Abstract
Objective
Substance use disorder is a growing concern in the USA, especially among pregnant women. This study was undertaken to assess the impact of substance use disorder on adverse pregnancy outcomes using a nationwide sample of inpatient pregnancy hospitalizations in the USA, and to elucidate the influence on each type of adverse pregnancy outcome.
Study design
A cross-sectional analysis of inpatient pregnancy hospitalizations in the USA from the Healthcare Cost and Utilization Project National Inpatient Sample from 2016 to 2020 was conducted. International Classification of Diseases – 10th revision and diagnosis-related group codes were used to identify inpatient pregnancy-related delivery hospitalizations with a substance use disorder and/or adverse pregnancy outcomes. Propensity score matching and multiple logistic regression analyses were undertaken to predict the likelihood of adverse pregnancy outcomes among pregnancy hospitalizations with and without substance use disorder. Subgroup analyses were performed to estimate the impact of substance use disorder on each adverse pregnancy outcome.
Results
From 3,238,558 hospitalizations, the prevalence of adverse pregnancy outcomes was substantially higher among pregnancy hospitalizations with substance use disorder (35.6 %) compared with pregnancy hospitalizations without substance use disorder (25.1 %, p < 0.001). After matching and model adjustment for sociodemographic covariates, substance use disorder was identified as an independent predictor of adverse pregnancy outcomes [adjusted odds ratio (aOR) 1.47, 95 % confidence interval (CI) 1.45–1.49]. In subgroup analyses based on type of adverse pregnancy outcome, the greatest exposure risks were fetal growth restriction (aOR 1.96, 95 % CI 1.91–2.01), antepartum hemorrhage (aOR 1.79, 95 % CI 1.73–1.85) and preterm birth (aOR 1.65, 95 % CI 1.62–1.68).
Conclusion
Patients with substance use disorder are at higher risk of adverse pregnancy outcomes, particularly fetal growth restriction, antepartum hemorrhage and preterm birth.
Keywords: Adverse pregnancy outcome, Fetal growth restriction, Antepartum hemorrhage, Preterm birth, Substance use disorder
Highlights
-
•
Pregnancy hospitalizations with substance use have a greater risk of adverse pregnancy outcomes.
-
•
The highest risk was noted for fetal growth restriction, antepartum hemorrhage and preterm birth.
-
•
While adjusting for sociodemographic factors, substance use remained an independent predictor of adverse pregnancy outcomes.
1. Introduction
Substance use disorder (SUD) is a growing concern, with the impact spanning broadly into all stages of life. Pregnancy represents a timeframe with increased access to health care, during which patients with SUD may have more opportunities to seek treatment given the heightened motivation to work towards optimizing maternal and fetal health [1], [2], [3].
The Diagnostic and Statistical Manual of Mental Disorders Fifth Edition defines SUD as a ‘cluster of cognitive, behavioral, and physiological symptoms indicating that the individual continues using the substance despite significant substance-related problems’ [4]. This definition encompasses all drug classes including alcohol, caffeine, cannabis, hallucinogens, inhalants, opioids, sedatives, hypnotics, anxiolytics, stimulants, tobacco, and other or unknown substances [4]. According to the 2020 National Survey of Drug Use and Health in the USA, 15.4 % (38.7 million) of individuals aged ≥ 18 years had SUD, and between 8 % and 11 % of pregnant women had used illicit drugs, tobacco or alcohol in the past month [5]. Recognizing the significant number of pregnant patients affected by SUD, it is critical to understand the impact on adverse pregnancy outcomes (APOs).
Previous studies have established the detrimental effects of substance use on neonatal outcomes, including increased risk of congenital anomalies, impaired growth, neonatal abstinence syndrome, and potential long-term intellectual disability [6]. Pregnant patients with SUD are at risk of insufficient or late presentation to prenatal care, increased utilization of emergency department visits, and APOs [7], [8]. Previous studies have failed to provide an in-depth understanding of the impact of SUD on individual APOs.
This study was undertaken to evaluate whether the prevalence of APOs was significantly different in a nationwide inpatient sample of pregnancy-related delivery hospitalizations in the USA with and without SUD, and to determine if SUD was an independent risk factor for predicting risk of an APO. This study also sought to elucidate the impact on each of the individual components of the APO composite. It was hypothesized that admissions among pregnant women with SUD will have a higher prevalence of APOs, and that SUD will be identified as an independent risk factor for predicting APOs.
2. Materials and methods
A cross-sectional analysis of the Healthcare Cost and Utilization Project (HCUP) National Inpatient Sample (NIS), the largest publicly available all-payer database released annually in the USA, was performed. HCUP-NIS provides national estimates of inpatient utilization, access, cost, quality and outcomes with diagnosis and procedure codes, with a weighted evaluation of 35 million hospitalizations [9]. This study used the 2016–2020 NIS datasets. The years of study were chosen after the transition from International Classification of Diseases 9th revision (ICD-9) to ICD-10 in order to avoid inaccuracies in the conversion between coding standards. Hospital admissions were identified using ICD-10 diagnosis and procedure codes pertaining to an ongoing pregnancy. Diagnosis-related group (DRG) codes were also used to specifically identify admissions with a DRG code related to a delivery episode, in order to avoid duplication of patients with multiple inpatient hospitalizations over the course of pregnancy. ICD-10 codes were also used to determine APO diagnoses and the presence of SUD (Appendix A). SUD – the exposure of interest in this study – was defined as a composite between alcohol, tobacco, cannabis, cocaine, sedatives, opioids, hallucinogens, stimulant and polysubstance use disorder. APO – the primary outcome of interest in this study – was also structured as a composite to include hypertensive disorders of pregnancy (HDP), antepartum hemorrhage (AH), postpartum hemorrhage (PPH), preterm birth (PTB) and fetal growth restriction (FGR). HDP notably included gestational hypertension, pre-eclampsia, eclampsia, and pre-existing hypertension with superimposed pre-eclampsia. Similarly, AH as a category included ICD-10 codes for the clinical diagnoses of placenta previa and placental abruption, among other types of antenatal hemorrhage. Inpatient hospitalizations were categorized as pregnancies with and without SUD.
The dataset was limited to records with no missing data for the variables of interest. The prevalence rates of APOs in admissions with and without SUD were reported for the study sample overall, by cohort year, and by specific type of SUD. Hospital admission rates for APOs with and without SUD were computed and analyzed by age, gender, race, insurance payer (governmental, private, other and missing), median household income for patient’s zip code, and year of the cohort. Hospital characteristics, including hospital region, urban–rural location, bed size and teaching status, were also considered in this analysis [10], [11], [12].
Patients without SUD were matched to patients with SUD in a 2:1 ratio using propensity scores derived from logistic regression using cohort year and all individual and hospital characteristics described above. After matching, survey-weighted proportions were computed for variables of interest in the whole sample, and stratified by SUD exposure. Rao Scott Chi-squared tests were used to compare the distribution of patients across the levels of exposure. Logistic regression analyses were used to estimate the adjusted odds ratios (aOR) and associated 95 % confidence intervals (CI) of the patient characteristics influencing APOs and each of the APO subcategories. All analyzes accounted for the complex survey structure of HCUP by incorporating hospital ID as a clustering variable, survey year and NIS stratum as strata, and discharge weight as survey weight, providing the national estimates [13]. Statistics were computed using SAS Version 9.4 (SAS Institute Inc., Cary, NC, USA) and R Version 4.3.1 [14]. Matching and complex survey design were performed using the R packages MatchIt [15] and survey [16], respectively. The NIS database is de-identified, so this study was deemed exempt by the local institutional review board.
