Abstract
Objective:
Rates of mother’s own milk (MOM) provision in the neonatal intensive care unit (NICU) vary widely, despite acceptance as the gold standard for nutrition in preterm infants. Direct breastfeeding (DBF) supports long-term provision of MOM, but factors that support DBF in preterm infants are unknown. The purpose of this study was to identify factors that predict DBF at oral feeding initiation and at NICU discharge.
Methods:
This was a retrospective cohort study of preterm infants born at ≤ 32 weeks who were receiving MOM at 32 weeks corrected gestational age (cohort 1) and at discharge to home (cohort 2). The primary outcomes were rates of DBF at oral feeding initiation (cohort 1) and at hospital discharge (cohort 2). We examined bivariate associations between infant characteristics, maternal sociodemographic factors, and hospital practices (e.g., lactation visit timing and frequency) with DBF outcomes and then built logistic regression models to determine the adjusted odds ratio and 95% confidence interval ([adjusted odds ratio [aOR] [95%CI]) for independent predictors of the DBF outcomes.
Results:
Sixty-four percent of eligible infants initiated DBF, and 51% were DBF at discharge. Sociodemographic, NICU, and lactation support factors were associated with both outcomes. Post hoc analysis showed that similar factors also influenced lactation support provision.
Conclusions:
Lactation support, NICU and sociodemographic variables influence DBF initiation and DBF at discharge. Interventions that optimize efficient use of available lactation support, address bias, and provide ample opportunity for DBF practice could improve rates.
Introduction
Breastfeeding and provision of mother’s own milk (MOM) remain the gold standard for nutrition in all newborns. 1 MOM especially reduces morbidity and mortality in preterm and critically ill infants admitted to the neonatal intensive care unit (NICU), but rates of MOM provision in preterm infants remain suboptimal and lower than their term counterparts.2–4 Direct breastfeeding (DBF) is an important intervention that can improve the duration and exclusivity of MOM provision both during the NICU stay and after discharge.5,6
A single DBF session per day during the NICU stay is associated with a higher likelihood of receiving MOM at NICU discharge and a longer duration of the breastfeeding relationship once home.5,6 One study showed that of mothers who directly breastfed on average 1 or more times per day, 84% and 44% were still providing breast milk for 1 and 4 months post-NICU discharge, respectively. This is in stark contrast to 23% and 6% providing milk at 1 and 4 months post-NICU discharge, respectively, in mothers providing <1 DBF on average per day. 7 Similarly, another study revealed that NICU infants who were DBF received MOM feeds for almost 1 year, versus 5 months for infants receiving only expressed breast milk (EBM). 8 Feeding EBM also carries a higher risk of nutritional and immunological degradation compared with DBF and is associated with less physiological stability and fewer opportunities for mother–infant bonding.9–17
Similarly, DBF has been shown to provide specific benefits when infants receive their first oral exposure at the breast rather than bottle. One study demonstrated a shorter transition to oral feeding and younger postmenstrual age at full oral feeding and discharge when infants were DBF before bottle introduction, further reinforcing the importance of establishing a strong breastfeeding foundation at oral feeding introduction. 18
Despite the importance of DBF, the focus in most U.S. NICUs is on any human milk provision, and DBF is not as aggressively championed. Dedicated lactation support is vital to DBF success but less than half of U.S. NICUs staff an International Board Certified Lactation Consultant (IBCLC), and consensus is lacking on the most efficient use of this resource.19,20
Staff may also encourage the use of bottles over DBF, believing this will hasten discharge. 21 Evidence to support this belief is lacking and sends a detrimental message to parents of vulnerable infants. 22 Misconceptions about infant vulnerability and the developmental ability of preterm infants to breastfeed further exacerbate the problem. 23 Limited research suggests lower socioeconomic status, younger age, unmarried status, smoking, and lack of private insurance are associated with a lower likelihood of successful establishment of any DBF in the NICU, and racial and ethnic disparities continue to disproportionately affect certain populations.24–27
In addition, elucidation of modifiable predictors that can be used to guide future interventions to support successful DBF initiation and continuation in the NICU is especially sparse. Thus, the purpose of this study is to determine feeding characteristics and modifiable factors that contribute to (1) DBF initiation at the time oral feeds are introduced and (2) DBF continuation at NICU discharge.
