Classifying early infant feeding status from clinical notes using natural language processing and machine learning.

The objective of this study is to develop and evaluate natural language processing (NLP) and machine learning models to predict infant feeding status from clinical notes in the Epic electronic health records system. The primary outcome was the classification of infant feeding status from clinical notes using Medical Subject Headings (MeSH) terms. Annotation of notes was completed using TeamTat to uniquely classify clinical notes according to infant feeding status. We trained 6 machine learning models to classify infant feeding status: logistic regression, random forest, XGBoost gradient descent, k-nearest neighbors, and support-vector classifier. Model comparison was evaluated based on overall accuracy, precision, recall, and F1 score. Our modeling corpus included an even number of clinical notes that was a balanced sample across each class. We manually reviewed 999 notes that represented 746 mother-infant dyads with a mean gestational age of 38.9 weeks and a mean maternal age of 26.6 years. The most frequent feeding status classification present for this study was exclusive breastfeeding [n = 183 (18.3%)], followed by exclusive formula bottle feeding [n = 146 (14.6%)], and exclusive feeding of expressed mother's milk [n = 102 (10.2%)], with mixed feeding being the least frequent [n = 23 (2.3%)]. Our final analysis evaluated the classification of clinical notes as breast, formula/bottle, and missing. The machine learning models were trained on these three classes after performing balancing and down sampling. The XGBoost model outperformed all others by achieving an accuracy of 90.1%, a macro-averaged precision of 90.3%, a macro-averaged recall of 90.1%, and a macro-averaged F1 score of 90.1%. Our results demonstrate that natural language processing can be applied to clinical notes stored in the electronic health records to classify infant feeding status. Early identification of breastfeeding status using NLP on unstructured electronic health records data can be used to inform precision public health interventions focused on improving lactation support for postpartum patients.

COVID-19 booster enhances IgG mediated viral neutralization by human milk in vitro.

Background: Facilitated by the inability to vaccinate, and an immature immune system, COVID-19 remains a leading cause of death among children. Vaccinated lactating mothers produce specific SARS-CoV-2 antibodies in their milk, capable of neutralizing the virus in vitro. Our objective for this study is to assess the effect of COVID-19 booster dose on SARS-CoV-2 antibody concentration and viral neutralization in milk, plasma, and infant stool. Methods: Thirty-nine mothers and 25 infants were enrolled from December 2020 to May 2022. Milk, maternal plasma, and infants' stool were collected at various time-points up to 12 months following mRNA COVID-19 vaccination. A subgroup of 14 mothers received a booster dose. SARS-CoV-2 antibody levels and their neutralization capacities were assessed. Results: Booster vaccination led to significantly higher IgG levels within human milk and breastfed infants' stool. In vitro neutralization of VSV-gfp-SARS-CoV-2-S-gp, a laboratory safe SARS-CoV-2 like pseudovirus, improved following the booster, with a 90% increase in plasma neutralization and a 60% increase in milk neutralization. We found that post-booster neutralization by human milk was highly correlated to SARS-CoV-2 IgG level. In support of our correlation result, Protein G column depletion of IgG in milk yielded a significant reduction in viral neutralization (p = 0.04). Discussion: The substantial increase in neutralizing IgG levels in milk and breastfed infants' stool post-booster, coupled with the decrease in milk neutralization capabilities upon IgG depletion, underscores the efficacy of booster doses in augmenting the immune response against SARS-CoV-2 in human milk.

Milk Biomarkers of Secretory Activation in Breast Pump-Dependent Mothers of Preterm Infants: An Integrative Review.

Background: Lack of mother's own milk (MOM) at discharge from the neonatal intensive care unit (NICU) is a global problem and is often attributable to inadequate MOM volume. Evidence suggests that the origins of this problem are during the first 14 days postpartum, a time period that includes secretory activation (SA; lactogenesis II, milk coming in). Objectives: To describe and summarize evidence regarding use of MOM biomarkers (MBMs) as a measure of SA in pump-dependent mothers of preterm infants in the NICU and to identify knowledge gaps requiring further investigation. Methods: An integrative review was conducted using Whittemore and Knafl methodology incorporating the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) checklist. A search using electronic databases MEDLINE (through PubMed) and CINAHL (Cumulative Index to Nursing and Allied Health Literature) and reference lists of included articles was conducted. Results: Of the 40 articles retrieved, 6 met the criteria for inclusion. Results revealed the following five findings: (1) Achievement of SA defined by MBMs is delayed and/or impaired in mothers of preterm infants. (2) MBMs are associated with pumped MOM volume. (3) Achievement of SA defined by MBMs is associated with pumping frequency. (4) Delayed and/or impaired achievement of SA defined by MBMs may be exacerbated by maternal comorbidities. (5) There is a lack of consensus as to which MBM(s) and analysis techniques should be used in research and practice. Conclusions: MBMs hold tremendous potential to document and monitor achievement of SA in mothers of preterm infants, with multiple implications for research and clinical practice.

