Key PointsQuestion
What is the association of state-level pregnancy-specific alcohol policies with infant morbidities and maltreatment?
Findings
In this population-level cohort study of 1 432 979 birthing person–infant pairs in the US, most state-level pregnancy-specific alcohol policies were not associated with decreased odds of infant maltreatment or morbidities. The few policies that were associated with decreased odds of maltreatment or morbidities were associated with an increase in the odds of another adverse outcome.
Meaning
These findings suggest that state-level pregnancy-specific alcohol policies do not appear to be effective at reducing harms to infants.
Importance
Research has found associations of pregnancy-specific alcohol policies with increased low birth weight and preterm birth, but associations with other infant outcomes are unknown.
Objective
To examine the associations of pregnancy-specific alcohol policies with infant morbidities and maltreatment.
Design, Setting, and Participants
This retrospective cohort study used outcome data from Merative MarketScan, a national database of private insurance claims. The study cohort included individuals aged 25 to 50 years who gave birth to a singleton between 2006 and 2019 in the US, had been enrolled 1 year before and 1 year after delivery, and could be matched with an infant. Data were analyzed from August 2021 to April 2023.
Exposures
Nine state-level pregnancy-specific alcohol policies obtained from the National Institute on Alcohol Abuse and Alcoholism’s Alcohol Policy Information System.
Main Outcomes and Measures
The primary outcomes were 1 or more infant injuries associated with maltreatment and infant morbidities associated with maternal alcohol consumption within the first year. Logistic regression, adjusting for individual-level and state-level controls, and fixed effects for state, year, state-specific time trends, and SEs clustered by state were used.
Results
A total of 1 432 979 birthing person–infant pairs were included (mean [SD] age of birthing people, 32.2 [4.2] years); 30 157 infants (2.1%) had injuries associated with maltreatment, and 44 461 (3.1%) infants had morbidities associated with alcohol use during pregnancy. The policies of Reporting Requirements for Assessment/Treatment (adjusted odds ratio [aOR], 1.28; 95% CI, 1.08-1.52) and Mandatory Warning Signs (aOR, 1.18; 95% CI, 1.10-1.27) were associated with increased odds of infant injuries but not morbidities. Priority Treatment for Pregnant Women Only was associated with decreased odds of infant injuries (aOR, 0.83; 95% CI, 0.76-0.90) but not infant morbidities. Civil Commitment was associated with increased odds of infant injuries (aOR, 1.26; 95% CI, 1.08-1.48) but decreased odds of infant morbidities (aOR, 0.57; 95% CI, 0.53-0.62). Priority Treatment for Pregnant Women and Women With Children was associated with increased odds of both infant injuries (aOR, 1.12; 95% CI, 1.00-1.25) and infant morbidities (aOR, 1.08; 95% CI, 1.03-1.13). Reporting Requirements for Child Protective Services, Reporting Requirements for Data, Child Abuse/Neglect, and Limits on Criminal Prosecution were not associated with infant injuries or morbidities.
Conclusions and Relevance
In this cohort study, most pregnancy-specific alcohol policies were not associated with decreased odds of infant injuries or morbidities. Policy makers should not assume that pregnancy-specific alcohol policies improve infant health.
The adverse outcomes of pregnant people’s alcohol consumption remain a substantial public health concern in the US and globally.1-3 Despite most US states having multiple policies in effect that attempt to reduce alcohol use during pregnancy and related harms,4 rates of use during pregnancy and related harms have remained steady.5-10
Previous research11 has examined associations of pregnancy-specific alcohol policies with alcohol use during pregnancy and birth outcomes and found that pregnancy-specific alcohol policies are largely not associated with alcohol use during pregnancy. Research has also found that multiple pregnancy-specific alcohol policies—including Mandatory Warning Signs, Child Abuse/Neglect, Limits on Criminal Prosecution, and Priority Treatment for Pregnant Women—are associated with increased low birth weight and preterm birth and decreased prenatal care (see eTable 1 in Supplement 1 for definitions of the pregnancy-specific alcohol policies referenced in this article).12,13 This research suggests that pregnancy-specific alcohol policies are mostly ineffective and possibly harmful to general public health outcomes.
