Neonatal Abstinence Syndrome Practices in the United States

Abstract

OBJECTIVE

To survey current practices for the treatment of neonatal abstinence syndrome (NAS) among institutions in the United States to identify changes in national practice over time.

METHODS

Previous NAS management reports were referenced in the development of our 26-question electronic survey, which was distributed in the fall of 2019 to pediatric practitioners of 2 national clinical pharmacy organizations via email list servers. Not all questions required a response and responses from incomplete surveys were included. Institution demographics and NAS management strategies, including location of care, observation period, and inpatient and outpatient pharmacotherapy, were queried.

RESULTS

Seventy respondents representing institutions from all US geographic regions participated in the survey The most commonly reported inpatient observation durations were 3 (18 of 61, 29%) and 5 (22 of 61, 36%) days. Respondents indicated that neonates were typically transferred to the NICU if pharmacologic management was required (38 of 56, 68%). According to participants, first-line agents used for NAS management were morphine (45 of 56, 80%), methadone (5 of 56, 9%), clonidine (2 of 56, 4%), and buprenorphine (2 of 56, 4%). Among respondents, only 20% (11 of 56) reported that infants may be discharged home on pharmacotherapy, including morphine (n = 6), phenobarbital (n = 3), clonidine (n = 1), and methadone (n = 1).

CONCLUSIONS

Opioids are the most commonly used first-line agents for NAS management in the United States. The primary site of NAS management is the inpatient setting, as only 20% of institutions report discharging patients on pharmacotherapy.

Introduction

In the United States, the rise in opioid abuse has coincided with an increase in the incidence of neonatal abstinence syndrome (NAS). Past literature expresses inconsistent evidence for NAS treatment best practices, including optimal first-line inpatient pharmacologic treatment, adjunct treatment, use of outpatient pharmacotherapy, minimum in-hospital observation period, and location of care. To our knowledge, the most recent national survey of NAS treatment practices in the United States was published in 2006.¹ Historically, opioids have been used as first-line pharmacologic agents for inpatient treatment of NAS.^1,2 Previous surveys of inpatient NICU practices also indicated that phenobarbital or clonidine are often used in addition to an opioid for refractory NAS or in other institutionally prespecified situations.^1,3 The American Academy of Pediatrics (AAP) does not address outpatient NAS management and recent reports have indicated that institutions seldom or never discharge infants on pharmacotherapy.^2,4 No studies in the United States have assessed typical duration of a minimum observation period to our understanding. The most frequent locations of care for infants in the inpatient setting include the NICU or pediatric unit; however, practices appear to be inconsistent between institutions according to previous literature.^3,5 Our study aimed to survey current practices for the treatment of NAS to identify changes in national practice over time and facilitate standardization of institutional NAS treatment protocol.

Materials and Methods

An electronic survey consisting of 26 questions was developed to assess NAS treatment protocols, assessment tool used, minimum observation period, location of care, inpatient and outpatient pharmacologic management, non-pharmacologic therapies, respondent job title, and institution demographics (Supplemental Table). Previous NAS management surveys were referenced for preparing survey questions.^1,3 A pilot survey was not conducted. Survey distribution occurred as follows: An introduction email providing the recipient with the survey objective, researcher contact information, and link to the survey was sent via the Pediatric Pharmacy Association (PPA) email list on October 24, 2019. A reminder email was sent through the PPA email list on November 12, 2019. Additionally, the same introduction email was also sent through the American College of Clinical Pharmacy (ACCP) Pediatric Practice and Research Network (PRN) email list on December 5, 2019. No reminder email was sent to the ACCP PRN email list. These 2 organizations were selected to reach the greatest number of pediatric specialists.

Participation was voluntary and anonymous. Consent to participate was obtained in the first question of the survey requiring a yes or no response. If the respondent agreed to participate, the next question asked if their institution manages patients with NAS. A yes response allowed the participant to complete the rest of the survey, whereas a no response terminated participation in the remaining survey questions. All subscribers to the PPA and ACCP PRN email lists, including those beyond the neonatal specialty, were eligible to take part in the survey if they agreed to participate and reported that their institution manages patients with NAS.

