Abstract
Neonatal abstinence syndrome (NAS) is an automatically qualifying diagnosis for enrollment in early intervention (EI) in Massachusetts (MA) given the increased risk of developmental challenges among infants and toddlers exposed to illicit substances in utero. Rates of referral and enrollment among these infants and toddlers with and without a diagnosis of NAS have not been well established at the population level.
Objective:
This study aimed to assess EI referral and enrollment rates among substance-exposed newborns (SENs) with and without a diagnosis of NAS in MA and to compare rates to infants and toddlers without known substance exposure.
Methods:
We analyzed the 2013–2020 Pregnancy to Early Life Longitudinal dataset, which includes maternal-infant birth hospitalization records linked with EI data. SEN and NAS were identified using ICD-9/10 codes. We conducted descriptive statistics and bivariate analysis to compare EI referral and enrollment between (1) SENs with and without NAS diagnoses and (2) SENs and the unexposed population.
Results:
Among 6565 SENs, 85.3% (n = 5602) were referred to EI, and of those referred, 49.5% (n = 2775) enrolled. Compared with SENs without an NAS diagnosis, SENs with an NAS diagnosis were more likely to be referred to (87.6%, n = 4242 vs 79.0%, n = 1360) and enroll in (53.2%, n = 2255 vs 38.2%, n = 520) EI. Among infants and toddlers without identified substance exposure, referral rates were lower (34.6%), but enrollment rates were higher among those referred (56.7%).
Conclusion:
Substance-exposed newborns in MA had relatively high rates of EI referral but lower rates of enrollment compared with the general population, particularly among SENs without a diagnosis of NAS, suggesting a referral-to-enrollment gap.
Index terms: early intervention, substance-exposed infants, neonatal abstinence syndrome, developmental pediatrics, early intervention enrollment, early intervention referral
INTRODUCTION
The number of newborns with in utero substance exposure has grown significantly over the past 2 decades.1 Substance-exposed newborns (SENs), particularly those exposed to opioids, have an increased risk of cognitive, language, and motor delays,2,3 as well as long-term challenges related to school performance, behavioral problems, and executive functioning.4,5 The reasons for these associations are not limited to a single factor such as in utero substance exposure; rather, they are likely caused by a complex interplay between multiple factors,6 including infant sex, socioeconomic status, genetics/epigenetics, and co-occurring maternal mental health diagnoses.7 Families with infants and toddlers affected by substance use disorders therefore often require a variety of parent- and child-focused services.
Infants with neonatal abstinence syndrome (NAS), defined as symptoms because of physiological dependence on substance(s) secondary to in utero exposure (most commonly opioids), are a subgroup of SENs. In the past decade, there has been a focus on developing clinical guidelines specifically for the care of infants and toddlers with NAS because of adverse neurodevelopmental and morbidity outcomes. Guidelines have included both pharmacological and nonpharmacological interventions, such as environmental modifications, during the newborn period.8 However, recommendations have also extended to posthospitalization care including regular developmental surveillance and intervention. Guidance is often more variable for substance-exposed infants and toddlers without NAS.
The Individuals with Disabilities Education Act Part C Early Intervention (EI) is a federally funded program for children with or at risk for developmental delay.9 A diagnosis of NAS became an automatic qualifying diagnosis in 2009 for up to 1 year of EI services in Massachusetts (MA). Other parental substance misuse and mental health diagnoses are considered risk factors, but qualification is not automatic. Although EI home visiting is intended to address common logistical barriers to clinic-based treatment, barriers to referral and enrollment persist. Despite the broad inclusive eligibility criteria in MA compared with other states, single-site studies in MA demonstrated enrollment rates of <50% among infants and toddlers with a diagnosis of NAS.10,11 However, less is known about the referral and enrollment rates of SENs without a NAS diagnosis. Qualitative studies suggest that fear, stigma, and uncertainty of benefits are all potential reasons that parents with substance use disorders may choose to decline EI services,12 yet larger survey studies suggest that individual-level parent variables alone do not explain the success of service enrollment.13
The Pregnancy to Early Life Longitudinal (PELL) dataset managed by the MA Department of Public Health (MDPH) presents a unique opportunity to accurately assess EI engagement at the population level by linking birth certificate and birth hospitalization records with EI referral and enrollment data. Although these data have been used to assess factors associated with EI participation in the general population and among preterm infants and toddlers, they have not yet been leveraged to study engagement among SENs.14-21 The aims of this study were to (1) assess referral and enrollment rates among SENs in MA with and without a diagnosis of NAS and (2) compare enrollment rates among SENs to those without known substance exposure. Duration of enrollment was a secondary outcome variable to assess engagement among those enrolled.
