Pregnancy characteristics and outcomes among birthing individuals with a diagnosis of fetal alcohol syndrome

Abstract

Background:

Fetal alcohol syndrome (FAS) can have adverse effects on health outcomes throughout the life course. Adults with FAS have an increased risk of chronic and infectious diseases. Although these conditions can affect reproductive health, few have described perinatal outcomes among individuals with an FAS diagnosis.

Methods:

We analyzed data from the Study of Mothers and Infants, an administrative birth cohort derived from California birth certificates linked with a hospital discharge database. The cohort consisted of 7.3 million singleton, live births between 2005-2021. FAS was identified by International Classification of Disease (ICD) codes in maternal hospital discharge records. Pregnancy and birth outcomes were captured via ICD codes in maternal or infant records. We performed descriptive analyses for pregnancy and birth outcomes by maternal FAS diagnosis.

Results:

There were 35 babies born to 30 individuals with an FAS diagnosis between 2005-2021 (0.5 per 100,000 live births). The prevalence of births to individuals with an FAS diagnosis increased over the period. Compared to birthing individuals without an FAS diagnosis, those with FAS were more likely to identify as non-Hispanic White, or ‘other/multiple’ race, and less likely to be Hispanic. They were more likely to be publicly insured and less than 18 years old. Birthing individuals with FAS were more likely to have diagnoses of mental health disorders, epilepsy, substance use disorders, use nicotine during pregnancy, have preexisting or gestational hypertension, and have sexually transmitted infections or other infections complicating pregnancy. Infants born to individuals with FAS were more likely to be born premature or small for gestational age and be admitted to the neonatal intensive care unit.

Conclusions:

Our work highlights the need for improved recognition of FAS among birthing people. Results suggest that individuals with FAS would benefit from early and sustained medical care prior to pregnancy to optimize perinatal outcomes.

Introduction

Alcohol is the most commonly used drug worldwide, consumed by over 2.3 billion people as of 2018, as reported by the World Health Organization (Hammer et al., 2018). Alcohol exposure during pregnancy can lead to Fetal Alcohol Spectrum Disorder (FASD), a spectrum of disorders characterized by behavioral and intellectual dysfunction, altered facial features, and restriction of growth (Jones, 2011). FASD includes four main diagnoses, each depending on the extent of growth, facial dysmorphia, central nervous system function, and prenatal alcohol exposure (Hoyme et al., 2016). The four categories include alcohol-related neurodevelopmental disorder, partial fetal alcohol syndrome, alcohol-related birth defects, and, most severe, fetal alcohol syndrome (FAS) (Hoyme et al., 2016). The effects of prenatal alcohol exposure can vary based on the dose of alcohol, length of in-utero exposure, nutrition, maternal health, and other factors, and FASD is often unrecognized or misdiagnosed (Chasnoff et al., 2015; May et al., 2014).

Previous research has primarily focused on health outcomes among young children with FASD. However, a small but growing collection of scientific literature has documented the disparities faced by adults with FASD (Moore and Riley, 2015). Notably, many of these findings arose from studies conducted in animal models (Pennington et al., 2002; Polanco et al., 2010; Zhang et al., 2012) as human adult FASD studies are uncommon. The most common disparities include an increased risk for immune-mediated diseases, such as diabetes or cancers (Himmelreich et al., 2020; Pennington et al., 2002; Polanco et al., 2010; Zhang et al., 2012), use of tobacco and alcohol use disorders (Landgren et al., 2019; Streissguth et al., 2004), and mental health diagnoses (Famy et al., 1998; Landgren et al., 2019). Also, psychosocial adversities disproportionately affect adults with FASD. In a qualitative study of 13 women with FASD who had substance use disorders or a child with FASD, most had involvement of child protective services, were survivors of violence or trauma, had mental health disorders, experienced poverty and job insecurity, and had difficulty accessing medical care (Rutman, 2013).

Very few studies have characterized the reproductive outcomes of individuals with FASD (Akison et al., 2019). In a systematic review of adverse reproductive outcomes associated with prenatal alcohol exposure, authors found five studies in humans, from which three included females (Akison et al., 2019). These studies primarily focused on age at menarche as the outcome. In addition, a large survey of adults with FASD catalogued the occurrence of a comprehensive set of health outcomes, including reproductive health outcomes (Himmelreich et al., 2020). These outcomes included the prevalence of recurrent miscarriages, giving birth to a premature baby, breastfeeding problems, and having a child with special needs (Himmelreich et al., 2020).