Survey questions were multiple choice, except for the final fill-in-the-blank question requesting the name of the respondent's institution. Questions were either or no, single response, multiple response, or multiple choice with an other fill-in-the-blank option. Sixteen questions were forced response and inquired about the presence of a written protocol, assessment tool used, location of care, first-line pharmacologic treatment, respondent job title, and institution demographics. The survey collected the following institution demographic information: geographic region of United States where the institution is located, category of the institution (i.e., non-government not-for-profit community hospital, non-government not-for-profit teaching hospital/academic health center), pediatric services offered at the institution, number of pediatric beds, NICU level, and description of NICU configuration. Institution name was requested, although not required, to combine information if received more than once from the same institution. Select questions were automatically bypassed if not relevant to the participant as based on a previous answer. A survey was considered complete if the respondent finished the survey form with all forced-response questions answered. When a respondent agreed to participate in the survey yet terminated their participation prior to completing all forced-response questions, the survey was considered incomplete. Responses from incomplete surveys were recorded and included in the final data. Data were collected by using the Qualtrics platform (Provo, UT). Descriptive statistics were calculated for each of the survey questions.

Results

Institution and Respondent Demographics. Our survey had 70 respondents. Fifty-five (79%) respondents completed all 26 survey questions; 14 (20%) were partially completed, but contained usable data; and 1 was omitted because of institution duplication. Of the 55 respondents who answered the question of job title, most identified themselves as pharmacists (96%), while 2 participants reported that they were neonatologists.

Fifty-five respondents provided information about their respective institutions. Most individuals practiced at a children's hospital that was located with an adult facility (53%), while 18% practiced at a freestanding children's hospital. A total of 29% were located in a facility with minimal pediatric services. The number of pediatric beds available at the 55 institutions represented in the survey varied from <50 (18%), 50 to 100 (27%), 101 to 200 (42%), and >200 (13%). All respondents noted that their institution had a NICU: Level II (5%), Level III (56%), and Level IV (38%). Fifty-three respondents reported their geographic region as follows: 19% West (e.g., AZ, OR); 30% South (e.g., VA, AL, OK); 15% Northeast (e.g., NY, VT), and 36% Midwest (OH, SD).

NAS Assessment Tool. Sixty-four respondents answered the question related to their institution use of NAS assessment tool. Most (94%) reported using a version of the Finnegan Neonatal Abstinence Scoring Tool. This encompassed the original, modified, or short form, either alone or in conjunction with another assessment tool.

Twenty percent reported that their institution used more than 1 assessment tool. The most frequently used standalone assessment tool was the Modified Finnegan Neonatal Abstinence Scoring Tool (52%). The Eat, Sleep, Console (ESC) method was used by 20% of respondents, either alone (n = 3) or in combination with another assessment tool (n = 10). Assessment tools reported as being used in combination with the ESC method included the Finnegan Neonatal Abstinence Scoring Tool (n = 2), the Modified Finnegan Neonatal Abstinence Scoring Tool (n = 7), and the Neonatal Withdrawal Inventory (n = 1).

Inpatient Pharmacologic Treatment. Survey responses for institutions' primary first-line and adjunct pharmacotherapy for inpatient treatment are described in Table 1. Eighty percent of institutions reported using morphine as their primary first-line pharmacologic treatment. Other first-line pharmacotherapy included methadone, buprenorphine, clonidine, and other. The other write-in response suggested that the methadone was used in the NICU and morphine was used in the nursery.

Table 1.

Survey Responses Regarding Inpatient and Outpatient Pharmacotherapy

Parameter	n (%)
Inpatient first-line treatment (n = 56)
Morphine	45 (80)
Methadone	5 (9)
Buprenorphine	2 (4)
Clonidine	2 (4)
Other	2 (4)
Inpatient adjunct treatment (n = 56)
Clonidine	33 (59)
Phenobarbital	17 (30)
Morphine	2 (4)
Phenobarbital or clonidine	1 (2)
Does not use an adjunct	3 (5)
Discharging on outpatient treatment (n = 56)
Most of the time	1 (2)
About half the time	0 (0)
Sometimes	10 (18)
Never	45 (80)
Outpatient treatment (n = 11)
Morphine	6 (54)
Phenobarbital	3 (27)
Methadone	1 (9)
Clonidine	1 (9)

Adjunct pharmacologic treatment was more variable than first-line pharmacologic treatment and was noted by 56 respondents. When asked which adjunct pharmacologic agent was primarily used at their institution, most respondents noted clonidine, followed by phenobarbital. Five percent of respondents reported that their institution did not use an adjunct pharmacologic treatment.

Outpatient Pharmacologic Management. Frequency of discharging on outpatient pharmacotherapy is summarized in Table 1. Most respondents (80%) stated patients from their institution were never discharged on outpatient pharmacotherapy (Table 1); however, 20% stated patients were discharged on outpatient pharmacotherapy most of the time (2%) or sometimes (18%). Of the respondents who answered most of the time or sometimes, outpatient pharmacotherapy included morphine, phenobarbital, methadone, and clonidine. First-line inpatient pharmacologic treatment was morphine for all but one of the institutions who discharged on outpatient pharmacotherapy.