METHODS
Data Source and Sample
Data were derived from the Massachusetts (MA) Pregnancy to Early Life Longitudinal (PELL) Data System. PELL is a longitudinally linked and relational data system with information on mothers and their children from delivery and birth through early childhood. The PELL data system uses deterministic and probabilistic methodologies to link vital statistics records (birth and death certificates), hospital utilization (including diagnostic codes), and public health programs’ participation data by using LinkPro software (InfoSoft, Inc, Winnipeg, MB, Canada). The PELL data system is linked to early intervention (EI) data on a yearly basis using a 9-step algorithm matching EI data and birth certificate records with variables including child’s first and last names, date of birth, mother’s date of birth, child gender, and/or residence zip code.19,22 EI records missing data on key linkage variables were excluded. The process of PELL linkage to EI data has been previously described.15 The maternal-child dyad linkage rate for PELL during the years 2013 to 2020 ranged between 96.1% and 99.3%. The PELL and EI databases linkage rate ranged between 90.9% and 92.8%. Records that were unlinked were excluded. Deidentified data were used for this analysis. This study was approved by both the Baystate Medical Center and MA Department of Public Health (MDPH) institutional review boards.
The study sample was selected from 352,375 in-state, MA resident births occurring between January 1, 2013 and December 31, 2017. These records were linked to EI program data for 154,853 individuals who engaged in some capacity with EI from January 1, 2013 to December 31, 2020, reflecting the period of birth to 3 years of age for our study cohort. Infants with a diagnosis of neonatal abstinence syndrome (NAS) were defined by infant International Classification of Diseases (ICD)-9/10 codes for NAS (779.5/P96.1). Consistent with prior studies, substance-exposed newborns (SENs) without a NAS diagnosis were defined by infant ICD-9/10 codes for unspecified intra uterine exposure (760.72/P04.49) during the birth hospitalization.11,23 Infants with tobacco exposure (760.79/P04.2), alcohol exposure (760.71/P04.3), and cocaine exposure (760.75/P04.41) were not specifically selected for study inclusion. Infants with chromosomal anomalies identified by ICD-9/10 codes 758/Q99 (5621 for all births, 97 for substance-exposed infants) were excluded as they had other high-risk diagnoses for EI referral in addition to in utero substance exposure. Infants with missing birthweight data were excluded (n = 22). This resulted in a sample of 6565 SENs, 4843 with a NAS diagnoses, 1722 without a NAS diagnosis, and 340,189 newborns without known substance exposure.
Study Sample Characteristics
To characterize the study population, additional variables derived from birth certificates included infant: race, sex, gestational age (GA), birth weight, feeding type (breast milk/formula), and if discharged home with birth mother. When GA or birthweight were missing from birth certificate data, hospitalization records were used to provide these data. Small for gestational age (SGA)/intrauterine growth restriction variables were created within PELL from the birth certificate or hospitalization record data. In addition, birth hospitalization records were used to determine the presence of complex chronic conditions identified using ICD-9/10 codes,24 level of newborn care, and length of hospital stay. Health care claims were also used to identify hospitalizations that occurred after the birth hospital stay.
Maternal variables derived from birth certificates included: age at time of infant birth, education level, marital status, mode of delivery, and presence of adequate prenatal care (calculated using Kotelchuck25’s Adequacy of Prenatal Care Utilization Index). Additionally, a binary variable was derived from the birth certificate data indicating whether a father was named on the birth certificate. Maternal variables derived from hospitalization records included: mental health diagnoses (by ICD-9/10 code) and insurance type.