Previous literature on associations of pregnancy-specific alcohol policies with alcohol use is constrained by small numbers of pregnant people in national surveys that measure alcohol consumption and reliance on self-reported measures of alcohol consumption,11 in which willingness to report use could be influenced by the policy environment itself. This previous research also reports on outcomes (eg, low birth weight and preterm birth)13 that are not specific foci of policies and may include neither the specific birth defects associated with alcohol use during pregnancy nor child maltreatment, which pregnancy-specific alcohol policies target.
Research to understand whether findings from extant literature extend to other outcomes, including those specifically targeted by pregnancy-specific alcohol policies, is needed. The present study used a private insurance claims database to examine the associations of pregnancy-specific alcohol policies with infant injuries associated with maltreatment, infant morbidities, maternal morbidities, and infant health care utilization.
This retrospective cohort study examined the associations of state-level pregnancy-specific alcohol policies with infant and maternal outcomes and infant health care utilization. The University of California, San Francisco, institutional review board considered this deidentified data study exempt, and the Penn State Institutional Review Board considered this study not human participants research; thus, informed consent was not needed in accordance with 45 CFR §46. We followed Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
This study used policy data from the National Institute on Alcohol Abuse and Alcoholism’s Alcohol Policy Information System4 and outcome data from the Merative MarketScan Commercial Claims and Encounters database, a commercially available health insurance claims database. MarketScan contains claims for a sample of privately insured people in all 50 US states and the District of Columbia, including demographic characteristics, health care utilization, dates of service, diagnosis codes, procedure codes, and facility type. Data represent claims that have been adjudicated for payment and are obtained directly from a convenience sample of health plans and large employers that agree to participate in MarketScan.
Study Population and Exposures
The study population included all reproductive-aged female beneficiaries in the database aged 12 to 50 years who gave birth to a singleton between 2006 and 2019, had been continuously enrolled 1 year before and 1 year after delivery, could be matched under the same household with an infant who had at least 1 claim within the first month after delivery and was continuously enrolled for 1 year after birth, and resided in a US state or Washington, DC. Primary independent variables include dichotomous pregnancy-specific alcohol policies as specified by the National Institute on Alcohol Abuse and Alcoholism4: Reporting Requirements for Child Protective Services (CPS), Reporting Requirements for Assessment/Treatment, Reporting Requirements for Data, Mandatory Warning Signs, Child Abuse/Neglect, Limits on Criminal Prosecution, Civil Commitment, Priority Treatment for Pregnant Women Only, and Priority Treatment for Pregnant Women and Women With Children (see eTable 1 in Supplement 1 for definitions of the pregnancy-specific alcohol policies).
Primary outcomes include infant injuries associated with maltreatment and infant morbidities associated with alcohol use during pregnancy, each dichotomous. Infant injuries include having 1 or more International Classification of Diseases, Ninth Revision (ICD-9) and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes that previous research14 has found to have positive predictive value of greater than 50% for child maltreatment (eTable 2 in Supplement 1). Infant morbidities include having 1 or more ICD-9 and ICD-10 codes for infant morbidities that previous research15 has identified as being related to alcohol use during pregnancy (eTable 3 in Supplement 1).
Secondary outcomes include severe maternal morbidities, defined as having 1 or more ICD-9 and ICD-10 codes for a de novo severe maternal morbidity within 6 weeks after discharge from the delivery hospital (eTable 4 in Supplement 1).16,17 They also include 3 infant health care utilization outcomes: inadequate well-child visits (<4 well-child visits before the first birthday, informed by American Academy of Pediatrics18 and California Department of Health Care Services19 guidelines and data distribution) and 2 or more emergency department (ED) visits and 2 or more inpatient admissions (each dichotomous of ≥2 of this type of health care utilization during infant’s first year).