3. Results
From 2016 to 2020, a total of 3,407,166 inpatient pregnancy hospitalizations were identified from the NIS database, of which 3,238,558 (95 %) contained complete information (Appendix B). Of these admissions, 194,930 (6.0 %) were complicated by SUD. Compared with their counterparts, pregnancy hospitalizations among women with SUD were more likely to involve individuals of White ethnicity, between the ages of 20 and 30 years, more often receiving governmental insurance, and within the lowest income quartile (Table 1). Upon evaluation of hospital level characteristics, inpatient admissions among pregnant women with SUD were noted more often among hospitalizations from the Midwestern states, in large hospitals (based on bed size), and in rural hospitals. Distribution of pregnancy hospitalizations was balanced overall over the 5 years, being evaluated with a small annual increase in number of hospitalizations with SUD from 2016 to 2019 (Table 1).
Table 1.
Descriptive statistics of demographic and hospital characteristics.
| Without SUD | With SUD | p-value | SMD | |
|---|---|---|---|---|
| n | 3,043,628 | 194,930 | ||
| Year (%) | < 0.001 | 0.062 | ||
| 2016 | 582,749 (19.1) | 33,903 (17.4) | ||
| 2017 | 586,501 (19.3) | 35,771 (18.4) | ||
| 2018 | 606,590 (19.9) | 38,714 (19.9) | ||
| 2019 | 646,043 (21.2) | 44,069 (22.6) | ||
| 2020 | 621,745 (20.4) | 42,473 (21.8) | ||
| Age (%) | < 0.001 | 0.257 | ||
| < 20 | 133,106 (4.4) | 10,083 (5.2) | ||
| 20–29 | 1,450,598 (47.7) | 115,594 (59.3) | ||
| 30–39 | 1,356,289 (44.6) | 65,228 (33.5) | ||
| ≥ 40 | 103,635 (3.4) | 4025 (2.1) | ||
| Ethnicity (%) | < 0.001 | 0.529 | ||
| White | 1,574,917 (51.7) | 134,617 (69.1) | ||
| Black | 444,869 (14.6) | 36,571 (18.8) | ||
| Hispanic | 660,528 (21.7) | 14,322 (7.3) | ||
| Other | 363,314 (11.9) | 9420 (4.8) | ||
| Insurance payer (%) | < 0.001 | 0.793 | ||
| Government | 1,243,769 (40.9) | 149,615 (76.8) | ||
| Private | 1,639,097 (53.9) | 38,305 (19.7) | ||
| Other | 160,762 (5.3) | 7010 (3.6) | ||
| Median zip code income (%) | < 0.001 | 0.503 | ||
| 0–25th percentile | 823,635 (27.1) | 82,698 (42.4) | ||
| 26th–50th percentile | 759,446 (25.0) | 59,326 (30.4) | ||
| 51st–75th percentile | 760,132 (25.0) | 37,160 (19.1) | ||
| 76th–100th percentile | 700,415 (23.0) | 15,746 (8.1) | ||
| Hospital size (%) | < 0.001 | 0.066 | ||
| Small | 580,511 (19.1) | 38,824 (19.9) | ||
| Medium | 928,869 (30.5) | 53,616 (27.5) | ||
| Large | 1,534,248 (50.4) | 102,490 (52.6) | ||
| Hospital type (%) | < 0.001 | 0.271 | ||
| Rural | 249,207 (8.2) | 33,309 (17.1) | ||
| Urban non-teaching | 622,491 (20.5) | 34,759 (17.8) | ||
| Urban teaching | 2,171,930 (71.4) | 126,862 (65.1) | ||
| Hospital region (%) | < 0.001 | 0.263 | ||
| Northeast | 497,041 (16.3) | 28,813 (14.8) | ||
| Midwest | 599,726 (19.7) | 57,579 (29.5) | ||
| South | 1,220,944 (40.1) | 76,401 (39.2) | ||
| West | 725,917 (23.9) | 32,137 (16.5) | ||
| Adverse pregnancy outcome (%) | 763,235 (25.1) | 69,491 (35.6) | < 0.001 | 0.231 |
| Hypertensive disorders of pregnancy | 414,435 (13.6) | 32,074 (16.5) | < 0.001 | 0.079 |
| Antepartum hemorrhage | 48,131 (1.6) | 5726 (2.9) | < 0.001 | 0.091 |
| Postpartum hemorrhage | 92,047 (3.0) | 6002 (3.1) | 0.173 | 0.003 |
| Preterm birth | 299,634 (9.8) | 32,773 (16.8) | < 0.001 | 0.206 |
| Fetal growth restriction | 101,803 (3.3) | 14,084 (7.2) | < 0.001 | 0.174 |
SUD, substance use disorder; SMR, standardized mean difference.
Over the study period, there were 832,726 (25.7 %) pregnancy-related hospital admissions with an associated diagnosis of an APO. The prevalence of APOs was substantially higher among pregnancy hospitalizations with SUD [69,491/194,930 (35.6 %)] compared with pregnancy hospitalizations without SUD [763,235/3,043,628 (25.1 %); p < 0.001] overall, as well as for each individual type of SUD (Appendix C). At population level, the prevalence of APOs increased over the 5-year study period among pregnancy hospitalizations overall, with and without SUD (Table 2).
Table 2.
Rates of adverse pregnancy outcomes across time in pregnancy inpatients, Healthcare Cost and Utilization Project National Inpatient Sample, 2016–2020.
| Year | APO prevalence n = 3,238,558 |
APO prevalence among admissions with SUD exposure n = 194,930 |
APO prevalence among admissions without SUD exposure n = 3,043,628 |
p-value |
|---|---|---|---|---|
| 2016–2020 | 832,726/3,238,558 (25.7 %) | 69,491/194,930 (35.6 %) | 763,235/3,043,628 (25.1 %) | < 0.0001 |
| 2016 | 126,492/616,652 (20.5 %) | 10,012/33,903 (29.5 %) | 116,480/582,749 (20.0 %) | < 0.0001 |
| 2017 | 140,680/622,272 (22.6 %) | 11,451/35,771 (32.0 %) | 129,229/586,501 (22.0 %) | < 0.0001 |
| 2018 | 161,602/645,304 (25.0 %) | 13,483/38,714 (34.8 %) | 148,119/606,590 (24.4 %) | < 0.0001 |
| 2019 | 201,134/690,112 (29.1 %) | 17,148/44,069 (38.9 %) | 183,986/646,043 (28.5 %) | < 0.0001 |
| 2020 | 202,818/664,218 (30.5 %) | 17,397/42,473 (41.0 %) | 185,421/621,745 (29.8 %) | < 0.0001 |
APO, adverse pregnancy outcome; SUD, substance use disorder.
Percentages reflect the prevalence of APOs in respective groups.
Propensity score matching led to near-perfect balance in all measured sociodemographic and hospital characteristics (Appendix D). Adjusting for relevant sociodemographic and hospital characteristics in the matched data set, SUD was identified as an independent predictor of APOs (aOR 1.47, 95 % CI 1.45–1.49) (Fig. 1, Table 3). Additional independent predictors of APOs included age > 40 years (aOR 1.64, 95 % CI 1.57–1.72), Black ethnicity (aOR 1.39, 95 % CI 1.37–1.42), government insurance (aOR 1.05, 95 % CI 1.03–1.06) and lower income, with the largest income-related predicting factor being income in the 0–25th percentile (aOR 1.17, 95 % CI 1.15–1.20) (Table 3).
Fig. 1.
Post-matching adjusted odds ratios between substance use disorder and adverse pregnancy outcomes. This figure shows adjusted odds ratios, after matching, for substance use disorder and each adverse pregnancy outcome in inpatient pregnancy admissions in the Healthcare Cost and Utilization Project National Inpatient Sample, 2016–2020.
Table 3.