Primary Outcomes
Early intervention referral and enrollment (from birth to age 3 years) were our primary outcome variables, with duration of enrollment as a secondary outcome variable of interest. The MDPH criteria for EI eligibility include: (1) established developmental delay, (2) diagnosis of a qualifying condition (e.g., NAS), (3) presence of 4 or more risk factors for developmental delay as defined by EI policies, or (4) clinical judgement of service needed.26 Referral by a medical professional is not required; thus, anyone including family members can make EI referrals. After referral, a formal evaluation is conducted to determine eligibility and recommend specific treatments as a standardized step in the enrollment process. Although all infants and toddlers with NAS automatically qualify for treatment, individual EI programs still make the ultimate determination regarding an infant’s eligibility for their specific program. Despite automatic eligibility with an NAS diagnosis, an EI program can find a family ineligible, for example if they do not receive documentation of NAS diagnosis or if they do not perceive a need for EI services. Eligible families agree to voluntary EI participation by completing an individualized family service plan (IFSP). Completion date of the IFSP was used to define enrollment. Although the MDPH collects data about EI eligibility, this variable was not included in our analytic dataset.
Analysis
Descriptive statistics (means for continuous variables and percentages for categorical variables) were used to summarize infant and maternal characteristics for the total cohort of SENs, and separately for those with and without a NAS diagnosis. Characteristics were compared between infants and toddlers with and without a NAS diagnosis using χ2 tests for categorical variables and Wilcoxon rank sum tests for continuous variables. The percent of infants and toddlers referred to and enrolled in EI were summarized and compared using χ2 tests. We then compared demographic maternal-child characteristics and EI referral/enrollment rates among infants and toddlers without any substance exposure to the substance-exposed newborn cohort.
RESULTS
We present the characteristics of the 6565 maternal-infant dyads with in-utero substance exposure in Table 1. Compared with maternal-infant dyads without a neonatal abstinence syndrome (NAS) diagnosis, infants with a diagnosis of NAS were more likely to be non-Hispanic White, small for gestational age (SGA), have a higher level of newborn care, and have a longer length of hospital stay. Infants with a diagnosis of NAS were less likely to be born preterm, be exclusively breastfed, or be discharged home with the mother. When comparing mothers of infants with and without a NAS diagnosis, mothers of infants with a diagnosis of NAS were significantly more likely to have a mental health disorder diagnosis and less likely to have received adequate prenatal care.
Table 1.
Infant and Maternal Birth Characteristics Among Substance-Exposed Newborns With and Without NAS Born January 1, 2013 Through December 31, 2017 in MA
| Category | Any Substance Exposure n = 6565 |
With NAS Diagnosis 4843 |
Without NAS Diagnosis 1722 |
Without Substance Exposure 340,189 |
p (Substance-Exposed Newborns with NAS vs without NAS) |