Methods
This was a retrospective cohort study including infants born at less than or equal to 32 weeks gestational age (GA) at a 60-bed, Level III NICU with approximately 2,500 deliveries and 600 NICU admissions per year in a Baby Friendly designated facility. The study was reviewed and approved by Cincinnati Children’s Hospital Institutional Review Board, with reliance approval by the University of Cincinnati Medical Center Institutional Review Board.
In our facility, lactation consults were placed as a standing order for all infants admitted to the NICU. Visits were completed both in person and by phone follow-up as needed and directed by the parent and/or medical team. When it was determined that a preterm infant was medically and developmentally appropriate to begin oral feeding attempts, a lactation consultant (LC) was engaged for mothers expressing a desire to directly breastfeed. Efforts would be made to have the LC present for the first DBF, but due to staffing limitations this was not always feasible. DBF attempts would continue based on the mother’s wishes and availability at bedside.
Study population
All infants born at 32 weeks or less and admitted to the University of Cincinnati Medical Center NICU from January 2017 through December 2019 were included. We excluded infants: (1) who were transferred to another facility or deceased before hospital discharge, (2) who were not enterally fed by 32 weeks corrected gestational age (CGA), (3) with congenital malformations that precluded oral feeding, including anatomical defects such as craniofacial anomalies, pharyngeal clefts/webs, esophageal atresia/tracheoesophageal fistula, and gastrointestinal abnormalities such as omphalocele, gastroschisis, congenital diaphragmatic hernia, pyloric stenosis, and intestinal malrotation, (4) with confirmed genetic diagnosis or condition, (5) not discharged on full oral feeds, or (6) of mothers with a history of illicit drug use or who are HIV positive. The admission log for all infants admitted to the NICU was referenced to generate a list of potentially eligible patients.
The resulting cohort was further divided into two cohorts. Cohort 1 included those infants who were receiving any MOM between 31 and 32 weeks CGA. This GA was chosen to account for an existing chronic lung disease prevention strategy of using continuous positive airway pressure (CPAP) until 32 weeks CGA, thus limiting any oral feeding attempts until after this support was discontinued. Cohort 2 included those infants who were receiving any MOM within 7 days of discharge to home.
Outcomes
The primary outcomes were DBF at oral feeding initiation (cohort 1) and DBF at hospital discharge (cohort 2). DBF at oral feeding initiation was defined as either DBF as the first oral feeding or any DBF within 7 days of the first bottle. This definition was selected because of low frequency of DBF as the first oral exposure, while still maintaining early DBF exposure as part of the outcome. DBF at discharge was defined as any DBF within 7 days of discharge to home. A post hoc analysis of factors related to receiving a lactation consult within 24 hours before starting oral feeds was also examined because of its upstream effects on DBF at oral feeding initiation.
Predictors
While we were primarily interested in predictor variables that are potentially modifiable through quality improvement initiatives aimed at improving DBF outcomes (hereafter referred to as “modifiable” variables), additional maternal and infant characteristics were evaluated to describe the population and for inclusion in multivariable models as known, non-modifiable predictors of DBF.
Maternal characteristics included age, race/ethnicity, education, insurance status, zip code, prior breastfeeding experience, and comorbid diagnoses (gestational diabetes, prolonged rupture of membranes, and obesity).
Infant characteristics included birthweight, GA, delivery type, sex, 1- and 5-minute Apgar score, total days on CPAP or mechanical ventilation, length of stay (LOS), multiple gestation and comorbid diagnoses (sepsis, necrotizing enterocolitis, retinopathy of prematurity, intraventricular hemorrhage [IVH], and chronic lung disease/bronchopulmonary dysplasia).
Modifiable variables of interest for predicting DBF at oral feeding initiation (outcome 1) included: CGA at first oral feeding, defined as CGA at first DBF or first bottle feed; CGA at first skin-to-skin contact (SSC); SSC in the first 72 hours and 21 days of life; recent lactation visit, defined as presence or absence of lactation consult within 24 hours before starting oral feeds; and average weekly number of lactation consultant visits from birth to the first oral feeding. Typical practice for SSC during the study period was to initiate as soon as possible following delivery and continue attempts throughout the hospital stay based on parental availability at bedside.
Modifiable variables of interest for DBF at discharge (outcome 2) included: DBF as the first oral exposure, defined as the first DBF occurring before the first bottle feed; DBF at oral feeding initiation (outcome 1); CGA at first DBF; interval between first bottle and first DBF, defined as the number of days between first DBF and first bottle feed (positive integer if DBF occurred first, negative integer if bottle feeding occurred first); average weekly number of lactation consultant visit from the first oral feeding to discharge home; and total and average weekly number of DBFs during NICU stay.