Predictors of adverse pregnancy outcomes in inpatient pregnancy admissions.
| Characteristic | OR | 95 % CI | p-value |
|---|---|---|---|
| Substance use disorder | 1.47 | 1.45–1.49 | < 0.001 |
| Year | < 0.001 | ||
| 2016 | — | — | |
| 2017 | 1.09 | 1.06–1.13 | |
| 2018 | 1.26 | 1.22–1.30 | |
| 2019 | 1.49 | 1.44–1.53 | |
| 2020 | 1.60 | 1.55–1.65 | |
| Age (years) | < 0.001 | ||
| < 20 | — | — | |
| 20–29 | 0.89 | 0.86–0.91 | |
| 30–39 | 1.10 | 1.07–1.13 | |
| ≥ 40 | 1.64 | 1.57–1.72 | |
| Ethnicity | < 0.001 | ||
| White | — | — | |
| Black | 1.39 | 1.37–1.42 | |
| Hispanic | 0.97 | 0.95–1.00 | |
| Other | 1.02 | 0.99–1.05 | |
| Insurance payer | < 0.001 | ||
| Government | — | — | |
| Private | 0.95 | 0.94–0.97 | |
| Other | 0.96 | 0.93–0.99 | |
| Median zip code income | < 0.001 | ||
| 0–25th percentile | — | — | |
| 26th–50th percentile | 0.93 | 0.92–0.95 | |
| 51st–75th percentile | 0.90 | 0.88–0.91 | |
| 76th–100th percentile | 0.85 | 0.83–0.87 | |
| Hospital size | < 0.001 | ||
| Small | — | — | |
| Medium | 1.11 | 1.08–1.14 | |
| Large | 1.36 | 1.32–1.39 | |
| Hospital type | < 0.001 | ||
| Rural | — | — | |
| Urban non-teaching | 1.11 | 1.08–1.14 | |
| Urban teaching | 1.54 | 1.50–1.58 | |
| Hospital region | < 0.001 | ||
| Northeast | — | — | |
| Midwest | 1.05 | 1.02–1.09 | |
| South | 1.15 | 1.11–1.19 | |
| West | 1.01 | 0.98–1.05 |
Additional analyses evaluating the impact of SUD on individual components of the APO composite are presented in Fig. 1. The presence of SUD in pregnancy hospitalizations was found to be associated with an increase in the likelihood for each category of APO, except for PPH (aOR 1.02, 95 % CI 0.99–1.06), with the highest risk noted for FGR (aOR 1.96, 95 % CI 1.91–2.01) followed by AH (aOR 1.79, 95 % CI 1.73–1.85) and PTB (aOR 1.65, 95 % CI 1.62–1.68). Pregnancy-related hospitalizations among women with SUD were also noted to increase the risk of HDP (aOR 1.08, 95 % CI 1.07–1.10), albeit to a smaller extent (Fig. 1). Full results for each APO component model are provided in Appendices E–I, Appendices E–I, Appendices E–I, Appendices E–I, Appendices E–I.
4. Discussion
4.1. Principal findings
This nationally representative study in the USA, evaluating the relationship between SUD and APOs, found that hospital admissions among pregnancies complicated by SUD had a substantially higher prevalence of APOs compared with pregnancies without SUD, with an overall difference of > 10 % noted in the study sample. After performing propensity score matching of sociodemographic factors of admissions with and without substance use, the impact of SUD as an independent predictor of APO was demonstrated. SUD was found to impact various types of APOs, with the highest risk noted for the occurrence of FGR, AH and PTB.
4.2. Results in context
Despite the high prevalence of substance use in women of reproductive age, there are limited data to help understand the impact of substance use on pregnancy overall, especially in the context of deriving insights from a nationwide inpatient database. However, perinatal outcomes have been evaluated internationally, and at state/local level, and support the present findings of increased risk of APOs in this clinical cohort [8], [17], [18], [19], [20], [21].
This study found that SUD was an independent predictor of APOs, with the highest risk noted for FGR. This finding is in alignment with several studies looking at the effect of specific substances on fetal growth. A study by Bailey et al., which examined the impact of opioid use on newborn outcomes, found that infants exposed to opioids were more than twice as likely to be diagnosed with growth restriction and decreased birth weight [22]. Additionally, Janisse et al. examined the effect of several types of substances on PTB and FGR using path analysis of the impact of substance abuse on decreased birth weight. Among the substances studied, exposure to alcohol, cigarettes and marijuana were noted to have the greatest negative impact on fetal growth [23]. These findings support the present results.
SUD was also found to predict increased likelihood of AH and PTB. Previous studies supported these findings, by identifying drug dependence as a risk factor for placental abruption, in addition to some evidence supporting increased risk of placenta previa with tobacco and cocaine use [24], [25]. Additionally, SUD in pregnancy has been found to increase the risk of PTB, although this is based on studies with smaller populations [8], [17], [19]. Finally, this study found that SUD was predictive of increased risk of HDP, although the strength of this association was weaker compared with FGR, AH and PTB, with no significant increase in the risk of PPH. Future research is warranted to examine the differential impact of specific SUD on perinatal outcomes, with a specific impact on maternal health.
4.3. Clinical implications
In this large nationwide sample of inpatient hospitalizations, the results provide supporting evidence that SUD is an independent risk factor for adverse pregnancy outcomes. In contrast to previous research endeavors focused on an individual type of illicit substance or a specific adverse outcome per se [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], the methodology employed in the present study was more expansive by using a nationwide database, and attempted to limit bias and account for confounding variables through propensity score matching. Hence, the results add significantly to the existing evidence base examining the impact of SUD on APOs, while also allowing the risk of each individual component of the APO composite to be dissected. Clinicians and policy makers can use this information to target interventions at those with the highest unmet need.
4.4. Research implications
While this study provides supporting evidence to enhance the understanding of complications faced by pregnancies affected by SUD, it is just the beginning in implementing solutions to address the growing disease burden of SUD in pregnancy. Further research is needed to examine the impact of individual types of SUD on pregnancy outcomes; however, this is challenging given the high prevalence of polysubstance use in this population. The impact of co-occurring SUDs on perinatal outcomes is poorly understood, and represents an extension of the research agenda for the future. Further research identifying individual, interpersonal, health system level and community/societal level predictors of adverse perinatal outcomes among women with SUD is an essential next step. This will enable the identification and mitigation of risk factors that may otherwise remain neglected, such as understanding the impact of social determinants of health on APOs in the context of SUD.
4.5. Strengths and limitations
This study provides in-depth analyses on the risk of APOs among pregnant women with SUD. The strength of this study stems from its ability to evaluate over three million pregnancies from a representative nationwide sample. By using propensity score matching, bias was mitigated to produce more generalizable results. This study therefore bolsters the existing evidence base about the relationship between SUD and APOs by providing insights from a population health level.
However, the study is not without limitations. Despite the benefits of using a large database, it was not possible to confirm or refute inaccuracies in ICD-10 coding at the hospital or individual patient level, and it is recognized that coding may be susceptible to errors. There were also limitations in the ability to understand the context, timing and etiology of diagnoses and outcomes. ICD-10 codes were used for identification of the exposures and outcomes of interest. Therefore, the definitions of SUD and APO used for this study are contingent upon data interpretation and abstraction from electronic medical records, and application of the appropriate ICD-10 codes. Given the reliance on ICD-10 codes, it was not possible to confirm or refute the clinical patterns of substance use in the study cohort. ICD-10 codes within these pregnancy-related admissions cannot distinguish between etiology, timing and misclassification of events. ICD-10 codes limit the ability to distinguish between true SUD and sporadic substance use. This study aimed to choose ICD-10 codes which would be the most inclusive to identify and encompass all individuals impacted by SUD. By operating within the scope of a database, it is recognized that not all predictors of APO may have been addressed, given that the analysis was limited to the collected sociodemographic factors as described.