p (Any Substance Exposure vs without Substance exposure) |
|---|---|---|---|---|---|---|
| Child | N (Column Percent) or Mean (SD) |
|||||
| Race and ethnicitya | ||||||
| Hispanic | 755 (11.8) | 412 (8.7) | 343 (20.5) | 62,531 (18.8) | <0.001 | <0.001 |
| Non-Hispanic White | 5220 (81.6) | 4111 (87.0) | 1109 (66.3) | 203,129 (60.9) | ||
| Non-Hispanic Black | 343 (5.4) | 151 (3.2) | 192 (11.5) | 34,135 (10.2) | ||
| Asian/Pacific Islander | 43 (0.7) | 31 (0.7) | 12 (0.7) | 31,577 (9.5) | ||
| Native American/Aleutian | 35 (0.6) | 18 (0.4) | 17 (1.0) | 1996 (0.6) | ||
| Male gender | 3318 (50.5) | 2488 (51.4) | 830 (48.2) | 174,172 (51.2) | 0.024 | 0.490 |
| GA at birth (wk) | 38.5 (2.4) | 39 (2.3) | 38 (2.6) | 39.0 (1.9) | 0.265 | <0.001 |
| <32 | 63 (1.0) | 42 (0.9) | 21 (1.2) | 1708 (0.5) | 0.173 | <0.001 |
| 32–37 | 1753 (26.7) | 1273 (26.3) | 480 (27.9) | 54,668 (16.1) | ||
| >37 | 4749 (72.3) | 3528 (72.9) | 1221 (70.9) | 283,813 (83.4) | ||
| Birth weight (g) | 3033.8 (586.1) | 3041.0 (573.8) | 3013.7 (619.0) | 3355 (591.1) | 0.388 | <0.001 |
| <2000 | 245 (3.7) | 152 (3.1) | 93 (5.4) | 4980 (1.5) | <0.001 | <0.001 |
| ≥2000 and <3000 | 2775 (42.3) | 2074 (42.8) | 701 (40.7) | 73,034 (21.5) | ||
| ≥3000 | 3545 (54.0) | 2617 (54.0) | 928 (53.9) | 26,215 (77.1) | ||
| SGA/IUGR | 834 (12.7) | 648 (13.4) | 186 (10.8) | 12,371 (3.6) | 0.006 | <0.001 |
| CCC | ||||||
| Neuromuscular | 110 (1.7) | 92 (1.9) | 18 (1.0) | 1144 (0.3) | 0.018 | <0.001 |
| CVD | 215 (3.3) | 203 (4.2) | 12 (0.7) | 2098 (0.6) | <0.001 | <0.001 |
| Renal | 64 (1.0) | 47 (1.0) | 17 (1.0) | 3580 (1.1) | 0.952 | 0.542 |
| Congenital/genetic | 40 (0.6) | 27 (0.6) | 12 (0.7) | 3525 (1.0) | 0.587 | |
| Neonatal | 204 (3.1) | 150 (3.1) | 54 (3.1) | 11,965 (3.5) | 0.937 | <0.001 |
| Any of the above | 664 (10.1) | 556 (11.5) | 108 (6.3) | <0.001 | <0.001 | |
| Level of newborn nursery careb | ||||||
| Level I | 1650 (25.4) | 1002 (20.9) | 648 (38.1) | 57,139 (16.9) | <0.001 | <0.001 |
| Level II | 2391 (36.8) | 1840 (38.4) | 551 (32.4) | 129,566 (38.4) | ||
| Level III | 2449 (37.7) | 1945 (40.6) | 504 (29.6) | 150,507 (44.6) | ||
| Length of hospital stay, days | 14.1 (11.8) | 17.4 (11.7) | 4.9 (5.3) | 3.2 (3.9) | <0.001 | <0.001 |
| >5 days | 4435 (67.6) | 4075 (84.1) | 360 (20.9) | 16,664 (4.9) | <0.001 | <0.001 |
| Feeding typec | ||||||
| Breast milk only | 1722 (26.3) | 1087 (22.5) | 635 (36.9) | 226,310 (66.7) | <0.001 | <0.001 |
| Formula only | 3335 (50.8) | 2613 (54.0) | 722 (42.0) | 42,188 (12.4) | ||
| Other (all other combinations) | 1503 (22.9) | 1141 (23.6) | 362 (21.1) | 70,670 (20.8) | ||
| Discharged home with mother | 3217 (49.1) | 1993 (41.2) | 1224 (71.2) | 319,016 (93.8) | <0.001 | <0.001 |
| Child posthospitalization measures | ||||||
| Nonbirth hospitalization | 893 (13.6) | 698 (14.4) | 195 (11.3) | 30,719 (9.0) | 0.001 | <0.001 |
| Mother | ||||||
| Age at time of infant birth (yr)d | 28.6 (4.9) | 29.1 (4.8) | 27.3 (5.1) | 30.5 (5.6) | <0.001 | <0.001 |
| Education levele | ||||||
| <High school | 1249 (19.5) | 869 (18.5) | 380 (22.6) | 31,203 (9.4) | <0.001 | 0.002 |
| High school/General Educational Development (GED) | 2265 (35.4) | 1683 (35.8) | 582 (34.5) | 53,294 (16.1) | ||
| Some college | 2515 (39.3) | 1893 (40.2) | 622 (36.9) | 84,391 (25.4) | ||