Adverse psychosocial and health outcomes among adults with FASD have the potential to negatively impact their pregnancies, yet to date, this remains poorly documented. Thus, the objective of this paper was to characterize the prevalence of birthing individuals with an FAS diagnosis from a large cohort of births in California, and describe their maternal, pregnancy, and birth outcomes. Throughout this paper, we use the term maternal to refer to the person giving birth with acknowledgement that all individuals do not identify with this title.

Methods

Study sample

We used data from 2005-2021 from the Study of Mothers and Infants (SOMI), an administrative birth cohort derived from a complete collection of California birth records. These records, maintained by California Vital Statistics, were linked to the California Department of Health Care Access and Information (HCAI) hospital, emergency department, and ambulatory surgery records for the birthing person (in the year before the birth through one-year post-delivery) and the infant (first year of life). The HCAI data includes diagnosis and procedure codes recorded using the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9), and the International Classification of Diseases, 10th Revision, Clinical Modification (ICD-10). Between 2005-2021, there were birth records for 7,396,840 infants delivered between 20-44 weeks gestation linked to HCAI records. There were no stillbirths to pregnant people with an FAS diagnosis, thus the dataset was limited to livebirths (n=7,366,293).

The SOMI study was approved by the Committee for the Protection of Human Subjects within the Health and Human Services Agency of the State of California and the University of California San Diego Human Research Protections Program.

Exposure and outcomes

We identified birthing people with an FAS diagnosis through ICD codes (ICD 9: 760.71; ICD-10: Q86.0) on maternal HCAI discharge records. Maternal demographic characteristics included maternal race/ethnicity (self-reported: non-Hispanic White, Hispanic, non-Hispanic Black, Asian/Pacific Islander, and multiple/other), age (<18, 18-34, >34), payer (private insurance, public insurance, or other sources), and education (<12 years, 12+ years), all assessed from birth records. Medical conditions included anxiety disorder (ICD9: 300; ICD10: F4), major depressive disorder (ICD9: 296.2, 296.3, 311; ICD10: F32.2, F32.3, F33.2, F33.3), bipolar disorder (ICD9: 296.0, 296.1, 296.4, 296.5, 296.6, 296.7, 296.8, 296.9; ICD10: F30-31), schizophrenia (ICD9: 295.7; ICD10: F20), cannabis use disorder (ICD9 305.2, 304.3; ICD10: F12), alcohol use disorder (ICD9: 303, 305.0; ICD10: F10), epilepsy and other seizure disorders (ICD9: 649.4, 345, ICD10: G40-47), sexually transmitted infections (ICD9: 647.0-647.2, ICD10: O98.1, O98.2, O98.3), and other infections complicating pregnancy (ICD9: 646.5, 646.6, 647.3 – 647.9, ICD10: O98.0, O98.4, O98.5, O98.6, O98.7, O98.8, O98.9, O23.0, O23.1, O23.2, O23.3, O23.4), which were assessed from HCAI records. Preexisting diabetes (ICD9: 648.0, 249, 250; ICD10: E10, E11, E12, E13, E14, O24.0, O24.1, O24.2, O24.3), gestational diabetes (ICD9: 648.8; ICD10: O24.4), preexisting hypertension (ICD9: 642.0-642.2, 642.7; ICD10: O10, O11), gestational hypertension (ICD9: 642.3, 642.4, 642.5, 642.6; ICD10: O13, O14.0, O14.1, O14.2, O14.9,O15) and nicotine use (ICD9 305.1, 649.0; ICD10: Z72.0, F17.2) were coded if they appeared in hospital discharge summaries or birth records.

Infant outcomes included preterm birth (<37 weeks gestational age at delivery), small for gestational age offspring (<10th centile for weight, calculated from birthweight, gestational age, and sex) (Talge, 2014), and neonatal intensive care unit (NICU) admission, which were assessed from birth records.

Statistical analysis

We calculated the prevalence of singleton, live births per year to individuals with a documented FAS diagnosis and performed a descriptive analysis comparing maternal, pregnancy, and birth outcomes between birthing individuals with and without a FAS diagnosis. We first calculated chi-square estimates, using Fisher’s exact when cell sizes were <5, for baseline demographic characteristics. Per our agreement with the California Department of Public Health, cells with frequencies <5 were masked, along with another row to preserve the masking. Next, unadjusted risk ratios with 95% confidence intervals were estimated for maternal, pregnancy, and birth outcomes. Again, cells with frequencies <5 were required to be masked. When this occurred, prevalence was listed as ‘<14%’ (5/35) and RRs were masked to preserve back calculation. Further, RRs were displayed as >1 or <1, and statistically significant (SS) or not statistically significant (NS) was noted based on a p value of <0.05 (with corresponding confidence intervals that did not contain the null).