Mother's Own Milk Versus Donor Human Milk: What's the Difference?

Mother's own milk (MOM) is known to decrease complications in preterm infants and when unavailable, it is recommended that preterm very low-birth weight infants be fed donor human milk (DHM). Due to the pasteurization, processing, and lactation stage of donors, DHM does not contain the same nutritional, immunologic, and microbial components as MOM. This review summarizes the differences between MOM and DHM, the potential effects on health outcomes, and the clinical implications of these differences. Finally, implications for research and clinical practice are discussed.

A Scoping Review of the Oral Microbiome in Preterm Infants.

The purpose of this scoping review was to examine the oral microbiome composition in preterm infants, sampling and collection methods, as well as exposures associated with oral microbiome composition and health implications. We conducted a scoping review of the literature using the Arskey and O'Malley framework. We identified a total of 13 articles which met our inclusion criteria and purpose of this scoping review. Articles included in this review compared the oral microbiome in preterm infants to term infants, examined alterations to the oral microbiome over time, compared the oral microbiome to different body site microbiomes, and explored associations with clinically relevant covariates and outcomes. Exposures associated with the diversity and composition of the oral microbiome in preterm infants included delivery mode, oral feeding, oropharyngeal care, skin-to-skin care, and antibiotics. Day of life and birth weight were also associated with oral microbiome composition. The oral microbiome may be associated with the composition of the tracheal and gut microbiomes, likely due to their proximity. Alpha and beta diversity findings varied across studies as well as the relative abundance of taxa. This is likely due to the different sampling techniques and timing of collection, as well as the wide range of infant clinical characteristics. Multiple factors may influence the composition of the oral microbiome in preterm infants. However, given the heterogeneity of sampling techniques and results within this review, the evidence is not conclusive on the development as well as short- and long-term implications of the oral microbiome in preterm infants and needs to be explored in future research studies. KEY POINTS: · Day of life is a critical factor in oral microbiome development in preterm infants.. · The oral microbiome may be associated with tracheal and gut microbiome colonization.. · Future research should examine sampling methodology for examining the oral microbiome.. · Future research should explore associations with the oral microbiome and adverse health outcomes..

Untargeted Metabolomic Analysis of Lactation-Stage-Matched Human and Bovine Milk Samples at 2 Weeks Postnatal.

Epidemiological data demonstrate that bovine whole milk is often substituted for human milk during the first 12 months of life and may be associated with adverse infant outcomes. The objective of this study is to interrogate the human and bovine milk metabolome at 2 weeks of life to identify unique metabolites that may impact infant health outcomes. Human milk (n = 10) was collected at 2 weeks postpartum from normal-weight mothers (pre-pregnant BMI < 25 kg/m2) that vaginally delivered term infants and were exclusively breastfeeding their infant for at least 2 months. Similarly, bovine milk (n = 10) was collected 2 weeks postpartum from normal-weight primiparous Holstein dairy cows. Untargeted data were acquired on all milk samples using high-resolution liquid chromatography-high-resolution tandem mass spectrometry (HR LC-MS/MS). MS data pre-processing from feature calling to metabolite annotation was performed using MS-DIAL and MS-FLO. Our results revealed that more than 80% of the milk metabolome is shared between human and bovine milk samples during early lactation. Unbiased analysis of identified metabolites revealed that nearly 80% of milk metabolites may contribute to microbial metabolism and microbe-host interactions. Collectively, these results highlight untargeted metabolomics as a potential strategy to identify unique and shared metabolites in bovine and human milk that may relate to and impact infant health outcomes.

Sociodemographic Factors and Intestinal Microbiome Development in Preterm, Very Low Birth Weight Infants.