Non-pharmacologic Management. The survey question addressing non-pharmacologic NAS management strategies asked respondents to select all non-pharmacologic NAS therapies used at their institution. Response options included 12 prespecified non-pharmacologic therapies and a fill-in-the-blank option. Fifty-six individuals responded to this question. Most participants (82%) reported their institution used 6 or more different non-pharmacologic therapies for NAS management. The most used non-pharmacologic strategies included soothing techniques (98%), low stimulation environment (95%), skin-to-skin contact (89%), clustering of cares (84%), parental presence at bedside (80%), and breastfeeding (77%). Two fill-in-the-blank responses were recorded. The first respondent stated that their institution used aromatherapy and the second indicated that non-pharmacologic strategies vary depending on patient tolerance (i.e., massage, music, swings in particular).

Minimum Observation Period. Minimum observation period duration responses are shown in the Figure. Significant variation was reported for minimum duration of observation. Duration of observation ranged from 1 day to 7 days. The most common observation period was 5 days (43%), followed by 3 days (33%).

Figure.

Survey responses regarding minimum observation period duration (n = 51).^*

Location of Care. As shown in Table 2, primary location of care was assessed for 3 separate NAS management situations: observation, non-pharmacologic management only, and pharmacologic management. Most participants (79%) reported that their institutions roomed neonates with their mothers prior to maternal discharge. Rooming with the mother occurred frequently during observation (69%) and when the infant received non-pharmacologic management only (48%). However, rooming with the mother rarely took place once the infant required pharmacologic treatment (5%). Following maternal discharge or when the infant did not room with the mother, the newborn nursery was the primary location of care for neonates during observation (54%) and while receiving non-pharmacologic management only (48%). Once a neonate required pharmacologic treatment, the neonate was typically transferred to the NICU (68%). A portion of respondents also indicated that neonates requiring pharmacologic treatment at their institution might be treated in the NICU or the pediatric unit, depending on the infant's condition (30%). The only situation in which the pediatric unit alone was reported as the primary site of treatment was during non-pharmacologic management only (4%).

Table 2.

Survey Responses Regarding Location of Care

Parameter	n (%)
Room with the mother (n = 56)
Observation	39 (69)
Non-pharmacologic management	27 (48)
Pharmacologic treatment	3 (5)
Never room with the mother	12 (21)
Location of care during observation (n = 56)
Newborn nursery only	30 (54)
NICU only	18 (32)
Pediatric unit only	0 (0)
NICU or pediatric unit	8 (14)
Location of care during non-pharmacologic management only (n = 56)
Newborn nursery only	27 (48)
NICU only	19 (34)
Pediatric unit only	2 (4)
NICU or pediatric unit	8 (14)
Location of care during pharmacologic treatment (n = 56)
Newborn nursery only	1 (2)
NICU only	38 (68)
Pediatric unit only	0 (0)
NICU or pediatric unit	17 (30)

Discussion

Between 2004 to 2014, the number of infants born with NAS increased from 1.5 to 8 per 1000 hospital births.⁶ With NAS incidence on the rise, understanding current trends in national NAS management practices can help facilitate fewer interventions, improve infant outcomes, reduce health care costs, and direct future research on NAS management practices. In 2012, the AAP released guidance for clinicians managing neonatal drug withdrawal, which included information on clinical presentation, options for therapeutic treatment, and guidance for inpatient management of neonates who require weaning from analgesics or sedatives.² The AAP encourages institutions that care for infants with NAS to develop and adhere to standardized protocols for the screening, evaluation, and treatment of infants exhibiting signs and symptoms of drug withdrawal.²

Our survey revealed several variations in current practices for NAS treatment across the United States. The largest differences in current practice were seen when respondents were asked how long their institutions observed a neonate at risk for NAS and which pharmacologic agents may be used for management in the outpatient setting. The minimum in-hospital observation period for infants suspected to be at risk of developing NAS is variable and partially dependent on the half-life of the maternal drug, as well as social factors.² Among surveyed institutions in 2017 in Canada, 77.4% had a minimum observation period for asymptomatic infants with durations ranging from 24 to 168 hours; however, optimal observation period duration had not been established.³ To assess duration of the minimum observation period, participants of our survey were asked: “At your institution, how many days is the minimum observation period for asymptomatic infants who are suspected to have a high risk of developing NAS?” This question offered a drop-down response menu with selections of 1 day, 2 days, 3 days, 4 days, 5 days, 6 days, 7 days, 8 days, 9 days, ≥10 days, or unsure. Our survey found that institutions observed patients for a range of 1 to 7 days, which mirrored the findings of the Canadian report.³ Most institutions in our survey followed a minimum observation period of 3 or 5 days.

