than the reference group to have co-occurring ADHD, anxiety disorders, depressive disorders, headache or migraine, altered sensory experiences, altered sexual function, and sleep disturbance, after adjusting for age group, race/ethnicity, ID, poverty status, urbanicity, and enrollment status (Fig. 1).

Fig. 1.

Conditions occurring more frequently among symptomatic menopausal autistic females compared to autistic reference females aged 35–70 years old. Adjusted odds ratios and 95% confidence intervals generated from the author’s analysis of 2014–2016 Medicare and Medicaid claims data. All models were adjusted for age, race and ethnicity, urban/rural status, poverty status, intellectual disability, and Medicare and/or Medicaid enrollment. Green bars not overlapping with 1 indicate conditions that are statistically different in the symptomatic menopause group compared to the reference group

Discussion

This is the first quantitative study to identify rates of symptomatic menopause in a large sample (~ 27,000) of autistic people in the menopausal age range. We found that approximately 4% of autistic females aged 46–70 years had symptomatic menopause claims. This is half of the 8% cumulative incidence of symptomatic menopause in women from the general population reported in a similarly designed study using medical claims (Moser et al., 2020). We would expect to see higher rates of symptomatic menopause because autistic people report significant challenges with menopausal symptoms (2020b; Groenman et al., 2022; Karavidas & de Visser, 2021; Moseley et al., 2020a). Our findings point to the need to identify the specific barriers that autistic females face when trying to obtain care and support for menopausal symptoms. Possible reasons for under-reporting of symptomatic menopause experiences include lack of access to timely or needed care, lack of communication accommodations or resources to promote dialogue, or lack of clinician skill or knowledge in addressing this important topic. Further, diagnosis of autistic females is less likely to be known or acknowledged by healthcare providers, and autistic female concerns and priorities of care are often dismissed (e.g., Miller et al., 2022; Raymaker et al., 2017; Zerbo et al., 2015). Lastly, autistic people report that menopause exacerbates existing challenges often attributed to their autism (e.g., executive functioning, sensory), suggesting that provider’s diagnostic over-shadowing is occurring (e.g., 2020b; Moseley et al., 2020a).

In our study, approximately 68% of the autistic female sample had ID. This is not unusual for this age cohort, born between 1944 and 1979. The earliest prevalence report on autism in the U.S. included children born between 1986 and 1993 and reported the rate of co-occurring ID to be 68% (Yeargin-Allsopp et al., 2003), consistent with our sample. In our study, we found that females on the autism spectrum with co-occurring ID were 28% less likely to have a symptomatic menopause claim on record compared to those without ID. There is no clear physiological reason why autistic people with intellectual disability should have differences in symptomatic menopause, suggesting that development of innovative ways to support communication for individuals for whom self-report of symptoms may be limited is an important next step. The interaction between communication, alexithymia, and interoception are factors that need to be explored, particularly for autistic individuals with different communication needs. Autistic females with ID who may be unable to communicate location and experience of symptoms could be underdiagnosed or misdiagnosed. Better tools to support communication and evaluate pain are areas for future research (El-Tallawy et al., 2023).

Conditions that differed between symptomatic and non-symptomatic autistic females included ADHD, sexual function, anxiety, depression, headache/migraine, altered sensory experiences, and sleep disturbance. For these conditions specifically, when compared to the general population of menopausal females, rates of anxiety, depression, and sleep disturbance are much higher among autistic individuals, but are even greater among menopausal autistic females. This triad of co-occurring conditions is particularly important, as both Hu et al. (2016) and Tandon et al. (2022) note that in the general public these three co-occurring conditions (anxiety, depression, and sleep disturbance) are inter-related among menopausal females. A lifetime of sleep disturbances, exacerbated by menopause, has the potential to intensify or contribute to other physical and mental health conditions that affect daily living, employment, and quality of life. More work is needed to understand how sleep changes among autistic people experiencing menopause, and the way in which altered sleep affects wellbeing and mental health.

Rates of reported altered sexual function among autistic individuals are much lower than the non-autistic general menopausal population, however, were reported about 3 times more frequently among symptomatic menopausal autistic females than non-symptomatic autistic females. This outcome has received very little attention in the literature despite intimacy, sex, and interpersonal relationships being identified by autistic people among the top 10 priority outcomes that matter (Benevides et al., 2020). Sexual health and gender-related priorities are expressed as key areas for future research, and this should not be limited to transition-age youth (Dewinter et al., 2023).

Sensory experiences are often examined in autistic children but limited work to understand how to support autistic adults in managing changes in how sensory experiences change in adulthood are warranted. No available research in the general population describes altered sensory experiences among menopausal females except for vasomotor changes (e.g., hot flashes), which occurred in approximately 75% of the general population (Monteleone et al., 2018). The experience of vasomotor symptoms may differ among autistic females and greater attention to these symptoms is warranted.

Interestingly, dually-enrolled individuals in our study were more likely to have a symptomatic menopause claim. Our finding implies that dual-enrollment in Medicare and Medicaid may afford system resources that enhance discussion of these symptoms in a care setting. It is possible that dual-enrollees have access to providers with more expertise with aging individuals. Other reasons that dual-enrollment may afford resources is that the regulatory mechanisms for dual enrollees include aspects of care coordination that could affect people’s ability to obtain care more easily. Little is known about dual enrollees, however, more work to understand how dual enrollees are accessing care is needed (Benevides et al., 2021).

Limitations of our study include usual challenges with claims data including disentangling race and ethnicity. Further, this study did not contain an allistic/non-autistic control cohort and was not designed in a case–control fashion, which limits our conclusions about the autistic experience in menopause care compared to the general population. Future studies should consider examining incidence of new co-occurring conditions; medication usage for mental health, sleep, or other symptoms of menopause; and prospective studies that are able to examine patient and provider characteristics likely impacting care during the menopausal transition. Strengths of our study include comparison of those with and without menopausal symptoms in autistic females observed in a 3-year claims period. This design has allowed us to report for the first time, occurrence of symptomatic menopausal in the autistic population as defined by medical claims in a large sample.

In summary, autistic females experience symptomatic menopause at a rate lower than what we would expect, and in our Medicare and/or Medicaid-enrolled sample, those with co-occurring ID are identified as having symptomatic menopause at a lower rate than those without ID. Our study represents autistic females receiving Medicare and/or Medicaid and does not represent all autistic females. Future studies should explore age and cohort effects that may confound our understanding of autism, intellectual disability, and menopause. Individuals with under-reported menopausal symptoms may not be getting appropriate management and treatment for both symptoms of menopause or other co-occurring conditions. Lack of menopause knowledge, clinical supports, and available resources for autistic people hinders important preventive care related to cardiovascular and other risks tied to menopausal transition (Collaborative Group on Hormonal Factors in Breast Cancer, 2019; El-Khoudary et al., 2020) and the cognitive, emotional, and sensory features that affect overall quality of life of menopausal autistic adults (2020b; Groenman et al., 2022; Moseley et al., 2020a). We have identified important next steps that require further inquiry to advance the study of menopause and the autistic lived experience required to address both the interpersonal and systemic barriers autistic adults face when seeking support.

Supplementary Material

Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10803-024-06516-x.

Data Availability

The data used in this study are available for purchase but not for public release because they are governed by the requirements of a Data Use Agreement with the Centers for Medicare and Medicaid.