OBJECTIVE: Preterm very low birth weight (VLBW) infants are at risk for intestinal morbidities and dysbiotic development of the intestinal microbiome. Despite the influence of sociodemographic factors on premature infant health outcomes, whether they shape the intestinal microbiome early in life is not clear. The objective was to explore the associations between race, sex, and socioeconomic status and the intestinal microbiome of VLBW infants during the first 4 weeks of life. STUDY DESIGN: This was a secondary analysis of data from an ongoing randomized trial of 79 infants ≤30 weeks' gestation and ≤1,500 g. Stool samples were collected at week 1 through week 4, frozen to -80°C and analyzed by 16S rRNA sequencing of the V4 region using Illumina MiSeq. Reads were analyzed to measure α and ß diversity as well as relative abundance of bacteria in the intestinal microbiome. RESULTS: Of the 79 infants, 63 had at least one sample available. Twenty-three (37%) of infants were African American, 30 (48%) were male, and 44 (71%) had Medicaid insurance. There were no statistically significant (<0.05) differences in α diversity or ß diversity, and the differential abundance analysis suggests limited patterns of distinction in the intestinal microbiome between non-African American and African American infants, male and female infants, and infants with maternal private or Medicaid insurance. CONCLUSION: Our results suggest race, sex, and socioeconomic status shape colonization of specific microorganisms to a limited extent. Future studies should confirm these findings and determine clinical relevance through further study of differentially abundant microorganisms and additional factors contributing to colonization patterns. KEY POINTS: · Diversity of the gut microbiome was similar between infants of varying race, sex, and socioeconomic status.. · We observed sociodemographic-linked differences in colonization of individual taxa.. · Further study is required to confirm these results and the clinical relevance of these findings..

Oral Care in Critically Ill Infants and the Potential Effect on Infant Health: An Integrative Review.

BACKGROUND: Critically ill infants admitted to the neonatal intensive care unit are at risk for ventilator-associated pneumonia and abnormal oral colonization. Adherence to evidence-based guidelines for oral care in critically ill adults is associated with improved short- and long-term health outcomes. However, oral care guidelines for critically ill infants admitted to the neonatal intensive care unit have not been established, possibly increasing their risk of ventilator-associated pneumonia and other health complications. OBJECTIVE: To describe and summarize the evidence regarding oral care for critically ill infants admitted to the neonatal intensive care unit and to identify gaps needing further investigation. METHODS: The MEDLINE (through PubMed) and CINAHL databases were searched for observational studies and randomized controlled trials investigating the effect of oral care on oral colonization, ventilator-associated pneumonia, and health outcomes of infants in the neonatal intensive care unit. RESULTS: This review of 5 studies yielded evidence that oral care may promote a more commensal oral and endotracheal tube aspirate microbiome. It may also reduce the risk of ventilator-associated pneumonia and length of stay in the neonatal intensive care unit. However, the paucity of research regarding oral care in this population and differences in oral care procedures, elements used, and timing greatly limit any possible conclusions. CONCLUSIONS: Oral care in critically ill infants may be especially important because of their suppressed immunity and physiological immaturity. Further appropriately powered studies that control for potential covariates, monitor for adverse events, and use recommended definitions of ventilator-associated pneumonia are needed to make clinical recommendations.

Pasteurization of human milk affects the miRNA cargo of EVs decreasing its immunomodulatory activity.

In this report, we evaluated the effect of the pasteurization (P) process of mother's own milk (MOM) on the miRNA content of extracellular vesicles (EVs) and its impact on innate immune responses. Differences in size or particle number were not observed upon pasteurization of MOM (PMOM). However, significant differences were observed in the EV membrane marker CD63 and miRNA profiles. miRNA sequencing identified 33 differentially enriched miRNAs between MOMEV and PMOMEV. These changes correlated with significant decreases in the ability of PMOMEV to modulate IL-8 secretion in intestinal Caco2 cells where only MOMEV were able to decrease IL-8 secretion in presence of TNFα. While EVs from MOMEV and PMOMEV were both able to induce a tolerogenic M2-like phenotype in THP-1 macrophages, a significant decrease in the transcript levels of IL-10 and RNA sensing genes was observed with PMOMEV. Together, our data indicates that pasteurization of MOM impacts the integrity and functionality of MOMEV, decreasing its EVs-mediated immunomodulatory activity. This data provides biomarkers that may be utilized during the optimization of milk processing to preserve its bioactivity.

Leveraging mHealth and a milk expression frequency biomarker during postpartum to prolong lactation among parents of critically ill infants: a pilot study.

OBJECTIVE: To assess the feasibility and potential benefits of personalized biomarker-based text messages in prolonging lactation among parents of critically ill infants. STUDY DESIGN: Thirty-six participants were randomized to receive either daily texts with Mother's Own Milk (MOM) sodium levels or standard care. Surveys at months 1 and 3 assessed whether infants were receiving exclusive MOM feeding, any MOM feeding, and whether the parent was still lactating. Kaplan-Meier and log-rank tests were used for time-to-event analysis within and between intervention and control groups. RESULTS: Participants were predominantly on Medicaid (72%), delivered infants <1500 g, and by c-section (56%). Kaplan-Meier probabilities at month 3 suggest prolonged MOM feeding (63% [0.95CI, 0.43-0.91] vs. 41% [0.95CI, 0.21-0.67]) and lactation (63% [0.95CI, 0.42-0.95] vs. 37% [0.95CI, 0.18-0.76]) in the enhanced group compared to the control group. CONCLUSION: Personalized biomarker-based text messages are feasible and may prolong lactation and MOM feeding among parents of critically ill infants.

