Purposeful Language Exposure by Neonatal Nurses and Caregivers in the NICU : Advances in Neonatal Care

Secondary Logo

Journal Logo

Neonatal Evidence-Based Review

Purposeful Language Exposure by Neonatal Nurses and Caregivers in the NICU

Newnam, Katherine M. PhD, RN, CPNP, NNP-BC; Muñoz, Lauren R. MSN

Editor(s): Gephart, Sheila PhD, RN, Section Editors; Newnam, Katherine PhD, RN, NNP-BC, CPNP, IBCLE, Section Editors

Author Information
doi: 10.1097/ANC.0000000000000833
  • Free


Worldwide, 15 million infants are born preterm each year. In the United States about 500,000 preterm infants are born annually, with the healthcare cost estimated at $68 billion.1 Efforts continue toward reducing the incidence of preterm birth in the United States particularly in the Southeastern United States, where rates of preterm birth are highest per capita.1 Significant technological and therapeutic advancements as well as improved medical and neurodevelopmental care in the neonatal intensive care unit (NICU) have shown positive results in both the mortality and morbidity of this population.2 Survival rates of infants born extremely low birth weight (ELBW) and/or less than 28 weeks' gestation have markedly improved. For example, survival of an infant born at 25 weeks' gestation has increased from 39% to 80% over a 25-year period.3 Yet, preterm complications, including death and disability, remain highest for the ELBW and the very low-birth-weight infant, whose birth weights are ≤1000 and ≤1500 g, respectively.4 Neonatal researchers and providers now pivot to discover interventions thought to improve neurodevelopmental outcomes.

To accomplish improved outcomes, the NICU environment, parental engagement, and clinical practice decisions must be reimagined to provide best practices for these fragile infants. Healthcare providers (HCPs) in the NICU provide interventions known to improve neurodevelopmental outcomes such as positioning, promoting quiet times through environmental noise reduction, clustering activities, and supporting breastfeeding practices. The bedside registered nurse (RN) is the most consistent member of the neonatal healthcare team, providing continuous patient monitoring, assessment, and physical care. Additionally, bedside RNs are in a unique position to role model methods of parental engagement including skin-to-skin holding, touch, and infant-directed speech (Table 1) by parents. The purpose of this evidence-based brief is to evaluate the current evidence to determine what is known about the characteristics of HCP communication to infants in the NICU.

TABLE 1. - Key Definitions
Descriptive study Research that is designed to closely observe and describe what happens in a situation; the researcher does not attempt to change the situation in any way
Expressive language An infant's ability to make sounds or words to communicate with others
Healthcare providers Members of the neonatal care team, including nurses, physicians, physical therapists, speech pathologists, etc
Health disparities Unequal share of illness or disability by some groups compared to others
Infant-directed speech Spoken language communicated directly and intentionally to an infant; also called language input
Language exposure Words or sounds spoken within earshot of the infant, whether directed to the infant or others; for example, a nurse speaking with a parent in the room while she assesses the infant
Language nutrition Starting while in utero and continuing throughout childhood, language of sufficient quantity and quality to feed a child's development5
Multimodal Composed of multiple interventions; for example, the auditory, tactile, visual, and vestibular intervention6
Neuroatypical Abnormal neurologic function or structure
Neurodevelopmental care Care that supports the normal systematic development of the central and peripheral nervous system
Open-bay units Neonatal intensive care units with large rooms/spaces housing multiple infants
Receptive language An infant's ability to understand words or sounds spoken to him or her
Single-family rooms Neonatal intensive care unit rooms that are designed to house only one infant and the infant's family


One major morbidity associated with preterm birth is recessive and expressive language delay (Table 1).7 Language outcomes are correlated with gestational age and birth weight and inversely correlated with the level of brain injury.8 Research has shown that the structures of the auditory system are formed by about 25 weeks' gestation, and the fetus begins the development of cognitive pathways related to language recognition during the final 15 weeks of pregnancy.9,10 Typical sounds heard by the fetus prior to birth include environmental sounds, human voices, and music. The repetition of those sounds during this timeframe allows for early recognition as the fetal brain begins to imprint the most commonly heard sounds, such as parents' voices.11 Even as early as term delivery, infants have been found to turn toward their parents' voices, clearly preferring them over other voices in the room.9,11,12

Early cognitive pathways are crucial first steps needed for infants to later process language and begin speaking.10 While in the NICU, preterm infants may not receive the critical language input believed to be an instrumental component of language acquisition.13,14 Healthcare providers, including bedside RNs, may have a unique opportunity to provide this important neurodevelopmental stimulus through infant-directed speech. Speaking to the infant during routine care promotes auditory function,12,15 reduces neonatal stress response,9,16,17 and may role model essential speaking behaviors to parents. Exploring language input from caregivers to neonatal patients is an important step in understanding the use of this widely available, low-cost intervention to promote neonatal language exposure during NICU hospitalization. In this evidence-based practice brief, the literature was explored to determine what is currently known and what knowledge gaps exist surrounding verbal communication of HCPs directed to infants hospitalized in the NICU. To answer the clinical question “What is known about the characteristics of HCP communication to infants in the NICU?” both seminal and contemporary literature were searched.