Individual-level controls include the birthing person’s age (categorized as 25-29, 30-34, 35-39, 40-44, and ≥45 years) and health status (Elixhauser Comorbidity Index,20 categorized as 0, 1, 2, and ≥3 comorbidities). State-level controls include unemployment, poverty, per capita alcohol consumption, and per capita tobacco consumption.21-25 For sensitivity analyses examining pregnancy-specific drug policies, we also included legal recreational cannabis.4 Pregnancy-specific alcohol policy data were merged with individual-level outcome data according to the estimated month and year the person became pregnant, accounting for preterm births.
Data were analyzed from August 2021 to April 2023. Multivariable logistic regression models examined associations between policies and individual-level outcomes. Models included all policy indicators simultaneously; adjusted for individual-level controls and, when Wald tests indicated state controls improved model fit, state-level controls; included fixed effects for state and year, as well as state-specific time trends; and accounted for clustering of SEs according to residence state. Unadjusted findings from models examining each policy in separate models are in eTable 5 in Supplement 1. The analysis assumes that missing data for the study cohort are rare because these are adjudicated billing claims used to determine payments to clinicians, hospitals, and pharmacies. Analyses were performed in Stata statistical software version 16.1 (StataCorp) and used a 2-sided statistical significance level of P < .05.
A series of sensitivity analyses were conducted. First, analyses examined whether findings were sensitive to changing the definition of when a policy was considered to be in effect for a given birthing person–infant pair from at conception to at birth. Second, analyses examined whether findings were sensitive to assumptions, including inclusion vs exclusion of birthing people younger than 25 years (because of the limited ability to match younger birthing people with infants), subsequent pregnancies, Civil Commitment and Limits on Criminal Prosecution (rarer policies that often coexist with other policies), and all state controls. Finally, pregnancy-specific drug policies were considered. Per a priori study plans, pregnancy-specific alcohol policies were established as main policies of interest. However, most state policies covering alcohol use during pregnancy also cover drugs.26 Because of this overlap, separately including pregnancy-specific alcohol and pregnancy-specific drug policies is infeasible. Instead, sensitivity analyses examined (1) pregnancy-specific drug and (2) pregnancy-specific alcohol and/or drug policies.
We identified 1 666 425 singleton birth and birthing person pairs that met inclusion criteria. We then excluded all births among people younger than age 25 years (91 228 births), because more than 70% were not matched with an infant. We excluded births where the infant did not have a claim within the first month after birth (142 218 births), because data on key study outcomes—particularly infant morbidities—appeared to be missing for infants without a claim within the first month. The final study cohort included a total of 1 432 979 birthing person-infant dyads (Figure 1).
In the study cohort, mean (SD) age of birthing people was 32.2 (4.2) years. Most (972 968 individuals [67.9%]) had no Elixhauser comorbidities whereas 37 630 (2.6%) had 3 or more comorbidities (Table 1). The number of states with each pregnancy-specific policy increased between 2005 and 2019, except Priority Treatment for Pregnant Women and Women With Children (Figure 2). More than 40% of birthing people were exposed to Reporting Requirements for CPS, Reporting Requirements for Data, or Reporting Requirements for Assessment/Treatment; Child Abuse/Neglect; or Mandatory Warning Signs. Between 10% and 25% were exposed to Priority Treatment, with more exposed to Priority Treatment for Pregnant Women Only than to Pregnant Women and Women With Children. Fewer than 10% were exposed to Civil Commitment or Limits on Criminal Prosecution (Table 1 and eTable 6 in Supplement 1).
Among the infants, 30 157 (2.1%) had an injury associated with maltreatment and 44 461 (3.1%) had a morbidity associated with alcohol use during pregnancy; 47 582 birthing people (3.3%) experienced a severe maternal morbidity. Regarding health care utilization, 148 170 infants (10.3%) had fewer than 4 well-child visits, 92 586 (6.5%) had 2 or more ED visits, and 32 345 (2.3%) had 2 or more inpatient hospitalizations.