Comparison of 3 Methods for Measuring Gastric Tube Length in Newborns: A Randomized Clinical Trial.

BACKGROUND: Incorrectly positioned gastric tubes occur in approximately 60% of infants hospitalized in the neonatal intensive care unit (NICU), increasing the risk of potentially serious complications. PURPOSE: To compare 3 methods of determining gastric tube insertion length in infants in the NICU. METHODS: In this randomized triple-blind clinical trial, 179 infants admitted to the NICU were randomized to have their gastric tube insertion length determined by 1 of 3 methods: (1) the nose, earlobe, mid-umbilicus (NEMU) method, (2) a weight-based method, or (3) an age-related height-based (ARHB) method. Positioning of the gastric tube was verified by radiograph. R software was used for analyses. To compare categorical variables, Fisher's exact test, χ 2 tests, and simulated χ 2 tests were used. RESULTS: Overall, infants had a mean gestational age of 35 weeks, 115 (58.8%) were male, and the mean birth weight was 2481.5 g. Upon radiological assessment, 145 gastric tubes (81.3%) were correctly positioned in the gastric body or greater curvature of the stomach with the weight-based method having the highest percentage of correctly positioned gastric tubes (n = 53; 36.6%), followed by the ARHB method (n = 47; 32.4%) and the NEMU method (n = 45; 31.0%). No significant differences were identified between groups ( P = .128). IMPLICATION FOR PRACTICE AND RESEARCH: Despite the NEMU method being the most commonly used method in clinical practice, the weight-based and ARHB methods to determine gastric tube insertion length may be more accurate.

Detection of SARS-CoV-2 IgA and IgG in human milk and breastfeeding infant stool 6 months after maternal COVID-19 vaccination.

OBJECTIVE: Assess presence, durability, and neutralization capacity of SARS-CoV-2-specific antibodies in breastfeeding infants' stool, mother's plasma and milk following maternal vaccination. DESIGN: Thirty-seven mothers and 25 infants were enrolled between December 2020 and November 2021 for this prospective observational study. All mothers were vaccinated during lactation except three, which were vaccinated during pregnancy. Milk, maternal plasma, and infants' stool was collected pre-vaccination and at periods up to 6 months following COVID-19 vaccine series initiation/completion. SARS-CoV-2 antibody levels and their neutralization capacities were assessed. RESULTS: SARS-CoV-2-specific IgA and IgG levels were higher in infant stool post-maternal vaccination amongst milk-fed compared to controls. Maternal SARS-CoV-2-specific IgA and IgG concentrations decreased over 6 months post-vaccination but remained higher than pre-vaccination levels. We observed improved neutralization capacity in milk and plasma after COVID-19 vaccination. CONCLUSIONS: The presence of SARS-CoV-2-specific antibodies in infant stool following maternal vaccination offers further evidence of the lasting transfer of these antibodies through breastfeeding.

Nonpharmacologic Factors Affecting Milk Production in Pump-Dependent Mothers of Critically Ill Infants: State of the Science.

BACKGROUND: Improved health outcomes for critically ill infants including neurodevelopmental, immunological, and cost benefits are dependent upon the dose and duration of mother's own milk feedings. However, mothers of infants admitted to the neonatal intensive care unit (NICU) must express their milk (pump-dependent) and often struggle with milk production. PURPOSE: To examine the state of the science on nonpharmacologic modifiable expression factors that may influence milk production in pump-dependent mothers of critically ill infants admitted to the NICU. DATA SOURCES: PubMed, Embase, and CINAHL databases from 2005 to 2020. SEARCH STRATEGY: Guided by the lactation conceptual model, the authors searched for peer-reviewed studies with terms related to milk volume, pump dependency, critically ill infants, and modifiable factors, which may influence milk volume and assessed 46 eligible studies. DATA EXTRACTION: Data were extracted by 3 reviewers with a systematic staged review approach. RESULTS: Evidence from 26 articles found expressed milk volume may be influenced by multiple potentially modifiable factors. Simultaneous expression with a hospital-grade electric pump at least 5 times per day beginning 3 to 6 hours after delivery, and adding complementary techniques including hand expression, hands-on-pumping, music, breast massage, warm compresses, skin-to-skin care, and the mother expressing near her infant may promote increased milk volume. IMPLICATIONS FOR PRACTICE AND RESEARCH: Healthcare providers should assist pump-dependent mothers with early initiation and frequent milk removal with a hospital-grade breast pump. Further research is needed to explore optimal frequency of expressions, dose and timing of skin-to-skin care, and other targeted strategies to improve expressed milk volume.