| College+ | 364 (5.7) | 263 (5.6) | 101 (6.0) | 162,941 (49.1) | ||
| Marital status | ||||||
| Married | 1067 (16.3) | 805 (16.6) | 262 (15.2) | 228,650 (67.3) | 0.178 | <0.001 |
| Not married | 5494 (83.7) | 4036 (83.3) | 1458 (84.7) | 111,110 (32.7) | ||
| Mode of delivery | ||||||
| Vaginal | 4208 (64.1) | 3089 (63.8) | 1119 (65.0) | 228,648 (67.2) | 0.197 | <0.001 |
| C-section | 1374 (20.9) | 1039 (21.5) | 335 (19.5) | 63,372 (18.6) | ||
| Unknown | 983 (15.0) | 715 (14.8) | 268 (15.6) | 48,169 (14.2) | ||
| Mental health diagnoses | ||||||
| Depressive disorders | 1067 (16.3) | 832 (17.2) | 235 (13.6) | 14,527 (4.3) | <0.001 | <0.001 |
| Anxiety disorders | 1250 (19.0) | 988 (20.4) | 262 (15.2) | 15,709 (4.6) | <0.001 | <0.001 |
| Bipolar | 482 (7.3) | 376 (7.8) | 106 (6.2) | 2772 (0.8) | 0.028 | <0.001 |
| Substance use disorder | 4798 (73.1) | 3739 (77.2) | 1059 (61.5) | 7346 (2.2) | <0.001 | <0.001 |
| Opioid use disorder | 3090 (47.1) | 2747 (56.7) | 343 (19.9) | 928 (0.3) | <0.001 | <0.001 |
| Adequate prenatal caref | 3679 (58.4) | 2637 (56.8) | 1042 (62.7) | 270,919 (83.1) | <0.0001 | <0.001 |
| Insurance type | ||||||
| Private | 555 (8.5) | 381 (7.9) | 174 (10.1) | 171,260 (50.3) | 0.010 | <0.001 |
| Public | 5924 (90.2) | 4394 (90.7) | 1530 (88.9) | 148,967 (43.8) | ||
| Self-pay/other | 86 (1.31) | 68 (1.4) | 18 (1.1) | 19,962 (5.9) | ||
| Father named | 4663 (71.0) | 3460 (71.4) | 1203 (69.9) | 316,084 (92.9) | 0.214 | <0.001 |
One hundred sixty-nine missing.
Three thousand thirty-nine missing.
One thousand twenty-six missing.
Eleven missing.
Eight thousand five hundred thirty-two missing.
Fourteen thousand four hundred thirty-four missing.
CCC, complex chronic conditions; CVD, cardiovascular disease; GA, gestational age; IUGR, intrauterine growth restriction; MA, Massachusetts; NAS, neonatal abstinence syndrome; SGA, small for GA.
Overall, for the total cohort of substance-exposed newborns (SENs), 85.3% (n = 5602) were referred to early intervention (EI), and of those, 49.5% (n = 2775) enrolled (Table 2). Infants and toddlers with the NAS diagnosis were more likely to be referred to EI compared with SENs without NAS (87.6%, n = 4242 vs 79.0%, n = 1360) and more likely to enroll (53.2%, n = 2255 vs 38.2%, n = 520 of referred infants). The duration of enrollment was similar between the 2 groups.
Table 2.
EI Referral and Enrollment for Infants and Toddlers Born January 1, 2013 Through December 31, 2017 in MA
| Outcome | Any Substance Exposure n = 6565 |
With NAS Diagnosis 4843 |
Without NAS Diagnosis 1722 |
Without Substance Exposure 340,189 |
p (Substance-Exposed Newborns with NAS vs without NAS) |
p (Any Substance Exposure vs without Substance exposure) |
|---|---|---|---|---|---|---|
| N (Column Percent) or Mean (SD) |
||||||
| Referred | 5602 (85.3) | 4242 (87.6) | 1360 (79.0) | 117,603 (34.6) | <0.001 | <0.001 |
| Enrolled (among referred) | 2775 (49.5) | 2255 (53.2) | 520 (38.2) | 66,688 (56.7) | <0.001 | <0.001 |
| Duration of enrollment (d) | ||||||
| <183 | 1158 (42.3) | 939 (42.2) | 219 (42.9) | 27,751 (42.5) | 0.916 | 0.004 |
| 183–365 | 1092 (39.9) | 888 (39.9) | 204 (39.9) | 27,382 (41.9) | ||
| >365 | 488 (17.8) | 400 (18.0) | 88 (17.2) | 10,196 (15.6) |
EI, early intervention; MA, Massachusetts; NAS, neonatal abstinence syndrome.