Results

From 2005-2021 there were 7,366,293 singleton live born infants in California with birth records linked to HCAI records. From those, 35 infants were born to 30 unique individuals with an FAS diagnosis (0.5 per 100,000 births). The number of infants born to individuals with an FAS diagnosis increased over the study period from 0.2 per 100,000 deliveries in 2005 to a high of 2.1 per 100,000 deliveries in 2020 (Figure 1).

Figure 1.

Prevalence of maternal FAS diagnosis per 100,000 singleton, live births in California (2005-2021).

Birthing individuals with an FAS diagnosis were more likely to identify as non-Hispanic White or ‘multiple/other’ race compared to those without an FAS diagnosis and were less likely to identify as Hispanic (Table 1). They were also more likely to be younger than 18 years of age. Most birthing individuals (91.4%) with an FAS diagnosis reported using public insurance (MediCal) compared to an almost even split between private and public insurance those without FAS. No significant difference was observed for years of education between the two groups.

Table 1.

Maternal sociodemographic characteristics among individuals giving birth in California (2005-2021) stratified by maternal FAS diagnosis

	Infants born to individuals with FAS		Infants born to individuals without FAS
	n=35		n=7,366,258
	n1	%	n	%	p-value
Sociodemographic factors
Maternal Race					<0.0001
White	17	48.6	1988136	27.0
Hispanic	7	20.0	3663111	49.7
Black	<5	<14%	363321	4.9
Asian	<5	<14%	989135	13.4
Multiple/other	7	20.0	326555	4.4
Maternal Age					0.002
<18 years	<5	<14%	154153	2.1
18-34 years	28	80.0	5739091	77.9
> 34 years	<5	<14%	1472805	20.0
missing	0	0.0	209	0.0
Payer					<0.0001
Private	<5	<14%	3536692	48.0
Medicaid	32	91.4	3395611	46.1
Other	<5	<14%	423372	5.7
missing	0	0.0	10583	0.1
Education					0.07
Less than 12 years	>9	>26%	1493066	20.3
More than 12 years	21	60.0	5564308	75.5
missing	<5	<14%	308884	4.2

¹Cells <5 are suppressed

Maternal physical and mental health

Birthing individuals with an FAS diagnosis had a higher prevalence of anxiety disorders (25.7% vs. 2.4%), major depressive disorder (28.6% vs. 2.0%), bipolar disorder (14.3% vs. 0.5%) and schizophrenia (14.3% vs. 0.1%) relative to individuals without an FAS diagnosis (Table 2). In addition, birthing individuals with FAS were more likely to have epilepsy or another seizure disorder (17.1% vs. 1.0%). Substance use disorders of cannabis, alcohol, or other substance use disorders were also much more common among birthing individuals with an FAS diagnosis. Nicotine and tobacco use were reported by 14.3% of birthing individuals with an FAS diagnosis compared to 2.1% of those without. Also, preexisting or gestational hypertension was also elevated among those with FAS compared to those without (22.9% vs. 8.8%). Birthing individuals with an FAS diagnosis were more likely to have a sexually transmitted infection in pregnancy, as well as other infections complicating the pregnancy (37.1% vs. 6.7%). Preexisting or gestational diabetes did not differ between the groups.

Table 2.

Maternal physical and mental health comorbidities and birth outcomes among individuals giving birth in California (2005-2021) stratified by maternal FAS diagnosis

	Infants born to individuals with FAS		Infants born to individuals without FAS
	n=35		n=7,366,258
	n1	%	n	%	RR, 95% CI2
Physical and mental health comorbidities
Anxiety	9	25.7	178791	2.4	10.5 (5.5, 20.3)
Depression	10	28.6	147843	2.0	14.2 (7.6, 26.4)
Bipolar disorder	5	14.3	33594	0.5	31.3 (13.0, 75.2)
Schizophrenia	5	14.3	4649	0.1	226.4 (94.2, 544.1)
Epilepsy or seizure disorders	6	17.1	75023	1.0	16.8 (7.6, 37.5)
Cannabis dependence	5	14.3	73215	1.0	14.4 (5.9, 34.5)
Alcohol dependence	<5	<14%	19481	0.3	>1.0 (SS)
Nicotine/tobacco	5	14.3	155580	2.1	6.7 (2.8, 16.2)
Other substance use	9	25.7	219319	3.0	8.6 (4.4, 16.5)
Preexisting or gestational diabetes	<5	<14%	643243	8.7	>1.0; (NS)
Preexisting or gestational hypertension	8	22.9	647733	8.8	2.6 (1.3, 5.2)
Sexually transmitted infection	<5	<14%	45167.0	0.6	>1.0 (SS)
Other infection complicating pregnancy	13	37.1	496701	6.7	5.5 (3.2, 9.4)
Infant outcomes
Preterm birth	6	17.1	529002	7.2	2.4 (1.1, 5.3)
Small for gestational age	9	25.7	657176	8.9	2.8 (1.5, 5.5)
Infant admitted to NICU	10	28.6	397827	5.4	5.2 (2.8, 9.8)