In adjusted models, Reporting Requirements for Assessment/Treatment was associated with increased odds of infant injuries (adjusted odds ratio [aOR], 1.28; 95% CI, 1.08-1.52), but not infant or maternal morbidities (Table 2 and eTable 7 in Supplement 1). Mandatory Warning Signs was associated with increased odds of infant injuries (aOR, 1.18; 95% CI, 1.10-1.27) and maternal morbidities (aOR, 1.87; 95% CI, 1.72-2.03), but not infant morbidities. Civil Commitment was associated with increased odds of infant injuries (aOR, 1.26; 95% CI, 1.08-1.48), but decreased odds of infant morbidities (aOR, 0.57; 95% CI, 0.53-0.62), and was not associated with maternal morbidities. Priority Treatment for Pregnant Women Only was associated with decreased odds of infant injuries (aOR, 0.83; 95% CI, 0.76-0.90), but not infant or maternal morbidities. Priority Treatment for Pregnant Women and Women With Children was associated with increased odds of infant injuries (aOR, 1.12; 95% CI, 1.00-1.25) and infant morbidities (aOR, 1.08; 95% CI, 1.03-1.13), but decreased odds of severe maternal morbidities (aOR, 0.83; 95% CI, 0.70-0.97). Reporting Requirements for CPS, Reporting Requirements for Data, Child Abuse/Neglect, and Limits on Criminal Prosecution were not associated with infant injuries or with infant or maternal morbidities.
The findings were somewhat sensitive to models examining policies in effect when people gave birth. Specifically, associations of Mandatory Warning Signs and Civil Commitment with infant injuries and of Priority Treatment for Pregnant Women and Women With Children with maternal morbidities were no longer statistically significant (eTable 8 in Supplement 1). Two reporting policies (Reporting Requirements for CPS and Reporting Requirements for Data) that had not been associated with outcomes in main models were associated with increased infant morbidities and severe maternal morbidities. Infant morbidity findings were generally robust to policy timing, except Reporting Requirements for Data (with increased infant morbidities) and Priority Treatment for Pregnant Women Only (with decreased infant morbidities), both of which were statistically significant only when examining policies in effect when people gave birth. Findings were robust for all policies when testing modeling assumptions (data not shown), with the exception of Priority Treatment for Pregnant Women and Women With Children, which was no longer statistically significant in 1 or more sensitivity analyses.
Most findings did not vary when we examined drug or drug and/or alcohol policies. When findings varied, it was typically loss of statistical significance in drug-focused policy models. Other differences included Civil Commitment, for which severe maternal morbidities were newly associated with increased morbidities in the alcohol and/or drug policy model. The only change in direction was Mandatory Warning Signs, which was associated with decreased infant injuries in drug policy models but was associated with increased infant injuries in alcohol policy and alcohol and/or drug policy models (eTable 9 and eTable 10 in Supplement 1).
In adjusted models, Reporting Requirements for Assessment/Treatment (aOR, 0.81; 95% CI, 0.69-0.95) and Priority Treatment for Pregnant Women and Women With Children (aOR, 0.87; 95% CI, 0.83-0.92) were associated with decreased odds of 2 or more inpatient admissions but not other utilization (Table 3 and eTable 11 in Supplement 1). Mandatory Warning Signs was associated with increased odds of inadequate well-child visits (aOR, 1.12; 95% CI, 1.01-1.24) and decreased odds of 2 or more inpatient admissions (aOR, 0.84; 95% CI, 0.79-0.90), but was not associated with 2 or more ED visits. Civil Commitment was associated with increased odds of 2 or more ED visits (aOR, 1.31; 95% CI, 1.21-1.43) but not other utilization. Limits on Criminal Prosecution was associated with decreased odds of 2 or more ED visits (aOR, 0.82; 95% CI, 0.77-0.87) and 2 or more inpatient admissions (aOR, 0.88; 95% CI, 0.82-0.94). Priority Treatment for Pregnant Women Only was associated with increased odds of inadequate well-child visits (aOR, 1.13; 95% CI, 1.03-1.23) but not other utilization. Reporting Requirements for CPS, Reporting Requirements for Data, and Child Abuse/Neglect were not associated with any utilization outcomes.