Referral rates were significantly lower among infants and toddlers without any identified substance exposure (34.6%, n = 117,603); however, those referred had higher rates of enrollment (56.7%, n = 66,688) compared with referred infants and toddlers with any substance exposure (49.5%). The duration of enrollment was more likely to be >365 days for infants and toddlers with substance exposure compared to without (17.8% vs 15.6%).
DISCUSSION
In a large cohort of infants and toddlers, we have demonstrated that substance-exposed newborns (SENs) in Massachusetts (MA) had relatively high rates of referral to early intervention (EI), but relatively low rates of enrollment, particularly among infants and toddlers without a NAS diagnosis. Awareness of the factors for this referral-to-enrollment gap may assist public health agencies in focusing resources to promote engagement with families affected by substance use disorders as well as those most in need of developmental support.
Importantly, rates of enrollment were significantly lower during the study period for referred infants and toddlers with prenatal substance-exposure compared with those who were not substance exposed. The difference was greatest for substance-exposed infants without the neonatal abstinence syndrome (NAS) diagnosis (38% vs 56% with NAS). Previous Pregnancy to Early Life Longitudinal (PELL) studies of infants with similar risk for developmental delay, such as late preterm birth, show enrollment rates of approximately 24%,19 which is slightly lower than the enrollment rate observed among SENs without the automatically qualifying NAS diagnosis. Although EI is a national, federally funded program, substantial state-level variation exists with respect to who is eligible to receive the services. The referral rate for infants and toddlers without substance exposure in MA is high at ~34%. The referral and enrollment rates for all SENs were also much higher compared to states with more restrictive eligibility criteria (referral rates ranging from 32% to 40% in Kentucky and Colorado, and enrollment at 9% in Kentucky).27,28 The rate of referral and enrollment for infants and toddlers with NAS at the population level during this study period fell between that observed in other single hospital studies in MA with differing timeframes.10,11
Over half of referred substance exposed infants and toddlers without an NAS diagnosis were never enrolled in EI services. It is possible that many of these infants and toddlers had normal developmental trajectories and were not perceived by providers and/or caregivers to require additional EI support. Alternatively, the lack of an NAS diagnosis and corresponding automatic qualifying criteria may create a barrier to enrollment for this subset of SENs. NAS symptomatology is not directly related to severity of maternal substance use disorder or other developmental risk factors; however, clinicians may be more likely to make a NAS diagnosis in infants with more severe symptoms requiring pharmacologic treatment. This diagnostic reasoning is notable in the context of nationwide trends from 2016 to 2020, during which NAS diagnosis rates decreased while rates of prenatal substance exposure rose, a trend that was also seen in MA.29
With maternal substance use and mental health disorders rising as leading causes of mortality and morbidity postpartum,30 along with their known effects on long-term child development3,31 and associations with other psychosocial risk factors,6 preventative services are critical even in the absence of a NAS diagnosis and identified developmental delays at birth. Still, there is limited ability to predict which infants and toddlers will ultimately experience developmental delay. Furthermore, families may not perceive benefits of enrollment if they experience stigma and fear,12 or if their unique needs are not addressed by therapies designed for the more general population.32 Strategies to encourage referred families to enroll and test services may be needed to close the referral-to-enrollment gap.
Strengths of this study include the use of a unique data linkage between maternal-infant dyads and EI services through the PELL dataset, allowing analysis of EI engagement at the population level. Dependence on ICD-9/10 for disease categorization is 1 limitation, which may result in misclassification, inconsistency, or missed individuals. For example, our use of broad codes to define the cohort of SENs without NAS may have missed infants with nonopioid drug exposure. Furthermore, we were not able to assess differences in eligibility among SENs without NAS and similar at-risk populations without any substance exposure or identify infants and toddlers who received services outside of EI as a potential cause for lower rates of enrollment. Additional contextual information such as referral source similarly was not available. We further acknowledge that our study period included the COVID-19 pandemic, during which referral and enrollment patterns may have been affected. Finally, the conclusions we draw from our analysis of MA data are not representative of nationwide trends, although they do help study EI engagement in the setting of more inclusive eligibility criteria.