¹Cells <5 are suppressed;

²When cells <5, RRs are denoted as greater than or less than 1, with statistical significance (p<0.05) expressed as (SS) and p ≥ 0.05 as (NS) to preserve masking of small cells

Infant outcomes

With respect to birth outcomes, babies born to individuals with FAS were more likely to be born preterm (17.1% vs. 7.2%) or small for gestational age (25.7% vs. 8.9%) and to be admitted to the NICU (28.6% vs. 5.4%).

Discussion

To our knowledge, this is the first large, population-based study of pregnancy and birth outcomes for individuals with an FAS diagnosis. Individuals giving birth with FAS were more likely to identify as non-Hispanic White or other races, rely on public insurance, and have a higher prevalence of mental health and physical comorbidities. Higher prevalence of substance use disorders was also observed. Finally, babies born to individuals with an FAS diagnosis were more likely to be delivered prior to 37 weeks of gestation, be born small for gestational age or be admitted to the NICU.

The US prevalence of FASD, estimated from a sample of first-grade children, ranges from 3.1-9.8% (May et al., 2018). FAS accounted for less than 20% of the overall prevalence of FASD cases, ranging from 0-0.8%. In this sample of birthing individuals, the prevalence of FAS was 0.0005%, or 0.5 per 100,000 deliveries, highlighting the under-ascertainment in this administrative data source. However, although highly variable, there was a monotonic increase in the prevalence of infants born to women with an FAS diagnosis during the study period.

In any given year in California over the study period, 0-8 individuals giving birth (average 2.2) had an FAS diagnosis noted in their hospital records, raising the possibility that most clinicians have not knowingly provided obstetric care to a person with FAS. The results of this descriptive study highlight the medical complexities of birthing individuals with FAS, which are important for clinicians to be aware of. Given the lack of published literature on pregnancy and birth outcomes to individuals with FAS, we compare our findings to other relevant literature. Mental health disorders were extremely common (present in 10-30% of pregnancies), which replicated findings from a cohort of 37 adults (15 females) with FASD in Sweden, where the prevalence of similar mental health disorders ranged from 10-50% of the cohort (Landgren et al., 2019). The results were also similar to a Himmelreich et al. survey of adults with FASD (53% female) where 17-88% reported similar mental health disorders (Himmelreich et al., 2020). Our findings of increased prevalence for substance use disorders in the population of birthing individuals with FAS (14-25%) was also directionally similar to findings from the other two adult studies (Himmelreich et al., 2020; Landgren et al., 2019). Furthermore, in our cohort, 17% of birthing individuals with FAS had a diagnosis for epilepsy or a seizure disorder, compared to 1% of the sample without FAS. This striking result mirrors findings by Himmelreich and colleagues in the survey of adults with FASD, which reported 20% of respondents endorsing seizure disorders (Himmelreich et al., 2020). Approximately 20% of birthing individuals with FAS had preexisting or gestational hypertension, which was also similar to the Swedish cohort where 28% had hypertension (Landgren et al., 2019) and the adult Himmelreich et al. survey where 16% had hypertension (Himmelreich et al., 2020). Finally, birthing individuals with FAS were much more likely to have a sexually transmitted infection or other infection during pregnancy. Although not the same types of infections, the Himmelreich et al. survey of adults with FASD reported infections across a wide spectrum (ear infections, chest infections, fungal infections, bladder infections) to be greatly elevated compared to the general population, and discussed the possibility that prenatal alcohol exposure had altered the immune response later in life (Himmelreich et al., 2020).