Additional limitations exist given the nature of the NIS database and information available for analysis. It is acknowledged that most investigations evaluating the impact of perinatal SUD would be remiss without a discussion of neonatal outcomes, including neonatal abstinence syndrome. However, the focus of this paper was APOs, particularly in the maternal context. The NIS database does not provide insights into neonatal outcomes, and does not allow the linkage of maternal data with neonatal information from related administrative datasets.
5. Conclusions
In summary, this study showed that pregnancies complicated by SUD are at a substantially higher risk of APOs, after accounting for confounding sociodemographic factors. The analysis provides evidence that SUD in pregnancy is an independent risk factor for APOs in adjusted analysis. Although this risk is increased with most types of APOs, the varying degrees of predictive risk stratification present an opportunity for provider education, and should be considered during patient counseling.
CRediT authorship contribution statement
Justine Keller: Conceptualization, Data curation, Writing – original draft. Niraj Chavan: Conceptualization, Data curation, Methodology, Supervision, Writing – review & editing. Alexandra Ragsdale: Conceptualization, Methodology, Writing – original draft, Writing – review & editing. Noor AL-HAMMADI: Conceptualization, Formal analysis, Methodology, Writing – original draft, Writing – review & editing. Sabel BASS: Conceptualization, Data curation, Formal analysis, Methodology. Travis M Loux: Data curation, Formal analysis, Methodology, Writing – review & editing.
Funding
This work was supported by the Saint Louis University Research Institute and resources from the Advanced HEAlth Data (AHEAD) Institute.
Declaration of Competing Interest
The authors have no conflicts of interest to declare.
Appendix A
Diagnosis and procedure codes used in the analysis.
| Condition | ICD-10 code |
|---|---|
| Pregnancy, childbirth and delivery | O00.0, O10X, O11, O12X, O13, O14X, O15X, O16, O20X, O21X, O22X, O23X, O24X, O25, O26X, O28X, O29X, O30X, O31X, O32X, O33X, O34X, O35X, O36X, O40, O41X, O42X, O43X, O44X, O45X, O46X, O47X, O48, O60X, O61X, O62X, O63X, O64X, O65X, O66X, O67X, O68X, O69X, O70X, O71X, O72X, O73X, O74X, O75X, O80X, O81X, O82X, O83X, O84X, O85, O86X, O87X, O88X, O89X, O90X, O94, O95, O96X, O97X, O98X, O99X |
| Alcohol use | F1010, F10120, F10121, F10129, F1014, F10150, F10151, F10159, F10180, F10181, F10182, F10188, F1019, F1019, F1020, F1021, F10220, F10221, F10229, F10230, F10231, F10231, F10239, F1024, F10250, F10251, F10259, F10280, F10281, F10282, F10288, F1029, F10920, F10921, F10929, F1094, F10950, F10951, F10959, F10980, F10981, F10982, F10988, F1099, O9931X, R780 |
| Opioid use | F1110, F11120, F11121, F11122, F11129, F1115, F11150, F11151, F11159, F11181, F11188, F1119, F1120, F1121, F11220, F11221, F11222, F11229, F1123, F1124, F11250, F11251, F11259, F11282, F11282, F11288, F1129, F1190, F11920, F11921, F11922, F11929, F1193, F1194, F11950, F11951, F11959, F11981, R781F11982, F11988, F1199 |
| Cannabis use | F1210, F12120, F12121, F12122, F12129, F12150, F12151, F12159, F12180, F12188, F1219, F1220, F1221, F12220, F12221, F12222, F12229, F12250, F12251, F12259, F12280, F12288, F1229, F1290, F12920, F12921, F12922, F12929, F12950, F12951, F12959, F12980, F12988, F1299 |
| Sedative use | F1310, F13120, F13121, F13129, F1314, F13150, F13151, F13159, F13180, F13181, F13182, F13188, F1319, F1320, F1321, F13220, F13221, F13229, F13230, F13231, F13232, F13239, F1324, F13250, F12251, F13259, F13280, F13281, F13282, F13288, F1329, F1390, F13920, F13921, F13929, F13930, F13931, F13932, F13939, FF1394, F13950, F13951, F13959, F13980, F13981, F13982, F12988, F1399 |
| Cocaine use | F1410, F14120, F14121, F14122, F13129, F1414, F1450, F14151, F14159, F14180, F14181, F14182, F14188, F1419, F1420, F1421, F14220, F14221, F14222, F14229, F1423, F1424, F14250, F14251, F14259, F14280, F14281, F14282, F14288, F1429, F1490, F14921, F14921, F14922, F14929, F1494, F14950, F14951, F14959, F14980, F14981, F14982, F14988, F1499, R782 |
| Other stimulant use | F1510, F15120, F15121, F15122, F15129, F1514, F15150, F15151, F15159, F15180, F15181, F15182, F15188, F1519, F1520, F1521, F15220, F15221, F15222, F15229, F1523, F1524, F15250, f15251, F15259, F1523, F1524, F15250, F15251, F15251, F15259, F15280, F15281, F15282, F15288, F1529, F1590, F15920, F15921, F15922, F15929, F1593, F1594, F15950, F15951, F15959, F15980, F15981, F15982, F15988, F1599, |
| Hallucinogen use | F1610, F16120, F16120, F16121, F16122, F16129, F1614, F16150, F16151, F16159, F16180, F16183, F16188, F1619, F1620, F1621, F16220, F16229, F1624, F16250, F16251, F16259, F16280, F16283, F16288, F1629, F1690, F16920, F16921, F16929, F1694, F16950, F16951, F16959, F16959, F16980, F16983, F16988, F1699, R783 |
| Nicotine use | F17200, F17201, F17203, F17208, F17209, F17210, F17211, F17213, F17218, F17291, F17220, F17221, F17223, F17228, F17229, F17290, F17291, F17293, F17298, F17299 |
| Inhalant use | F1810, F18120, F18121, F18219, F1814, F18250, F18151, F18259, F1817, F18180, F18188, F1819, F1820, F1821, F19220, F18221, F18229, F1824, F18250, F18251, F18259, F1827, F18280, F18288, F1829, F1890, F18920, F18921, F18929, F1894, F18950, F18951, F18959, F18980, F18988, F1899 |
| Other psychoactive substance use | F1910, F19120, F19121, F19122, F19129, F1914, F19150, F19151, F19159, F19180, F19181, F19182, F19188, F1919, F1920, F1921, F19220, F19221, F19222, F19229, F19230, F19231, F19232, F19239, F1924, F19250, F19251, F19259, F19280, F19281, F19282, F19288, F1929, F1990, F19920, F19921, F19922, F19929, F19930, F19931, F19932, F19939, F1994, F19950, F19951, F19959, F19980, F19981, F19982, F19988, F1999, R785 |
| Other drug use | O9932X, R784 |
| Hypertensive disorders of pregnancy | O10X, O11X, O13X, O14X, O15X, O16X |
| Antepartum haemorrhage | O20X, O44.0X, O44.1X, O44.3X, O44.4X, O44.5X, O45X, O46X, O67X |
| Postpartum haemorrhage | O72X |
| Preterm birth | O601X, O4201X, Z3A1X, Z3A2X, Z3A30-6 |
| Fetal growth restriction | O365X, Z364, P050X, P051X, P052, P059 |
Appendix B
Comparison of complete cases with cases with missing data.