Content validity of a Safe Nursing Care Checklist for a neonatal unit.

BACKGROUND: Nursing checklists have been shown to improve communication, reduce the occurrence of adverse events, and promote safe, quality care in different care settings. However, to date, there is no validated patient care safety checklist for nurses caring for infants in Neonatal Intensive Care Units (NICU). AIM: To describe development and content validation of the "Safe Nursing Care Checklist for Infants Hospitalized in the Neonatal Intensive Care Unit". STUDY DESIGN: Online Survey. METHODS: Based upon an integrative literature review, we developed a checklist focused on safe nursing care for infants in the NICU. Nursing experts participated in three rounds of a content validation process where they rated the items online. An agreement level ≥0.90 was required for inclusion in the checklist. Forty- three expert nurses with experience working in the NICU and who were certified in neonatal nursing or had a master's or doctoratal degree in child health provided content validation of the patient care checklist. RESULTS: The final checklist contained 45 items with content validation index scores greater than 90%. The instrument was structured into six dimensions including patient identification, effective communication, medication safety, infection prevention, fall prevention, and pressure injuries/skin injuries prevention. CONCLUSION: Content validity was established for the "Safe Nursing Care Checklist for Infants Hospitalized in the Neonatal Intensive Care Unit" which can identify strengths and weaknesses in safe nursing care for infants in the NICU as well as direct educational interventions to promote nursing care based on scientific evidence. RELEVANCE TO CLINICAL PRACTICE: This checklist can potentially be used by bedside nurses to promote provision of safe care to infants in the NICU and to guide corrective strategies and encourage evidence-based decision-making. Validation in the clinical setting is needed.

Influence of Sociodemographic and Occupational Factors on Safety Culture in Neonatal Intensive Care in Brazil.

BACKGROUND: Patient safety culture is influenced by factors such as professional category, experience, and age. Understanding these factors can inform initiatives to improve safety. PURPOSE: To evaluate the relationship between sociodemographic and occupational characteristics on health professionals' perception of patient safety culture. METHODS: A cross-sectional study involving 514 health care professionals from Brazilian neonatal intensive care units was conducted using the Hospital Survey on Patient Safety Culture. RESULTS: Several sociodemographic and occupational characteristics were associated with higher perceptions of safety culture, including older age and having a higher level of education. CONCLUSION: Sociodemographic and occupational factors may influence the safety culture in neonatal intensive care units and should be considered when developing and implementing strategies to improve safety.

Maternal Voice Exposure and Its Effect on Premature Infants' Feeding Milestones: A Systematic Review.

BACKGROUND: Premature infants experience alterations in maternal stimulation (including auditory sensory alteration such as talking or singing to the infant in the neonatal intensive care unit) due to admission to the neonatal intensive care unit. Because of their physiological and neurobehavioral immaturity, infants are at an increased risk of delays in reaching feeding milestones (a key developmental milestone), which often need to be achieved before discharge. PURPOSE: This systematic review evaluated the literature regarding the effect of maternal speech on achievement of feeding milestones in premature infants. DATA SOURCES: A systematic search of CINAHL, PubMed, Web of Science, and Google Scholar from 2010 to 2021. STUDY SELECTION: Studies were selected if they examined the effect of maternal voice interventions on premature infants' feeding milestones. DATA EXTRACTION: Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were used. RESULTS: Six studies were identified. This systematic review of the literature on the effects of maternal voice on feeding milestones in premature infants found equivocal results. IMPLICATIONS FOR PRACTICE: Given the inconsistent results, this systematic review does not support a change in clinical practice. However, encouragement of maternal visits is highly recommended as the additional benefits of the mother's presence may extend beyond exposure to maternal voice. IMPLICATIONS FOR RESEARCH: More research is needed including use of more homogenous samples, application of recommended decibel levels, and utilization of an adequately powered randomized controlled trial to further examine the effects of maternal voice on feeding milestones.

Skin microbiome sampling in the preterm neonate.

Despite advances in our understanding of the human microbiome, there exist significant knowledge gaps in our understanding of the skin microbiome of the preterm neonate. Herein, we describe skin microbiome sampling of six preterm neonates at multiple timepoints, and compare the skin microbiome samples to environmental (crib/isolette swabs) and negative controls. Samples of the same type (skin, crib, control) were more similar than when compared by week or by patient.