The findings were sensitive to models examining policies in effect when people gave birth. Findings for 4 policies for inadequate well-child visits, 3 policies for 2 or more ED visits, and 7 policies for 2 or more inpatient admissions differed when timing of policies was varied. For most, these were variations in statistical significance or effect magnitude; for Mandatory Warning Signs, however, the direction for 2 or more ED visits changed when examining policies in effect when people gave birth (eTable 8 in Supplement 1).
When we examined the modeling assumptions, all findings for inadequate well-child visits were robust, but findings for 2 or more ED visits and 2 or more inpatient admissions were not, although these changes were mostly statistical significance rather than effect magnitude or direction. The only exception was Civil Commitment, for which the association was in the opposite direction in the model without Limits on Criminal Prosecution (data not shown).
With the exception of 2 or more ED visits, utilization findings were somewhat sensitive to examinations of drug or alcohol and/or drug policies. Findings for 2 or more inpatient admissions were more sensitive to drug vs alcohol focus of policies. Most changes were statistical significance and not effect direction or magnitude. The only exception was for Mandatory Warning Signs, for which findings for drug-focused Mandatory Warning Signs was in the opposite direction (eTable 9 and eTable 10 in Supplement 1).
In this retrospective cohort study, 4 of the 9 state-level pregnancy-specific alcohol policies, including Reporting Requirements for CPS, Reporting Requirements for Data, Child Abuse/Neglect, and Limits on Criminal Prosecution, were not associated with infant maltreatment or infant or maternal morbidities. Two policies, Reporting Requirements for Assessment/Treatment and Mandatory Warning Signs, were associated with increased infant maltreatment and/or infant or maternal morbidities without being offset by a decrease in another outcome. To the extent any pregnancy-specific alcohol policy was associated with improvements in infant or maternal outcomes or health care utilization, that policy was associated with increased odds of another outcome. Although some individual associations of policies with outcomes were not robust across sensitivity analyses, sensitivity analyses did not change overall interpretation of findings.
Our findings that state-level pregnancy-specific alcohol policies are generally not associated with improved health outcomes and, in some cases, are associated with increased adverse outcomes are consistent with previous literature11,13 examining associations of these policies with alcohol use during pregnancy, birth outcomes, and prenatal care utilization. These findings are also consistent with the slightly larger body of literature examining pregnancy-specific drug policies, which also tends to find few infant health or health care utilization benefits associated with pregnancy-specific drug policies.27,28 Combined with previous research, these results provide further evidence that adopting more of the extant type of pregnancy-specific alcohol policies should be paused and that repeal, or at least revision and improvement, of some policies is warranted.
The fact that most findings did not differ between pregnancy-specific drug and pregnancy-specific alcohol policies likely reflects that pregnancy-specific substance-use policies typically cover both alcohol and drugs26 and that policy makers passing them may not know whether they cover alcohol, drugs, or both.29 The findings appear to apply to both pregnancy-specific alcohol and pregnancy-specific drug policies, noting that sensitivity analyses were not conducted for drug-focused policies and that the infant morbidity outcome related to alcohol consumption during pregnancy. The only exception is Mandatory Warning Signs, for which findings substantively differed between alcohol-focused vs drug-focused policies and which historically have existed for alcohol, but not drugs.26
Although it was not the main focus, the prevalence of infant injuries, infant morbidities, and severe maternal morbidities (2%-3%) provides additional evidence that infants and birthing people face health burdens. Pregnancy-specific alcohol policies do not appear to reduce these burdens for infants or birthing people and may exacerbate the problems. Alternative policy approaches that improve infant and maternal health and well-being are urgently needed.
Strengths and Limitations
This study has strengths. First, this study uses policy data rigorously coded for social science research. Second, outcomes are based on claims data, which is one of the few options for assessing study outcomes. Although an ideal study would examine fetal alcohol spectrum disorder directly, fetal alcohol spectrum disorder surveillance systems vary considerably across states and are not in place in all states; there are also considerable concerns with variations among and underreporting in these systems.30,31 Although national databases tracking child maltreatment exist,32 these databases track reports to CPS and outcomes of these reports rather than maltreatment. The reports are also likely influenced by the CPS reporting policies we propose to study33 and, thus, may measure policy compliance, rather than maltreatment. Third, this study examined infant and maternal outcomes together, rather than separately, which places focus on the family unit, a more relevant way to assess health and well-being of caregiver and child.