| Incomplete data | Complete data | p-value | SMD | |
|---|---|---|---|---|
| n | 168,608 | 3,238,558 | ||
| Year (%) | < 0.001 | 0.236 | ||
| 2016 | 45,038 (26.7) | 616,652 (19.0) | ||
| 2017 | 38,491 (22.8) | 622,272 (19.2) | ||
| 2018 | 28,393 (16.8) | 645,304 (19.9) | ||
| 2019 | 28,808 (17.1) | 690,112 (21.3) | ||
| 2020 | 27,878 (16.5) | 664,218 (20.5) | ||
| Median zip code income (%) | < 0.001 | 0.169 | ||
| Quartile 1 | 28,873 (20.9) | 906,333 (28.0) | ||
| Quartile 2 | 39,356 (28.5) | 818,772 (25.3) | ||
| Quartile 3 | 38,375 (27.8) | 797,292 (24.6) | ||
| Quartile 4 | 31,653 (22.9) | 716,161 (22.1) | ||
| Hospital size (%) | < 0.001 | 0.121 | ||
| Small | 40,552 (24.1) | 619,335 (19.1) | ||
| Medium | 49,323 (29.3) | 982,485 (30.3) | ||
| Large | 78,733 (46.7) | 1,636,738 (50.5) | ||
| Hospital type (%) | < 0.001 | 0.197 | ||
| Rural | 25,352 (15.0) | 282,516 (8.7) | ||
| Urban non-teaching | 33,168 (19.7) | 657,250 (20.3) | ||
| Urban teaching | 110,088 (65.3) | 2,298,792 (71.0) | ||
| Hospital region (%) | < 0.001 | 0.459 | ||
| Northeast | 16,440 (9.8) | 525,854 (16.2) | ||
| Midwest | 59,816 (35.5) | 657,305 (20.3) | ||
| South | 42,105 (25.0) | 1,297,345 (40.1) | ||
| West | 50,247 (29.8) | 758,054 (23.4) | ||
| Age in years (%) | < 0.001 | 0.029 | ||
| < 20 | 6672 (4.0) | 143,189 (4.4) | ||
| 20–29 | 83,163 (49.3) | 1,566,192 (48.4) | ||
| 30–39 | 73,503 (43.6) | 1,421,517 (43.9) | ||
| ≥ 40 | 5261 (3.1) | 107,660 (3.3) | ||
| Ethnicity (%) | < 0.001 | 0.199 | ||
| White | 14,917 (46.1) | 1,709,534 (52.8) | ||
| Black | 3987 (12.3) | 481,440 (14.9) | ||
| Hispanic | 9187 (28.4) | 674,850 (20.8) | ||
| Other | 4269 (13.2) | 372,734 (11.5) | ||
| Pay (%) | < 0.001 | 0.181 | ||
| Government | 59,525 (36.1) | 1,393,384 (43.0) | ||
| Private | 90,704 (55.1) | 1,677,402 (51.8) | ||
| Other | 14,458 (8.8) | 167,772 (5.2) | ||
| SUD (%) | 10,286 (6.1) | 194,930 (6.0) | 0.172 | 0.003 |
| APO (%) | 40,660 (24.1) | 832,726 (25.7) | < 0.001 | 0.037 |
| HDP (%) | 20,210 (12.0) | 446,509 (13.8) | < 0.001 | 0.054 |
| AH (%) | 2697 (1.6) | 53,857 (1.7) | 0.048 | 0.005 |
| PPH (%) | 4876 (2.9) | 98,049 (3.0) | 0.002 | 0.008 |
| PTB (%) | 17,252 (10.2) | 332,407 (10.3) | 0.676 | 0.001 |
| FGR (%) | 5993 (3.6) | 115,887 (3.6) | 0.610 | 0.001 |
SMD, standardized mean difference; SUD, substance use disorder; APO, adverse pregnancy outcome; HDP, hypertensive disorders of pregnancy; AH, antepartum haemorrhage; PPH, postpartum haemorrhage; PTB, preterm birth; FGR, fetal growth restriction.
Appendix C
Rates of adverse pregnancy outcomes in pregnant inpatients, Healthcare Cost and Utilization Project National Inpatient Sample, 2016–2020 (n = 3,238,558).
| SUD prevalence in all population | APO rates in specific SUD exposure | APO rates in specific non-SUD exposure | p-value | |
|---|---|---|---|---|
| n (%) | n (%) | n (%) | ||
| Alcohol | 4216/3,238,558 (0.13) | 1798/4216 (42.65) | 830,928/3,234,342 (25.69) | < 0.0001 |
| Tobacco | 141,221/3,238,558 (4.36) | 48,542/141,221 (34.37) | 784,184/3,097,337 (25.32) | < 0.0001 |
| Cannabis | 46,666/3,238,558 (1.44) | 17,962/46,666 (38.49) | 814,764/3,191,892 (25.53) | < 0.0001 |
| Cocaine | 5299/3,238,558 (0.16) | 2897/5299 (54.67) | 829,829/3,233,259 (25.67) | < 0.0001 |
| Opioids | 22,026/3,238,558 (0.68) | 9542/22,026 (43.32) | 823,184/3,216,532 (25.59) | < 0.0001 |
| Sedatives | 951/3,238,558 (0.03) | 441/951 (46.37) | 832,285/3,237,607 (25.71) | < 0.0001 |
| Hallucinogens | 201/3,238,558 (0.01) | 99/201 (49.25) | 832,627/3,238,357 (25.71) | < 0.0001 |
| Stimulants | 10,569/3,238,558 (0.33) | 5263/10,569 (49.80) | 827,463/3,227,989 (25.63) | < 0.0001 |
| Other | 79,936/3,238,558 (2.47) | 32,666/79,936 (40.87) | 800,060/3,158,622 (25.33) | < 0.0001 |
SUD, substance use disorder; APO, adverse pregnancy outcomes.
Appendix D
Descriptive statistics of demographic and hospital characteristics in matched sample.
| Without SUD | With SUD | p-value | SMD | |
|---|---|---|---|---|
| n | 389,860 | 194,930 | ||
| Year (%) | 0.977 | 0.002 | ||
| 2016 | 67,867 (17.4) | 33,903 (17.4) | ||
| 2017 | 71,588 (18.4) | 35,771 (18.4) | ||
| 2018 | 77,482 (19.9) | 38,714 (19.9) | ||
| 2019 | 88,276 (22.6) | 44,069 (22.6) | ||
| 2020 | 84,647 (21.7) | 42,473 (21.8) | ||
| Median zip code income (%) | 0.965 | 0.001 | ||
| Quartile 1 | 165,146 (42.4) | 82,698 (42.4) | ||
| Quartile 2 | 118,837 (30.5) | 59,326 (30.4) | ||
| Quartile 3 | 74,417 (19.1) | 37,160 (19.1) | ||
| Quartile 4 | 31,460 (8.1) | 15,746 (8.1) | ||
| Hospital size (%) | 0.954 | 0.001 | ||
| Small | 77,598 (19.9) | 38,824 (19.9) | ||
| Medium | 107,380 (27.5) | 53,616 (27.5) | ||
| Large | 204,882 (52.6) | 102,490 (52.6) | ||
| Hospital type (%) | 0.932 | 0.001 | ||
| Rural | 66,466 (17.0) | 33,309 (17.1) | ||
| Urban non-teaching | 69,569 (17.8) | 34,759 (17.8) | ||
| Urban teaching | 253,825 (65.1) | 126,862 (65.1) | ||
| Hospital region (%) | 0.993 | 0.001 | ||
| Northeast | 57,515 (14.8) | 28,813 (14.8) | ||
| Midwest | 115,175 (29.5) | 57,579 (29.5) | ||
| South | 152,847 (39.2) | 76,401 (39.2) | ||
| West | 64,323 (16.5) | 32,137 (16.5) | ||
| Age in years (%) | 0.685 | 0.003 | ||
| < 20 | 20,050 (5.1) | 10,083 (5.2) | ||
| 20–29 | 231,765 (59.4) | 115,594 (59.3) | ||
| 30–39 | 130,100 (33.4) | 65,228 (33.5) | ||
| ≥ 40 | 7945 (2.0) | 4025 (2.1) | ||
| Ethnicity (%) | 0.988 | 0.001 | ||
| White | 269,313 (69.1) | 134,617 (69.1) | ||
| Black | 73,122 (18.8) | 36,571 (18.8) | ||
| Hispanic | 28,666 (7.4) | 14,322 (7.3) | ||
| Other | 18,759 (4.8) | 9420 (4.8) | ||
| Pay (%) | 0.930 | 0.001 | ||
| Government | 299,248 (76.8) | 149,615 (76.8) | ||
| Private | 76,666 (19.7) | 38,305 (19.7) | ||
| Other | 13,946 (3.6) | 7010 (3.6) | ||
| APO (%) | 107,614 (27.6) | 69,491 (35.6) | < 0.001 | 0.174 |
| HDP (%) | 60,013 (15.4) | 32,074 (16.5) | < 0.001 | 0.029 |
| AH (%) | 6495 (1.7) | 5726 (2.9) | < 0.001 | 0.085 |
| PPH (%) | 11,733 (3.0) | 6002 (3.1) | 0.146 | 0.004 |
| PTB (%) | 42,897 (11.0) | 32,773 (16.8) | < 0.001 | 0.168 |
| FGR (%) | 14,925 (3.8) | 14,084 (7.2) | < 0.001 | 0.149 |
SMD, standardized mean difference; SUD, substance use disorder; APO, adverse pregnancy outcome; HDP, hypertensive disorders of pregnancy; AH, antepartum haemorrhage; PPH, postpartum haemorrhage; PTB, preterm birth; FGR, fetal growth restriction.