Information on feeding route, type, and frequency, along with demographics and other medical history, was obtained directly from maternal and infant charts and was recorded in a REDCap database. Chart abstraction was performed primarily by a single individual (MM) with assistance from a single research coordinator. To ensure consistency in data collection, 10% of the secondary reviewer’s charts were also reviewed by the primary reviewer (MM). Some variables were obtained from data already collected for infants ≤ 1500 g as part of an existing quality improvement project. Some infant demographics and maternal information were obtained through an automated data extraction report from the electronic medical record. Data from all these sources were merged to create a complete dataset of key variables for all infants in the cohorts.
Statistical analysis
Cohort characteristics are reported using descriptive statistics. We determined unadjusted associations between maternal and infant characteristics and modifiable variables of interest with the two outcomes. In addition, we performed a post hoc analysis of predictors associated with receiving a lactation consult within the 24 hours prior to the first oral feed. The Kruskal–Wallis test was used for continuous variables and Chi-square or Fisher’s exact test for categorical variables.
Following bivariate analysis, a hierarchical logistic regression modeling approach was used to estimate the odds ratio (OR) and 95% confidence interval (CI) for predictors of interest, first modeling non-modifiable maternal (Stage 1) variables, then adding infant factors (Stage 2) known to be associated with breastfeeding in preterm infants, and finally adding the predictors of interest sequentially (Stage 3).
For DBF at oral feeding initiation, maternal factors included in the model were initially selected based on a priori knowledge of existing literature examining maternal demographics that impact breastfeeding outcomes and included maternal race, insurance status, and age in Stage 1 of the model, with all three variables carried over to the next stage, irrespective of statistical significance. Infant factors were selected for inclusion in stage 2 based on those reflective of infant vulnerability and clinical status and included GA, and GA at first oral feeding. Stage 3 carried over all maternal and infant characteristics from stage 2 regardless of significance and added SSC in the first 72 hours of life and recent lactation visit.
For DBF at discharge, the same maternal/infant variables and hierarchical approach were used in Stages 1 and 2. Stage 3 carried over all variables from the first two stages regardless of significance and added DBF at oral feeding initiation, LOS, and average weekly number of lactation consultant visits from the first oral feeding to discharge. The variables of interest in stage 3 for both outcomes were selected based on their clinical and statistical significance when building the final regression model.
Some relevant variables in stages 2 and 3 were ultimately excluded from analysis for both outcomes due to concern for collinearity. Rates of IVH were significantly different between the groups on bivariate analysis but given that this is not a known predictor of DBF and the association was in the opposite direction than would be expected, IVH was not entered into the final prediction model.
In post hoc analysis, we were interested in evaluating upstream predictors associated with receiving a lactation consult, given that a recent lactation consult was strongly associated with DBF at oral feeding initiation. The model for the outcome of the recent lactation consultant visit included stage 1 and 2 maternal and infant variables only with the same hierarchical approach used for the two primary outcomes.
Sample size and power
We estimated approximately 80 infants <32 weeks admitted to the study NICU per year. Over a 3 year study period, this would provide 240 charts for review. Assuming rates of breast milk at discharge of approximately 50% (120 eligible) and DBF at discharge of 50% (60 infants) based on local quality improvement data, at p < 0.05 this would provide 80% power to detect a 1.7-fold or great odds of achieving DBF.
Results
DBF at oral feeding initiation
Cohort 1 included 256 meeting eligibility criteria, including infants receiving MOM at 32 weeks CGA. Of these, 164 (64%) were DBF at oral feeding initiation, and 92 (36%) did not.
Bivariate analysis comparing maternal, infant, and NICU variables by DBF status at oral feeding initiation is summarized in Table 1. Maternal characteristics were comparable between the groups, except for a significantly higher average maternal age in the DBF initiation group (29.8 years vs. 26.8 years, p = 0.001).
Maternal and Infant Characteristics by Direct Breastfeeding Outcomes
DBF at first oral exposure refers to DBF before bottle introduction, while DBF at oral feeding initiation includes DBF as first oral exposure and any DBF within 7 days of beginning oral feeds.
indicates p < 0.05.
indicates p < 0.01.
indicates p < 0.001 in comparing DBF outcome within each cohort, based on Kruskall–Wallis test for continuous variables and Chi-square or Fisher’s exact test for categorical variables.