Overall, SENs without a NAS diagnosis were less likely to be referred and, among those referred, less likely to enroll in EI than infants who received an NAS diagnosis. With the increasing clinical focus on infants with NAS within the cohort of substance-exposed infants, it is possible that these differences can be partly attributed to the implementation of more robust prehospitalization and posthospitalization clinical guidelines for identification and care of infants and toddlers with NAS, which likely affects participation in EI. SENs and toddlers without a diagnosis of NAS may subsequently receive less standardized, guideline-directed care that ultimately results in decreased enrollment despite referral. Greater understanding of the need for targeted interventions for this population may be required to inform such guidelines. When comparing all SENs to newborns without documented substance exposure, referral rates were substantially higher for SENs, but enrollment rates were significantly lower among those referred. These results indicate a need for further exploration of predictors, eligibility, and strategies to address the referral-to-enrollment gap in MA. Future studies that assess the developmental impact of EI on this population could inform programs nationally.
Acknowledgments
Supported by the National Institute on Drug Abuse (award no. K23DA050731). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Funded by the National Institutes of Health (NIH).
Footnotes
Disclosure: The authors declare no conflict of interest.
REFERENCES
- 1.Hirai AH, Ko JY, Owens PL, et al. Neonatal abstinence syndrome and maternal opioid-related diagnoses in the US, 2010–2017. JAMA. 2021;325:146–155. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Beckwith AM, Burke SA. Identification of early developmental deficits in infants with prenatal heroin, methadone, and other opioid exposure. Clin Pediatr (Phila). 2015;54:328–335. [DOI] [PubMed] [Google Scholar]
- 3.Merhar SL, McAllister JM, Wedig-Stevie KE, et al. Retrospective review of neurodevelopmental outcomes in infants treated for neonatal abstinence syndrome. J Perinatol. 2018;38:587–592. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.MacMillan KDL. Neonatal abstinence syndrome: review of epidemiology, care models, and current understanding of outcomes. Clin Perinatol. 2019;46:817–832. [DOI] [PubMed] [Google Scholar]
- 5.Rees P, Stilwell PA, Bolton C, et al. Childhood health and educational outcomes after neonatal abstinence syndrome: a systematic review and meta-analysis. J Pediatr. 2020;226:149–156.e16. [DOI] [PubMed] [Google Scholar]
- 6.Larson JJ, Graham DL, Singer LT, et al. Cognitive and behavioral impact on children exposed to opioids during pregnancy. Pediatrics. 2019;144:e20190514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Corr TE, Schaefer EW, Hollenbeak CS, et al. One-year postpartum mental health outcomes of mothers of infants with neonatal abstinence syndrome. Matern Child Health J. 2020;24:283–290. [DOI] [PubMed] [Google Scholar]
- 8.Mascarenhas M, Wachman EM, Chandra I, et al. Advances in the care of infants with prenatal opioid exposure and neonatal opioid withdrawal syndrome. Pediatrics. 2024;153:e2023062871. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Individuals with Disabilities Education Improvement Act. Pub. L; 2004. p. 108–446. [Google Scholar]
- 10.Mascarenhas M, Peacock-Chambers E, Fraiman YS, et al. Engagement with recommended developmental follow-up and supports among infants with intrauterine opioid exposure. J Dev Behav Pediatr. 2025;46(4): e375–e382. doi: 10.1097/DBP.0000000000001372. 40601929. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Peacock-Chambers E, Leyenaar JK, Foss S, et al. Early intervention referral and enrollment among infants with neonatal abstinence syndrome. J Dev Behav Pediatr. 2019;40:441–450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Peacock-Chambers E, Paterno MT, Kiely D, et al. Engagement in perinatal outpatient services among women in recovery from opioid use disorders. Subst Abus. 2021;42:1022–1029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Harrison AJ, Bhimani SK, Allen FE, et al. Identifying factors related to successful enrollment in early intervention and early childhood special education. Early Child Education J. 2023;52:2039–2051–2051. 10.1007/s10643-023-01535-2. [DOI] [Google Scholar]