Appendix E
Regression model of hypertensive disorder of pregnancy.
| Characteristic | OR | 95 % CI | p-value |
|---|---|---|---|
| SUD | 1.08 | 1.07–1.10 | < 0.001 |
| Year (%) | < 0.001 | ||
| 2016 | — | — | |
| 2017 | 1.25 | 1.19–1.30 | |
| 2018 | 1.46 | 1.40–1.52 | |
| 2019 | 1.69 | 1.62–1.77 | |
| 2020 | 1.88 | 1.80–1.96 | |
| Median zip code income (%) | < 0.001 | ||
| Quartile 1 | — | — | |
| Quartile 2 | 0.92 | 0.90–0.94 | |
| Quartile 3 | 0.89 | 0.87–0.92 | |
| Quartile 4 | 0.83 | 0.80–0.86 | |
| Age in years (%) | < 0.001 | ||
| < 20 | — | — | |
| 20–29 | 0.93 | 0.90–0.97 | |
| 30–39 | 1.32 | 1.27–1.37 | |
| ≥ 40 | 2.19 | 2.08–2.31 | |
| Ethnicity (%) | < 0.001 | ||
| White | — | — | |
| Black | 1.44 | 1.40–1.47 | |
| Hispanic | 0.91 | 0.88–0.94 | |
| Other | 0.89 | 0.85–0.94 | |
| Pay (%) | < 0.001 | ||
| Government | — | — | |
| Private | 1.11 | 1.09–1.13 | |
| Other | 0.89 | 0.85–0.93 | |
| Hospital size (%) | < 0.001 | ||
| Small | — | — | |
| Medium | 1.10 | 1.06–1.14 | |
| Large | 1.33 | 1.29–1.38 | |
| Hospital type (%) | < 0.001 | ||
| Rural | — | — | |
| Urban non-teaching | 1.09 | 1.05–1.13 | |
| Urban teaching | 1.55 | 1.50–1.60 | |
| Hospital region (%) | < 0.001 | ||
| Northeast | — | — | |
| Midwest | 1.03 | 0.98–1.07 | |
| South | 1.17 | 1.13–1.22 | |
| West | 0.99 | 0.94–1.03 |
SUD, substance use disorder; OR, odds ratio; CI, confidence interval.
Appendix F
Regression model of antepartum hemorrhage.
| Characteristic | OR | 95 % CI | p-value |
|---|---|---|---|
| SUD | 1.79 | 1.73–1.85 | < 0.001 |
| Year (%) | < 0.001 | ||
| 2016 | — | — | |
| 2017 | 0.93 | 0.86–1.00 | |
| 2018 | 1.05 | 0.98–1.13 | |
| 2019 | 1.34 | 1.26–1.43 | |
| 2020 | 1.41 | 1.32–1.50 | |
| Median zip code income (%) | < 0.001 | ||
| Quartile 1 | — | — | |
| Quartile 2 | 0.97 | 0.92–1.01 | |
| Quartile 3 | 0.87 | 0.82–0.91 | |
| Quartile 4 | 0.87 | 0.81–0.94 | |
| Age in years (%) | < 0.001 | ||
| < 20 | — | — | |
| 20–29 | 1.26 | 1.14–1.40 | |
| 30–39 | 1.82 | 1.64–2.01 | |
| ≥ 40 | 2.52 | 2.19–2.89 | |
| Ethnicity (%) | < 0.001 | ||
| White | — | — | |
| Black | 1.21 | 1.15–1.27 | |
| Hispanic | 1.08 | 1.00–1.16 | |
| Other | 1.15 | 1.06–1.25 | |
| Pay (%) | < 0.001 | ||
| Government | — | — | |
| Private | 0.81 | 0.77–0.85 | |
| Other | 1.11 | 1.02–1.22 | |
| Hospital size (%) | < 0.001 | ||
| Small | — | — | |
| Medium | 1.13 | 1.06–1.20 | |
| Large | 1.31 | 1.24–1.38 | |
| Hospital type (%) | < 0.001 | ||
| Rural | — | — | |
| Urban non-teaching | 1.15 | 1.07–1.24 | |
| Urban teaching | 1.56 | 1.46–1.65 | |
| Hospital region (%) | 0.11 | ||
| Northeast | — | — | |
| Midwest | 1.01 | 0.95–1.08 | |
| South | 1.06 | 0.99–1.13 | |
| West | 1.06 | 0.99–1.14 |
SUD, substance use disorder; OR, odds ratio; CI, confidence interval.
Appendix G
Regression model of postpartum hemorrhage.
| Characteristic | OR | 95 % CI | p-value |
|---|---|---|---|
| SUD | 1.02 | 0.99–1.06 | 0.2 |
| Year (%) | < 0.001 | ||
| 2016 | — | — | |
| 2017 | 1.00 | 0.91–1.09 | |
| 2018 | 1.38 | 1.28–1.50 | |
| 2019 | 2.23 | 2.07–2.39 | |
| 2020 | 2.46 | 2.29–2.64 | |
| Median zip code income (%) | 0.066 | ||
| Quartile 1 | — | — | |
| Quartile 2 | 1.02 | 0.98–1.06 | |
| Quartile 3 | 1.01 | 0.97–1.06 | |
| Quartile 4 | 1.09 | 1.02–1.17 | |
| Age in years (%) | < 0.001 | ||
| < 20 | — | — | |
| 20–29 | 0.95 | 0.89–1.02 | |
| 30–39 | 1.09 | 1.01–1.17 | |
| ≥ 40 | 1.35 | 1.21–1.51 | |
| Ethnicity (%) | < 0.001 | ||
| White | — | — | |
| Black | 1.22 | 1.17–1.28 | |
| Hispanic | 1.35 | 1.28–1.44 | |
| Other | 1.38 | 1.29–1.47 | |
| Pay (%) | 0.4 | ||
| Government | — | — | |
| Private | 1.00 | 0.96–1.04 | |
| Other | 1.06 | 0.97–1.16 | |
| Hospital size (%) | < 0.001 | ||
| Small | — | — | |
| Medium | 1.00 | 0.94–1.06 | |
| Large | 1.23 | 1.16–1.30 | |
| Hospital type (%) | < 0.001 | ||
| Rural | — | — | |
| Urban non-teaching | 0.88 | 0.82–0.94 | |
| Urban teaching | 1.31 | 1.24–1.39 | |
| Hospital region (%) | < 0.001 | ||
| Northeast | — | — | |
| Midwest | 1.07 | 1.00–1.14 | |
| South | 0.78 | 0.73–0.83 | |
| West | 1.15 | 1.07–1.24 |
SUD, substance use disorder; OR, odds ratio; CI, confidence interval.