Infants receiving MOM within 7 days of discharge.
Mothers providing breastmilk at 32 weeks CGA.
CGA, corrected gestational age; CLD/BPD, chronic lung disease/bronchopulmonary dyplasia; CPAP, continuous positive airway pressure; DBF, direct breastfeeding; GDM, gestational diabetes mellitus; IVH, intraventricular hemorrhage; LC, lactation consultant; MOM, mother’s own milk; NEC, necrotizing enterocolitis; PROM, prolonged rupture of membranes; ROP, retinopathy of prematurity; SD, standard deviation; SSC, skin-to-skin contact.
Infant characteristics were also similar, except for fewer total days on CPAP (18.9 vs. 24.2 days, p = 0.0001) and lower CGA at first oral feeding (33.9 vs 34.6 weeks, p = 0.0001) for the DBF initiation group. There was no significant difference for variables associated with illness severity and rates of prematurity-related comorbidities.
Infants in the DBF initiation group had more SSC occurrences at both the 72-hour and 21-day mark (average 1.4 vs. 0.9 occurrences, p = 0.001; 6.7 vs. 4.0 occurrences, p < 0.0001), though CGA at first SSC did not significantly differ. Eighty-three percent of infants who received a lactation consult within 24 hours before the first oral feeding were DBF at oral feeding initiation, versus 40% of those who had not (p < 0.001).
Multiple variable logistic regression analysis revealed higher odds of DBF initiation for non-Black maternal race ([adjusted odds ratio [aOR] 2.29, 95% CI 1.09–4.80, p = 0.028) and older maternal age (aOR 1.10, 95% CI 1.03–1.17, p = 0.002). Higher CGA at first oral feeding was associated with lower odds of DBF initiation (aOR 0.75, 95% CI 0.59–0.95, p = 0.018) (Table 2). Each SSC in the first 72 hours of life significantly increased the odds of DBF initiation by 1.8-fold (aOR 1.80, 95% CI 1.24–2.63, p = 0.002). Similarly, infants who had a recent lactation consult had nearly fivefold higher odds of DBF initiation versus those who had not (aOR 4.84, 95% CI 2.40–9.75, p < 0.0001) (Table 2).
Adjusted Odds of DBF Outcomes for Hypothesized Variables of Interest, with Adjustment for Key Maternal and Infant Characteristics
indicates p < 0.05.
indicates p < 0.01.
indicates p < 0.001.
CGA, corrected gestational age; DBF, direct breastfeeding; LC, lactation consultant; SSC, skin-to-skin contact.
DBF at NICU discharge
Of the 256 infants in Cohort 1, 163 (64%) continued to receive MOM at the time of NICU discharge, comprising Cohort 2. Eighty-three (51%) of Cohort 2 were DBF at discharge, and 80 (49%) were not (Table 1).
No significant differences were noted in maternal or infant characteristics between the Cohort 2 groups in bivariate analysis (Table 1). A statistically significant difference was noted in IVH rates between the groups, with those having IVH more likely to be DBF at discharge (68.6% vs. 44.7%, p = 0.014).
Infants in the DBF at discharge group had more SSC occurrences at the 72-hour (average 1.6 vs. 1.1, p = 0.045) and 21-day mark (8.0 vs. 5.3, p = 0.0004) compared with those who were not DBF, with no difference in CGA at first SSC (30.1 weeks vs. 30.1 weeks, p = 0.625).
CGA at first DBF, the time between first DBF and bottle introduction, and DBF as the first oral exposure did not differ between the groups. However, infants in the DBF at discharge group had three times as many total direct breastfeeds and more weekly direct breastfeeds during their NICU stay versus the infants who were not DBF at discharge (average 12.9 vs. 4.3, p < 0.0001; 4.3 vs. 1.5, p < 0.0001). Likewise, the number of lactation visits from first oral feeding to discharge were almost double in the DBF group (weekly average 1.3 vs. 0.7, p < 0.001).
In multiple variable logistic regression analysis, the odds of DBF at discharge were 0.5 lower per week of increasing GA at birth (aOR 0.50, 95% CI 0.32–0.79, p = 0.002) (Table 2). Infants who initiated DBF had significantly higher odds of DBF at discharge as compared with than those who were not DBF at the time of oral feeding initiation (aOR 17.82, 95% CI 3.28–96.91, p = 0.0009).