Our findings of an increased prevalence of adverse birth outcomes are unsurprising given the comorbidities present in the population, and highlight the need for comprehensive medical care before and during pregnancy. Infants born to individuals with FAS were 2-3 times more likely to be delivered prematurely or be small for gestational age, and almost five times more likely (28% vs. 5%) to be admitted to the NICU. In the Himmelreich et al. survey, the prevalence of preterm births (9.5%) did not differ from the general population (Himmelreich et al., 2020). They did, however, report women with FASD were more likely to have recurrent miscarriages (11% vs. 2% of the general population). While we were unable to examine early pregnancy losses, we did examine fetal deaths (death after 20 weeks of gestation), but there were none in the cohort among birthing individuals with an FAS diagnosis. Finally, although not directly comparable, there is a larger literature on pregnancy and birth outcomes among women with any intellectual or developmental disability (IDD), which includes FASD. Similar to our results, in a meta-analysis, birthing individuals with an IDD had an increased risk of hypertensive disorders of pregnancy (RR 1.8, 95% CI 1.2, 2.6), but not for gestational diabetes (Tarasoff et al., 2020). Also, in a Canadian administrative cohort, birthing individuals with an IDD had increased risks for preeclampsia (aRR 1.5, 95% CI 1.1–1.9), and their offspring had increased risks for preterm birth (aRR 1.6, 95% CI 1.5–1.8) and being born small for gestational age (aRR 1.4, 95% CI 1.3–1.5) (Brown et al., 2017). Although absolute risks were lower for offspring outcomes in the Canadian population (potentially due to the heterogeneity in grouping all IDDs), the elevated risks relative to the general population mirror our findings.

Individuals with FASD are subject to stigmatization from a young age, often due to the medical and social problems that accompany FASD. Further, they are born into families that face a tremendous amount of stigma due to society’s simplified beliefs of maternal culpability in alcohol use during pregnancy (Roozen et al., 2022). This stigma, experienced over the life course, may result in individuals with FASD receiving less medical care, and by extension, less reproductive health care. Additionally, there is evidence that birth mothers of individuals with FASD have a higher prevalence of substance use and mental health disorders (Singal et al., 2017), lower socioeconomic status, inadequate prenatal care, and psychological distress (Singal et al., 2019), highlighting the adversities individuals with FASD encounter in the earliest years of life. These adversities are exacerbated in their own pregnancies by the societal stereotype that individuals with FASD will have children with FASD. These stereotypes increasingly become reality when individuals with FASD are not expected to become parents, or provided access to safe, effective services to promote reproductive health. Many programs meant to improve health and social outcomes fail individuals with FASD, as they are based on behavioral learning models that are challenging to affected individuals (Rutman, 2016). This has led to the call for FASD-informed programs to meet the needs and challenges of individuals with FASD. Briefly, this includes having awareness of FASD, making person-centered accommodations, and being strengths-based (Rutman, 2016). Reproductive rights are universal, and to achieve equality in healthcare, research on FAS in reproductive aged individuals must continue. Understanding the disparities is the first step toward increasing awareness and producing actionable efforts such as improving FASD awareness, promoting studies of adults with FASD, and prioritizing the holistic care of individuals with FAS prior to and during their pregnancies.

Limitations

The following limitations should be considered when interpreting the results of this study. FAS is an underdiagnosed condition, and FAS may be captured as another intellectual disability or, more likely, not at all. This would lead to misclassification in the non-FAS group, potentially attenuating the differences between the two groups. Further, administrative data often provide incomplete capture of behaviors and exposures, primarily capturing those that rise to the level of a diagnosis (e.g., with substance use disorders). Thus, many of our prevalence estimates may be underreported, although we would not expect this to differ substantially by FAS diagnosis. Finally, although the total sample is quite larger, FAS diagnosis is rare and, importantly, fails to capture the full spectrum of disorders caused by prenatal alcohol exposure (e.g. alcohol-related neurodevelopmental deficits). The latter do not have corresponding ICD codes, so we were unable to identify individuals with any other FASD. Thus, we were unable to comment on the full spectrum of disorders with respect to pregnancy and birth outcomes.

Conclusion

In this study, we highlight the adverse maternal and birth outcomes among individuals with an FAS diagnosis giving birth in California. Future research should attempt to compile larger sample sizes to examine the reproductive health of individuals with FAS, and should further prioritize reproductive health outside of pregnancy, including the use of contraceptive methods, pregnancy planning, and frequency of early-onset menopause and miscarriages. Individuals with FAS deserve the same reproductive and healthcare rights as everyone else. Interventions, including early and sustained medical care, would optimize maternal and infant outcomes in this vulnerable population.