There are also limitations to this study. First, infant outcome measurement is imprecise. For infant injuries consistent with maltreatment, we used ICD-9 and ICD-10 codes with a positive predictive value of 50%, rather than 75%, with maltreatment. This meant we had sufficient sample for analyses but we made assumptions about injury causes. Second, we were unable to restrict analyses to people with ICD-9 and ICD-10 codes for alcohol-related diagnoses or medications because of the rarity of alcohol-related codes in the data. Thus, examined injuries and morbidities cannot be assumed to be due to alcohol consumption during pregnancy. Third, these claims data are from people with employer-sponsored insurance; the findings may not generalize to people with other insurance. Fourth, we are missing individual controls, such as economic status. Given that all data were from people covered by employer-sponsored insurance, there may be less variation in economic status. None of the missing individual-level control variables, however, would be expected to confound associations of state-level policies with outcomes. Fifth, although the analysis cohort can be assumed to have nearly complete data, given that these are adjudicated claims, having to exclude people from the cohort may have influenced the findings. Sixth, the main analyses excluded birthing people younger than 25 years because so few of them were matched with an infant. This likely reflects health insurance patterns for young people, many of whom are still covered under a parent’s insurance and, thus, are unable to place their infant on the same insurance. Sensitivity analyses examining whether inclusion of the young birthing people we could match with an infant did not substantively affect the findings.
In this retrospective cohort study, most pregnancy-specific alcohol policies were not associated with decreased infant injuries or morbidities or maternal morbidities, and were sometimes associated with increased odds of at least 1 adverse infant or maternal outcome. Policy makers should not assume that pregnancy-specific alcohol policies improve infant or maternal health. Policy approaches more likely to improve infant and maternal health are urgently needed.
Accepted for Publication: June 23, 2023.
Published: August 3, 2023. doi:10.1001/jamanetworkopen.2023.27138
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2023 Roberts SCM et al. JAMA Network Open.
Corresponding Author: Sarah C. M. Roberts, DrPH, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, 1330 Broadway, Ste 1100, Oakland, CA 94612 (sarah.roberts@ucsf.edu).
Author Contributions: Drs Roberts and Liu had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Roberts, Schulte, Leslie, Liu.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Roberts, Schulte, Corr.
Critical review of the manuscript for important intellectual content: Schulte, Zaugg, Leslie, Liu.
Statistical analysis: Roberts, Schulte, Leslie, Liu.
Obtained funding: Roberts.
Administrative, technical, or material support: Roberts, Schulte, Zaugg, Leslie, Liu.
Supervision: Roberts, Leslie.
Conflict of Interest Disclosures: Dr Roberts reported receiving grants from the Foundation for Opioid Response Efforts outside the submitted work. No other disclosures were reported.
Funding/Support: This work was supported by the US National Institute on Alcohol Abuse and Alcoholism at the National Institutes of Health (grant 2R01AA023267).
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Data Sharing Statement: See Supplement 2.
Additional Contributions: Sue Thomas, PhD, and Ryan Treffers, JD (both from Pacific Institute for Research and Evaluation), coded the pregnancy-specific alcohol and drug policy data. They were compensated for this work partly with funding from R01AA023267 as well as their ongoing contract for National Institute on Alcohol Abuse and Alcoholism’s Alcohol Policy Information System (https://alcoholpolicy.niaaa.nih.gov/about-apis).
2.May
PA, de Vries
MM, Marais
AS,
et al. The prevalence of fetal alcohol spectrum disorders in rural communities in South Africa: a third regional sample of child characteristics and maternal risk factors.
Alcohol Clin Exp Res. 2022;46(10):1819-1836. doi:
10.1111/acer.14922PubMedGoogle ScholarCrossref 5.Centers for Disease Control and Prevention (CDC). Alcohol consumption among pregnant and childbearing-aged women—United States, 1991 and 1995.