Unadjusted analysis did not show a difference in LOS between DBF at discharge groups, but after controlling for maternal/infant and lactation variables, each additional day of admission was found to be associated with a 5% reduction in the odds of DBF at discharge (aOR 0.95, 95% CI 0.91-0.99, p = 0.007). Average weekly lactation consults from first oral feeding to discharge also remained significant after adjustment (aOR 2.79, 95% CI 1.34-5.83, p = 0.006) (Table 2).
Recent lactation consult
Post hoc bivariate analysis of factors associated with presence or absence of a recent lactation consult revealed similar baseline characteristics between the groups, except for insurance status and total CPAP days. Of those with Medicaid or other government insurance, 53.2% had a recent lactation consult, versus 71% of those holding commercial insurance (p = 0.012). Infants who received a recent lactation consult also had fewer average days on CPAP (19.2 vs. 22.7, p = 0.018) and were younger at the first oral feeding (33.9 vs. 34.4, p = 0.0003) and first DBF (33.9 vs. 34.4, p = 0.004) than those who had not.
In the adjusted model, Medicaid insurance status remained significantly associated with a lower odds of having a recent lactation consult (aOR 0.44, 95% CI 0.22-0.88, p = 0.02), as did older GA at first oral feeding (aOR 0.73, 95% CI 0.60-0.90, p = 0.003) (Table 3).
Adjusted Odds of Lactation Support Outcomes for Selected Maternal/Infant Characteristics and Hypothesized Variables of Interest
indicates p < 0.05.
indicates p < 0.01.
CGA, corrected gestational age; DBF, direct breastfeeding.
Discussion
The results of our study provide new insight on modifiable and behavioral factors that influence DBF in the NICU, which has been shown to improve overall rates of MOM provision.5–8 Maternal characteristics, infant factors, and hospital practices (environmental factors) all play unique roles in determining whether an infant will DBF during their NICU stay.
We found that infants of older, non-Black mothers were more likely to initiate DBF, indicating that socio-economically disadvantaged mothers may face disproportionately more barriers to DBF initiation. This is consistent with previous studies that show a positive association of DBF in preterm infants with older maternal age, higher maternal education level, and white race.24–26 Likewise, factors associated with provision of lactation support itself reveal potential sources of disparity and bias, as a higher proportion of mothers from racial and ethnic minority groups are covered under Medicaid, which was itself associated with a lower likelihood of receiving timely LC visits.
Infant factors such as longer CPAP duration and older CGA when starting oral feeds were associated with lower rates of DBF initiation, suggesting that those infants who went on to directly breastfeed were either more developmentally mature or had a milder clinical course, allowing them to start feeds sooner than their non-DBF counterparts. However, no significant differences in major comorbidities were noted, and older infants were also less likely to receive appropriate LC support, revealing both a direct and indirect effect of CGA on DBF initiation. This perhaps indicates a tendency to provide better lactation support for infants who are more medically stable and able to orally feed sooner. Similarly, younger infants may be viewed as having more time to work on oral feeding, creating less pressure around timing of discharge related to DBF. Younger infants also achieved higher DBF rates at discharge, further reinforcing the growing concern over greater resource allocation for the youngest, most vulnerable NICU patients, leaving older infants at greater risk for poor breastfeeding rates. 28
Hospital practices such as lactation support and prioritization of SSC also heavily influenced DBF outcomes in our study. Dyads who saw an LC within 24 hours before the first oral feeding were more likely to initiate DBF than those who did not. Whether DBF is the result of timely LC support or the driver of LC engagement at oral feeding initiation is unclear. Unfortunately, many NICUs have limited or no access to specialized lactation support because of staffing limitations, despite evidence that shows increased odds of MOM initiation as well as any and exclusive MOM feeding with NICU-specific lactation support. 19 As a result, LCs are often only engaged on an “as needed” basis. But our findings suggest that once oral feedings have begun, a consistent LC presence can facilitate the frequent opportunities for DBF that are needed to maintain the breastfeeding relationship. A nonstandard approach to consultation not only engages LCs after mothers may already be experiencing barriers that lead to early DBF cessation, but it inadvertently creates disparity in resource allocation as well. Focusing efforts on the days leading up to the first anticipated oral feeding, setting a minimum standard for lactation visit frequency (e.g., one visit per week), and establishing criteria for consultation based on unit staffing and feasibility may provide the highest yield for optimal breastfeeding outcomes. Other interventions such as prenatal breastfeeding education and lactation training for the entire care team may further reduce the IBCLC workload and allow for more optimal use of their specialized expertise.