Responses to our survey regarding pharmacologic agents used in the outpatient management of NAS were also inconsistent. Morphine, phenobarbital, metha-done, and clonidine were reported in our survey as being used in the outpatient setting. These findings are similar to the variation of agents used in the outpatient management setting in a study by Maalouf et al⁷ of the 2009–2011 Tennessee Medicaid Program data, which found that agents used for outpatient NAS management included phenobarbital (89%), dilute opium tincture (40%), methadone (19%), and morphine (7%). Unlike the findings from Maalouf et al,⁷ however, our survey revealed that morphine was the most frequently used outpatient pharmacologic agent. The AAP does not address discharging infants on pharmacotherapy for the outpatient management of NAS.² Institutions may be apprehensive about managing NAS in the outpatient setting, because outpatient pharmacotherapy has been associated with increased length of therapy and emergency department use.⁷ A 2017 Canadian survey found that 36.7% of surveyed institutions discharged an infant home on a first-line medication.³ Similar to the findings of the Canadian survey, only 20% of participants in our survey reported that their institution discharged infants on outpatient pharmacotherapy most of the time or sometimes, signifying that outpatient management is not widely used in current practice.

First-line pharmacologic agent, assessment tool, and location of care exhibited less variation of responses within our survey than observation period and outpatient pharmacologic agent. First-line pharmacologic agents for NAS treatment have typically included either morphine or methadone, although recent studies have been unable to consistently suggest superiority of one agent over the other.^1,2,8–11 Eighty percent of respondents in our survey stated that morphine was the primary pharmacologic agent used for NAS treatment. Compared with our findings, a 2006 survey identifying NAS management practices in the United States found that 83% of respondents used an opioid and 17% used phenobarbital as their first-line treatment for opioid withdrawal.¹ In our survey, neither opium tincture nor phenobarbital were reported as first-line agents and only 9% stated that their first-line treatment was methadone. This suggests a major shift in inpatient pharmacologic treatment has occurred over the past 15 years and that institutions are moving away from opium tincture, phenobarbital, and methadone as first-line agents.

Notably, 2 respondents in our survey reported using buprenorphine as their first-line pharmacologic agent. Sublingual buprenorphine has been previously studied as a novel first-line treatment and has been found to be well tolerated and effective at decreasing length of opioid treatment and length of stay.^12,13 In 2006, no institutions in the United States reported using sublingual buprenorphine as a first- or second-line agent for pharmacologic management.¹ A more recent California survey found that, although no hospitals used buprenorphine as a first-line medication, 5% indicated using buprenorphine as the most common second-line medication.⁴ This finding, along with the results of our survey, suggests that sublingual buprenorphine use for NAS treatment in the United States has increased as compared with its non-existent role in NAS therapy 15 years ago.

The addition of a second-line agent may be considered over monotherapy in a variety of situations, including in neonates with refractory NAS; however, limited guidance exists outlining the implementation of adjunct agents in a NAS treatment regimen.¹⁴ Phenobarbital has been commonly included as a second-line agent in cases of refractory NAS with polysubstance exposure in utero.¹⁵ In 2006, the most common drug used as an adjunct treatment was phenobarbital.¹ In comparison, our survey found that 59% of institutions used clonidine most frequently as an adjunct agent, whereas 30% of institutions reported using phenobarbital. Because of a lack of sedative and respiratory depressive properties, clonidine requires a potentially shorter overall duration of monitoring and therapy than phenobarbital, which may make it a more attractive adjunct agent.^16,17 Similar to the first-line agents, trends in adjunct therapy appear to have changed in the past 15 years. Regarding first-line and adjunct pharmacologic agents, evidence from our survey suggests a limited need for further research between morphine and methadone as first-line pharmacologic agents for NAS treatment. Instead, focus for future research could consider evaluating buprenorphine as a novel first-line agent for NAS management and determining best practices for adjunct pharmacologic therapies.

Recommendations by the AAP for NAS scoring suggest that each institution should choose an assessment tool and then train staff on how to use it.² The most common assessment tool used in 2006 was the Finnegan method, either the original or modified version.¹ According to our survey, the Finnegan assessment tool was still the most popular scoring method, with 93% of institutions using a version of the Finnegan Neonatal Abstinence Scoring Tool either alone or in conjunction with another scoring method.