Appendix H
Regression model of preterm birth.
| Characteristic | OR | 95 % CI | p-value |
|---|---|---|---|
| SUD | 1.65 | 1.62–1.68 | < 0.001 |
| Year (%) | < 0.001 | ||
| 2016 | — | — | |
| 2017 | 1.00 | 0.96–1.04 | |
| 2018 | 1.07 | 1.02–1.11 | |
| 2019 | 1.13 | 1.09–1.17 | |
| 2020 | 1.13 | 1.08–1.17 | |
| Median zip code income (%) | < 0.001 | ||
| Quartile 1 | — | — | |
| Quartile 2 | 0.93 | 0.91–0.95 | |
| Quartile 3 | 0.87 | 0.84–0.89 | |
| Quartile 4 | 0.83 | 0.80–0.86 | |
| Age in years (%) | < 0.001 | ||
| < 20 | — | — | |
| 20–29 | 0.97 | 0.94–1.01 | |
| 30–39 | 1.21 | 1.17–1.26 | |
| ≥ 40 | 1.61 | 1.52–1.71 | |
| Ethnicity (%) | < 0.001 | ||
| White | — | — | |
| Black | 1.34 | 1.31–1.37 | |
| Hispanic | 1.06 | 1.02–1.09 | |
| Other | 1.10 | 1.05–1.14 | |
| Pay (%) | < 0.001 | ||
| Government | — | — | |
| Private | 0.84 | 0.82–0.86 | |
| Other | 1.06 | 1.02–1.11 | |
| Hospital size (%) | < 0.001 | ||
| Small | — | — | |
| Medium | 1.18 | 1.14–1.22 | |
| Large | 1.48 | 1.44–1.53 | |
| Hospital type (%) | < 0.001 | ||
| Rural | — | — | |
| Urban non-teaching | 1.26 | 1.21–1.31 | |
| Urban teaching | 1.84 | 1.77–1.90 | |
| Hospital region (%) | < 0.001 | ||
| Northeast | — | — | |
| Midwest | 1.10 | 1.06–1.15 | |
| South | 1.25 | 1.20–1.30 | |
| West | 1.06 | 1.02–1.11 |
SUD, substance use disorder; OR, odds ratio; CI, confidence interval.
Appendix I
Regression model of fetal growth restriction.
| Characteristic | OR | 95 % CI | p-value |
|---|---|---|---|
| SUD | 1.96 | 1.91–2.01 | < 0.001 |
| Year (%) | < 0.001 | ||
| 2016 | — | — | |
| 2017 | 1.06 | 1.00–1.12 | |
| 2018 | 1.13 | 1.06–1.20 | |
| 2019 | 1.13 | 1.06–1.20 | |
| 2020 | 1.17 | 1.10–1.24 | |
| Median zip code income (%) | < 0.001 | ||
| Quartile 1 | — | — | |
| Quartile 2 | 0.96 | 0.93–0.99 | |
| Quartile 3 | 0.94 | 0.90–0.97 | |
| Quartile 4 | 0.90 | 0.86–0.95 | |
| Age in years (%) | < 0.001 | ||
| < 20 | — | — | |
| 20–29 | 0.71 | 0.68–0.75 | |
| 30–39 | 0.63 | 0.60–0.66 | |
| ≥ 40 | 0.78 | 0.72–0.86 | |
| Ethnicity (%) | < 0.001 | ||
| White | — | — | |
| Black | 1.34 | 1.30–1.39 | |
| Hispanic | 0.78 | 0.73–0.82 | |
| Other | 1.04 | 0.98–1.11 | |
| Pay (%) | < 0.001 | ||
| Government | — | — | |
| Private | 0.87 | 0.84–0.90 | |
| Other | 0.78 | 0.72–0.84 | |
| Hospital size (%) | < 0.001 | ||
| Small | — | — | |
| Medium | 1.10 | 1.05–1.16 | |
| Large | 1.29 | 1.24–1.36 | |
| Hospital type (%) | < 0.001 | ||
| Rural | — | — | |
| Urban non-teaching | 1.14 | 1.08–1.21 | |
| Urban teaching | 1.38 | 1.31–1.45 | |
| Hospital region (%) | < 0.001 | ||
| Northeast | — | — | |
| Midwest | 1.04 | 0.99–1.10 | |
| South | 1.08 | 1.03–1.14 | |
| West | 0.89 | 0.84–0.94 |
SUD, substance use disorder; OR, odds ratio; CI, confidence interval.
References
- 1.Frazer Z., McConnell K., Jansson L.M. Treatment for substance use disorders in pregnant women: motivators and barriers. Drug Alcohol Depend. 2019;205 doi: 10.1016/j.drugalcdep.2019.107652. [DOI] [PubMed] [Google Scholar]
- 2.Latuskie K.A., Andrews N.C.Z., Motz M., et al. Reasons for substance use continuation and discontinuation during pregnancy: a qualitative study. Women Birth. 2019;32:e57–e64. doi: 10.1016/j.wombi.2018.04.001. [DOI] [PubMed] [Google Scholar]
- 3.Adams Z.M., Ginapp C.M., Price C.R., et al. “A good mother”: impact of motherhood identity on women's substance use and engagement in treatment across the lifespan. J Subst Abus Treat. 2021;130 doi: 10.1016/j.jsat.2021.108474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.American Psychiatric Association . Diagnostic and statistical manual of mental disorders. 5th ed. APA; Washington, DC: 2022. [Google Scholar]
- 5.Substance Abuse and Mental Health Services Administration . SAMHSA; Rockville, MD: 2020. National Survey of Drug Use and Health (NSDUH) releases. [Available at: 〈 https://www.samhsa.gov/data/release/2020-national-survey-drug-use-and-health-nsduh-releases〉] [Google Scholar]
- 6.Viteri O.A., Soto E.E., Bahado-Singh R.O., et al. Fetal anomalies and long-term effects associated with substance abuse in pregnancy: a literature review. Am J Perinatol. 2015;32:405–416. doi: 10.1055/s-0034-1393932. [DOI] [PubMed] [Google Scholar]
- 7.Lynch V., Clemans-Cope L., Howell E., Hill I. Diagnosis and treatment of substance use disorder among pregnant women in three state Medicaid programs from 2013 to 2016. J Subst Abus Treat. 2021;124 doi: 10.1016/j.jsat.2020.108265. [DOI] [PubMed] [Google Scholar]
- 8.Kotelchuck M., Cheng E.R., Belanoff C., et al. The prevalence and impact of substance use disorder and treatment on maternal obstetric experiences and birth outcomes among singleton deliveries in Massachusetts. Matern Child Health J. 2017;21:893–902. doi: 10.1007/s10995-016-2190-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Agency for Healthcare Research and Quality . AHRQ; Rockville, MD: 2022. Healthcare cost and utilization project (HCUP). HCUP NIS database documentation. [Available at: 〈 www.hcup-us.ahrq.gov/db/nation/nis/nisdbdocumentation.jsp〉] [Google Scholar]
- 10.Agency for Healthcare Research and Quality. Healthcare cost and utilization project (HCUP). HCUP NIS description of data elements. Rockville, MD: AHRQ; 2008. Available at 〈www.hcup-us.ahrq.gov/db/vars/h_region/nisnote.jsp〉.