MMWR Morb Mortal Wkly Rep. 1997;46(16):346-350.
PubMedGoogle Scholar 6.Centers for Disease Control and Prevention (CDC). Alcohol use among women of childbearing age—United States, 1991-1999.
MMWR Morb Mortal Wkly Rep. 2002;51(13):273-276.
PubMedGoogle Scholar 7.Centers for Disease Control and Prevention (CDC). Alcohol use among pregnant and nonpregnant women of childbearing age—United States, 1991-2005.
MMWR Morb Mortal Wkly Rep. 2009;58(19):529-532.
PubMedGoogle Scholar 9.Denny
CH, Acero
CS, Naimi
TS, Kim
SY. Consumption of alcohol beverages and binge drinking among pregnant women aged 18-44 years—United States, 2015-2017.
MMWR Morb Mortal Wkly Rep. 2019;68(16):365-368. doi:
10.15585/mmwr.mm6816a1PubMedGoogle ScholarCrossref 11.Roberts
SCM, Mericle
AA, Subbaraman
MS,
et al. State policies targeting alcohol use during pregnancy and alcohol use among pregnant women 1985-2016: evidence from the Behavioral Risk Factor Surveillance System.
Womens Health Issues. 2019;29(3):213-221. doi:
10.1016/j.whi.2019.02.001PubMedGoogle ScholarCrossref 13.Subbaraman
MS, Thomas
S, Treffers
R,
et al. Associations between state-level policies regarding alcohol use among pregnant women, adverse birth outcomes, and prenatal care utilization: results from 1972 to 2013 Vital Statistics.
Alcohol Clin Exp Res. 2018;42(8):1511-1517. doi:
10.1111/acer.13804PubMedGoogle ScholarCrossref 14.Syed
S, Ashwick
R, Schlosser
M, Gonzalez-Izquierdo
A, Li
L, Gilbert
R. Predictive value of indicators for identifying child maltreatment and intimate partner violence in coded electronic health records: a systematic review and meta-analysis.
Arch Dis Child. 2021;106(1):44-53. doi:
10.1136/archdischild-2020-319027PubMedGoogle ScholarCrossref 15.O’Leary
CM, Elliott
EJ, Nassar
N, Bower
C. Exploring the potential to use data linkage for investigating the relationship between birth defects and prenatal alcohol exposure.
Birth Defects Res A Clin Mol Teratol. 2013;97(7):497-504. doi:
10.1002/bdra.23142PubMedGoogle ScholarCrossref 20.Moore
BJ, White
S, Washington
R, Coenen
N, Elixhauser
A. Identifying increased risk of readmission and in-hospital mortality using hospital administrative data: the AHRQ Elixhauser Comorbidity Index.
Med Care. 2017;55(7):698-705. doi:
10.1097/MLR.0000000000000735PubMedGoogle ScholarCrossref 22.Nephew
TM, Yi
H, Williams
GD, Stinson
FS, Dufour
MC. US alcohol epidemiologic data reference manual, Volume 1, 4th Edition. US apparent consumption of alcoholic beverages based on state sales, taxation, or receipt data. NIH Publication No. 04-5563. June 2004. Accessed June 29, 2023.
https://pubs.niaaa.nih.gov/publications/DRM_04-5563/DRM_04-5563.pdf 30.Andrews
JG, Galindo
MK, Meaney
FJ,
et al. Recognition of clinical characteristics for population-based surveillance of fetal alcohol syndrome.
Birth Defects Res. 2018;110(10):851-862. doi:
10.1002/bdr2.1203PubMedGoogle ScholarCrossref 31.O’Leary
LA, Ortiz
L, Montgomery
A,
et al; FASSNetII. Methods for surveillance of fetal alcohol syndrome: the Fetal Alcohol Syndrome Surveillance Network II (FASSNetII)—Arizona, Colorado, New York, 2009-2014.
Birth Defects Res A Clin Mol Teratol. 2015;103(3):196-202. doi:
10.1002/bdra.23335PubMedGoogle ScholarCrossref