In 2017, Grossman et al¹⁸ published a novel scoring system called ESC, by which infants were assessed on their ability to eat 1 or more ounces per feed, sleep undisturbed for 1 or more hours, and be consoled within 10 minutes. If 1 or more of these conditions were not met, the non-pharmacologic therapies were amplified or treatment with morphine was initiated.¹⁸ The pilot study found that significantly fewer infants were treated with morphine with the ESC approach than with the Finnegan approach.¹⁸ Seventeen percent of respondents to our survey reported their institution used the ESC method alone or in combination with another assessment tool. The Finnegan assessment tool remains as the most popular assessment method; however, it appears that the ESC method may be gaining favor as NAS management research and practice approaches shift to prioritizing non-pharmacologic management.

Regarding location of care, our survey revealed that infants typically roomed with the mother or in the newborn nursery during observation and non-pharmacologic therapy only. To assess under what situations a neonate with or at a high risk of developing NAS roomed with the mother, participants were asked a multiple response question with answers including during the observation period, while receiving non-pharmacologic therapy only, while receiving pharmacologic treatment, and these infants do not room with mom. To assess the primary site of care following maternal discharge or when the neonate was not allowed to room with the mother, participants were asked a single response question for each situation (observation, non-pharmacologic management only, and pharmacologic management) with answers including newborn nursery only, NICU only, pediatric unit only, and NICU or pediatric unit depending on the infant's condition. The practice of rooming infants with NAS with the mother was variable and, according to the results of our survey, appeared to be dependent on the degree of management the infant requires. Similar to the responses received in our survey, about half of institutions in California roomed infants receiving nonpharmacologic management with the mother, whereas infants receiving pharmacologic management seldom roomed with the mother.⁴

According to our survey, once infants with NAS required pharmacologic treatment, they were commonly transferred to higher levels of care in the NICU or, less commonly, the pediatric unit. Our findings were supported by the 2017 Canadian survey where 89.2% of institutions treated infants on pharmacologic therapy in the NICU.³ In our survey, institutions reported managing neonates with NAS in the pediatric unit for observation, non-pharmacologic management, and pharmacologic treatment; however, use of the pediatric unit was reported less frequently for managing infants with NAS than the NICU. Controversy exists regarding whether the high level of care provided in the NICU is a necessity for patients with NAS. One institutional study found that infants with NAS treated on the pediatric floor had a shorter length of stay (14.2 vs 27.1 days) and a shorter length of pharmacologic treatment (9 vs 18 days) than infants treated in the NICU.⁵ Institutions may consider managing these patients in the pediatric unit to improve patient outcomes and reserve the NICU for more critically ill patients.

A strength of our study was that survey participants represented institutions across the United States from a variety of geographic regions, NICU levels, number of pediatric beds, and types of pediatric services offered. However, there were limitations to our study. Our survey was distributed via organization email lists servers. The total number of individuals who received the link to our survey was unable to be determined for multiple reasons. First, the number of subscribers to these email list servers was unknown. Second, organization members may have forwarded the survey link on to non-member colleagues. Consequently, it was not possible to calculate a survey response rate. Also, institutions that did not have members within these organizations were less likely to be represented. Although participants were requested to disclose their institution name to prevent duplication, 27 participants did not list their institution of practice. It was possible that a hospital's practices were represented more than once without our knowledge. Most respondents were pharmacists and, although the participants were asked to answer on the basis of practices of their institution, the data might not have reflected the practices of all members of the health care team involved in the management of infants with NAS, including neonatologists and nurses.

Conclusion

As NAS incidence continues to rise, the development of standardized protocols for NAS treatment becomes increasingly imperative. This survey indicated large variations in current NAS treatment practices between institutions in the United States, with the greatest differences reported for inpatient observation period duration and pharmacologic agent of choice for outpatient management. Our survey also suggested that several areas of current NAS treatment practices have shifted in comparison to previous literature regarding NAS management in the United States. Based on our responses, morphine as the primary first-line agent and the increasing use of clonidine as an adjunct agent are notable changes in practice that have occurred in the past 15 years. Further research is needed to understand buprenorphine's role in NAS treatment and to determine a superior adjunct pharmacologic therapy. Our survey showed that the primary site of NAS management was in the inpatient setting. Although institutions may be apprehensive about outpatient pharmacologic management, the findings of this study suggest hospitals are beginning to explore other settings of NAS management outside of the NICU.

ABBREVIATIONS

AAP

American Academy of Pediatrics

ACCP

American College of Clinical Pharmacy

ESC

Eat, Sleep, Console

NAS

neonatal abstinence syndrome

NICU

neonatal intensive care unit

PPA

Pediatric Pharmacy Association

PRN

Pediatric Practice and Research Network