- 11.Agency for Healthcare Research and Quality. Healthcare cost and utilization project (HCUP). HCUP NIS description of data elements. Rockville, MD: AHRQ; 2008. Available at: 〈www.hcup-us.ahrq.gov/db/vars/hosp_bedsize/nisnote.jsp〉.
- 12.Agency for Healthcare Research and Quality. Healthcare cost and utilization project (HCUP). HCUP NIS description of data elements. Rockville, MD: AHRQ; 2008. Available at: 〈www.hcup-us.ahrq.gov/db/vars/hosp_location/nisnote.jsp〉.
- 13.Agency for Healthcare Research and Quality. Healthcare cost and utilization project (HCUP). Producing national HCUP estimates – accessible version. Rockville, MD: AHRQ; 2018. Available at: 〈www.hcupus.ahrq.gov/tech_assist/nationalestimates/508_course/508course_2018.jsp〉.
- 14.R Core Team . R Foundation for Statistical Computing; Vienna: 2021. R: a language and environment for statistical computing. [Available at: 〈 https://www.R-project.org/〉] [Google Scholar]
- 15.Ho D.E., Imai K., King G., Stuart E.A. MatchIt: nonparametric preprocessing for parametric causal inference. J Stat Softw. 2011;42:1–28. [Google Scholar]
- 16.Lumley T. Analysis of complex survey samples. J Stat Softw. 2004;9:1–19. [Google Scholar]
- 17.Kennare R., Heard A., Chan A. Substance use during pregnancy: risk factors and obstetric and perinatal outcomes in South Australia. Aust N Z J Obstet Gynaecol. 2005;45:220–225. doi: 10.1111/j.1479-828X.2005.00379.x. [DOI] [PubMed] [Google Scholar]
- 18.Irner T.B., Teasdale T.W., Nielsen T., Vedal S., Olofsson M. Substance use during pregnancy and postnatal outcomes. J Addict Dis. 2012;31:19–28. doi: 10.1080/10550887.2011.642765. [DOI] [PubMed] [Google Scholar]
- 19.Oni H.T., Buultjens M., Mohamed A.L., Islam M.M. Neonatal outcomes of infants born to pregnant women with substance use disorders: a multilevel analysis of linked data. Subst Use Misuse. 2022;57:165. doi: 10.1080/10826084.2021.1958851. [Erratum in Subst Use Misuse 2022;57:165] [DOI] [PubMed] [Google Scholar]
- 20.Little M., Shah R., Vermeulen M.J., Gorman A., Dzendoletas D., Ray J.G. Adverse perinatal outcomes associated with homelessness and substance use in pregnancy. CMAJ. 2005;173:615–618. doi: 10.1503/cmaj.050406. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pereira C.M., Pacagnella R.C., Parpinelli M.A., et al. Drug use during pregnancy and its consequences: a nested case control study on severe maternal morbidity. Rev Bras Ginecol Obstet. 2018;40:518–526. doi: 10.1055/s-0038-1667291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Bailey B.A., Shah D.S., Boynewicz K.L., Justice N.A., Wood D.L. Impact of in utero opioid exposure on newborn outcomes: beyond neonatal opioid withdrawal syndrome. J Matern Fetal Neonatal Med. 2022;35:9383–9390. doi: 10.1080/14767058.2022.2035713. [DOI] [PubMed] [Google Scholar]
- 23.Janisse J.J., Bailey B.A., Ager J., Sokol R.J. Alcohol, tobacco, cocaine, and marijuana use: relative contributions to preterm delivery and fetal growth restriction. Subst Abus. 2014;35:60–67. doi: 10.1080/08897077.2013.804483. [DOI] [PubMed] [Google Scholar]
- 24.McDonald S.D., Vermeulen M.J., Ray J.G. Risk of fetal death associated with maternal drug dependence and placental abruption: a population-based study. J Obstet Gynaecol Can. 2007;29:556–559. doi: 10.1016/S1701-2163(16)32511-7. [DOI] [PubMed] [Google Scholar]
- 25.Handler A.S., Mason E.D., Rosenberg D.L., Davis F.G. The relationship between exposure during pregnancy to cigarette smoking and cocaine use and placenta previa. Am J Obstet Gynecol. 1994;170:884–889. doi: 10.1016/s0002-9378(94)70303-5. [DOI] [PubMed] [Google Scholar]
- 26.Addis A., Moretti M.E., Ahmed Syed F., Einarson T.R., Koren G. Fetal effects of cocaine: an updated meta-analysis. Reprod Toxicol. 2001;15:341–369. doi: 10.1016/s0890-6238(01)00136-8. [DOI] [PubMed] [Google Scholar]
- 27.Mbah A.K., Alio A.P., Fombo D.W., Bruder K., Dagne G., Salihu H.M. Association between cocaine abuse in pregnancy and placenta-associated syndromes using propensity score matching approach. Early Hum Dev. 2012;88:333–337. doi: 10.1016/j.earlhumdev.2011.09.005. [DOI] [PubMed] [Google Scholar]
- 28.Gouin K., Murphy K., Shah P.S. Knowledge synthesis group on determinants of low birth weight and preterm births. Effects of cocaine use during pregnancy on low birthweight and preterm birth: systematic review and metaanalyses. Am J Obstet Gynecol. 2011;204:340.e1–340.e12. doi: 10.1016/j.ajog.2010.11.013. [DOI] [PubMed] [Google Scholar]
- 29.Smid M.C., Metz T.D., Gordon A.J. Stimulant use in pregnancy: an under-recognized epidemic among pregnant women. Clin Obstet Gynecol. 2019;62:168–184. doi: 10.1097/GRF.0000000000000418. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Strauss A.A., Modaniou H.D., Bosu S.K. Neonatal manifestations of maternal phencyclidine (PCP) abuse. Pediatrics. 1981;68:550–552. [PubMed] [Google Scholar]
- 31.Shyken J.M., Babbar S., Babbar S., Forinash A. Benzodiazepines in pregnancy. Clin Obstet Gynecol. 2019;62:156–167. doi: 10.1097/GRF.0000000000000417. [DOI] [PubMed] [Google Scholar]
- 32.Piske M., Homayra F., Min J.E., et al. Opioid use disorder and perinatal outcomes. Pediatrics. 2021;148 doi: 10.1542/peds.2021-050279. [DOI] [PubMed] [Google Scholar]
- 33.Stanhope T.J., Gill L.A., Rose C. Chronic opioid use during pregnancy: maternal and fetal implications. Clin Perinatol. 2013;40:337–350. doi: 10.1016/j.clp.2013.05.015. [DOI] [PubMed] [Google Scholar]
- 34.Jarlenski M.P., Paul N.C., Krans E.E. Polysubstance use among pregnant women with opioid use disorder in the United States, 2007–2016. Obstet Gynecol. 2020;136:556–564. doi: 10.1097/AOG.0000000000003907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Wouldes T.A., Lester B.M. Stimulants: how big is the problem and what are the effects of prenatal exposure? Semin Fetal Neonat Med. 2019;24:155–160. doi: 10.1016/j.siny.2019.01.011. [DOI] [PubMed] [Google Scholar]
- 36.Jarlenski M., Krans E.E., Chen Q., et al. Substance use disorders and risk of severe maternal morbidity in the United States. Drug Alcohol Depend. 2020;216 doi: 10.1016/j.drugalcdep.2020.108236. [DOI] [PMC free article] [PubMed] [Google Scholar]