Skin-to-skin contact is another established predictor of any and exclusive MOM feeding in preterm infants at discharge, and our findings extend this relationship to improved DBF initiation rates as well. 29 Given that insufficient maternal milk supply is a common reason for early weaning, interventions such as SSC may encourage maintenance of lactation for long enough to achieve transition to DBF. 5
Our findings also indicate that about half of mothers who intend to directly breastfeed their infants discontinue these efforts at some point before discharge. Rigid feeding schedules, inadvertent promotion of bottle-feeding, limited access to lactation support, and the NICU environment interfere with mother–infant bonding and maintenance of the DBF relationship. A common misconception among NICU staff and parents is that bottle-feeding is less strenuous or easier to master for the infant and facilitates earlier discharge, but studies evaluating feeding milestones and LOS fail to show this association. 22 In fact, breastfeeding is associated with improved physiological stability and fewer bradycardia and desaturation events compared with bottle-feeding, which are often factors that can delay discharge in otherwise discharge-ready infants.12–15,19 In our study, longer LOS was associated with lower odds of DBF at discharge, perhaps a consequence of disease severity or slower achievement of oral feeding independence, maternal stress and fatigue associated with a longer NICU stay or the promotion of bottle-feeding to facilitate discharge. Our results continue to support that DBF is not associated with a longer hospital stay and should be encouraged whenever feasible as part of the discharge feeding plan.
Interventions designed to mitigate structural, institutional, and interpersonal contributors to socioeconomic disparities in DBF outcomes can and should be developed. The provision of targeted breastfeeding education and group-specific interventions to support breastfeeding among minority women have previously been shown to improve breastfeeding rates, and tailored incorporation of such programs based on individual unit demographics may help reduce disparities and increase breastfeeding in at-risk populations. 30
Strengths of this study include the evaluation of DBF at distinct time points—oral feeding initiation and at hospital discharge—providing insight into when specific predictors may be most influential. Evaluation of lactation support predictors further reveals disparities and subjective application of this valuable resource, providing evidence to support standardization and equitable health care delivery. Lastly, particular attention was given to the identification of modifiable predictors toward which future interventions can be targeted.
Limitations of this study include: 1) retrospective design, 2) low incidence of DBF in this population, and 3) completion at a single center. To address the disadvantages associated with a retrospective design, a staged logistic regression model was used to evaluate the independent effects of maternal, infant, and NICU factors. To address the low incidence of DBF in our unit, more liberal inclusion criteria were used. This could have resulted in unintended selection bias, but given that one DBF per week is considered a reasonable—or even optimal—frequency for preterm infants in the NICU, it was determined that these inclusion criteria would have minimal influence on the desired outcome. Lastly, while this study was conducted at a single center, the demographics of this cohort are similar to that which would be found in most Level 3 NICUs, improving the potential for generalizability.
Conclusion
DBF promotes optimal MOM feeding outcomes in preterm infants. Timing and frequency of lactation support, sociodemographic factors, and infant GA influence DBF initiation and continuation through NICU discharge. Interventions that optimize efficient use of available lactation support, address bias, encourage early SSC, and provide ample opportunity for DBF at oral feeding initiation and throughout the hospital stay can improve rates and support better long-term health outcomes for this vulnerable population.
Footnotes
Acknowledgment
The authors would like to acknowledge the contribution of lactation and nursing staff in providing important insight into the daily breastfeeding and human milk practices in the unit that helped shape the concept for this study.
Authors’ Contributions
M.M.: conceptualization, methodology, formal analysis, investigation, writing (original draft, review, and editing), and visualization. L.N.-R.: conceptualization, methodology, formal analysis, investigation, and writing (review and editing). H.K.: conceptualization, methodology, formal analysis, investigation, and writing (review and editing). C.L.: formal analysis. S.E.: formal analysis. L.W.: conceptualization, methodology, formal analysis, investigation, and writing (review and editing).
Disclosure Statement
No competing financial interests exist.
Funding Information
Limited funding was provided by the Division of Neonatology and Pulmonary Biology at Cincinnati Children’s Hospital for statistical analysis using resources dedicated to the support and execution of fellow research projects.