- 14.Barfield WD, Clements KM, Lee KG, et al. Using linked data to assess patterns of early intervention (EI) referral among very low birth weight infants. Matern Child Health J. 2008;12:24–33. [DOI] [PubMed] [Google Scholar]
- 15.Clements KM, Barfield WD, Kotelchuck M, et al. Birth characteristics associated with early intervention referral, evaluation for eligibility, and program eligibility in the first year of life. Matern Child Health J. 2006;10:433–441. [DOI] [PubMed] [Google Scholar]
- 16.Clements KM, Barfield WD, Ayadi MF, et al. Preterm birth-associated cost of early intervention services: an analysis by gestational age. Pediatrics. 2007;119:e866–e874. [DOI] [PubMed] [Google Scholar]
- 17.Clements KM, Barfield WD, Kotelchuck M, et al. Maternal socioeconomic and race/ethnic characteristics associated with early intervention participation. Matern Child Health J. 2008;12:708–717. [DOI] [PubMed] [Google Scholar]
- 18.Derrington TM. Development of the drug-exposed infant identification algorithm (DEIIA) and its application to measuring part C early intervention referral and eligibility in [STATE], 1998-2005. Matern Child Health J. 2013;17:1567–1575. [DOI] [PubMed] [Google Scholar]
- 19.Shapiro-Mendoza C, Kotelchuck M, Barfield W, et al. Enrollment in early intervention programs among infants born late preterm, early term, and term. Pediatrics. 2013;132:e61–e69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Diop H, Gopal D, Cabral H, et al. Assisted reproductive technology and early intervention program enrollment. Pediatrics. 2016;137:e20152007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Fauth RC, Kotake C, Manning SE, et al. Timeliness of early identification and referral of infants with social and environmental risks. Prev Sci. 2023;24:126–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Manning SE, Davin CA, Barfield WD, et al. Early diagnoses of autism spectrum disorders in [STATE] birth cohorts, 2001-2005. Pediatrics. 2011;127:1043–1051. [DOI] [PubMed] [Google Scholar]
- 23.Diop H, Cui X, Nielsen T, et al. Length of stay among infants with neonatal abstinence syndrome and risk of hospital readmission. Matern Child Health J. 2022;26:2020–2029. [DOI] [PubMed] [Google Scholar]
- 24.Feudtner C, Feinstein JA, Zhong W, et al. Pediatric complex chronic conditions classification system version 2: updated for ICD-10 and complex medical technology dependence and transplantation. BMC Pediatr. 2014;14:199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Kotelchuck M. The adequacy of prenatal care utilization index: its US distribution and association with low birthweight. Am J Public Health. 1994;84:1486–1489. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.[STATE] Early Intervention Division. Early Intervention Policies. Commonwealth of [STATE]. 2024. https://www.mass.gov/lists/early-intervention-policies. Accessed April 17, 2026. [Google Scholar]
- 27.Ferrell E, Marshall J, Bada H, et al. Children diagnosed with NAS in Kentucky and their involvement in early intervention. J Early Intervention. 2023. doi: 10.1177/10538151231204794. [DOI] [Google Scholar]
- 28.Bourque SL, Weikel BW, Hwang SS. Prevalence and predictors of early intervention referral among substance-exposed newborns. Hosp Pediatr. 2023;13:945–953. [DOI] [PubMed] [Google Scholar]
- 29.West KD, Ali MM, Blanco M, et al. Prenatal substance exposure and neonatal abstinence syndrome: state estimates from the 2016-2020 transformed Medicaid statistical information system. Matern Child Health J. 2023;27:14–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Trost S, Busacker A, Chandra G, et al. Pregnancy-Related Deaths: Data from Maternal Mortality Review Committees in 38 U.S. States; 2020:2024. [Google Scholar]
- 31.Ornoy A, Michailevskaya V, Lukashov I, et al. The developmental outcome of children born to heroin-dependent mothers, raised at home or adopted. Child Abuse Negl. 1996;20:385–396. [DOI] [PubMed] [Google Scholar]
- 32.Lowell AF, Suchman NE, Byatt N, et al. Parental substance use and home visiting programs: implementation considerations for relationship-based treatment. Infant Ment Health J. 2023;44:166–183. [DOI] [PMC free article] [PubMed] [Google Scholar]