Data Retrieval

The Cumulative Index to Nursing and Allied Health Literature (CINAHL), PubMed, PsycINFO, and Web of Science databases were used in the initial search for relevant literature. No data limitations were applied to searches, but sources were excluded if they were not in English. Only empiric, primary research articles were included in this review. No unpublished, gray, or secondary literature was utilized, including dissertations/theses, expert opinion pieces, commentaries, literature reviews, and conference proceedings.

Search terms were developed with the assistance of a health sciences research librarian. The following blocks of search terms were connected with the Boolean operator AND:

  • Block 1: (“healthcare providers” OR “health care providers” OR doctor* OR physician* OR nurs* OR nurse practitioner* OR therapist*) AND (communicat* OR word* OR verbal* OR spoke* OR speak* OR speech OR talk* OR language OR linguistic OR auditory OR vocal*)
  • Block 2: “neonatal intensive care” OR “NICU” OR “neonatal intensive nursery”
  • Block 3: (newborn* OR neonat* OR preterm OR premature* OR “high risk”) AND (infant* OR baby OR babies OR neonate*)

The final run of database searching was completed on March 29, 2020. The details of searching and selection of sources can be seen in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) diagram18 in Figure 1. In total, 2307 sources were found using databases, including 719 from CINAHL, 946 from PubMed, 243 from PsycINFO, and 399 from Web of Science. After duplicates and non-English sources were removed, 1351 remained. Titles and abstracts were examined by the second author for relevance, and 1239 were excluded because they did not pertain to health caregiver communication with infants. The second author examined the full text of 112 sources and excluded 101 articles for the reasons seen in Figure 1, leaving 11 sources for inclusion. After both authors read through the included articles a second time, however, 6 additional articles were excluded.

PRISMA diagram. HCP indicates healthcare provider; NICU, neonatal intensive care unit.

Forward searching by examining the reference lists of the 6 included articles from database searching was also employed to find relevant sources. One hundred sixty-one source titles were examined, and abstracts were assessed if titles seemed applicable to the research question. After duplicates and irrelevant sources were removed, 17 full-text sources were read, and an additional 3 articles were included for the literature review.

Data Extraction

The second author gathered the initial cadre of included articles, and both authors discussed which should remain in the inclusion group and which should be excluded from the analysis. The authors agreed to exclude articles that did not explicitly respond to the purpose of the review: “what is known about the characteristics of HCPs' communication to infants in the NICU.” Research that employed infant-directed speech only by the researcher or research assistants was not included in the formal analysis.6 Therefore, articles that made only brief mention of infant-directed speech by HCPs within the context of multimodal developmental interventions were also excluded.19 The authors were surprised by the age of many of the articles pertaining to the research question and decided not to pursue articles prior to 1980, as NICU environments and care delivery have vastly changed in the last 40 years.3

On the second full-text reading of each source, the second author extracted data and placed them into rows of a matrix following Garrard's Matrix Method (Table 2).20 Both authors independently appraised the quality of studies based on Sirriyeh et al's21 Quality Assessment Tool for Studies with Diverse Designs (QATSDD), a validated 16-item tool for examining a variety of studies, including qualitative, quantitative, and mixed-methods studies. Following independent scoring, a discussion allowed authors to reach consensus of each study's score (Table 2). Regardless of quality score, each study's data were utilized for the review although more emphasis was given during the discussion section to studies with higher scores. Overall, the QATSDD scores demonstrated the wide variety of articles and the fact that many were lacking important components of high-quality reports, representing a continued gap in the scientific evidence.

TABLE 2. - Literature Review Matrix
Author: Quality Score, Purpose Setting, Participants Design, Methods Findings
Marton et al22: 23/42

To quantify the amount and types of care low birth-weight infants receive during routine care in the NICU
The Hospital for Sick Children in Toronto, Canada

Convenience sample of 10 premature infants (<1500 g; 5 in the NICU and 5 in the postintensive care unit) without serious medical conditions
Study design: descriptive quantitative
  • Random observations in 80-min intervals

  • Total 12 h of observations from 0900 to 2100

  • Coding of observations included 12 categories

  • “Talk to infant” (p. 311) was 1 of 12 categories

  • Infants received contact with hospital staff an average of 8 min every hour

  • No difference was reported between NICU and the postintensive care unit

  • Postintensive care unit infants received greater amounts of soothing touch

  • NICU infants had more procedural contact compared to the postintensive care unit

  • Comparatively, infants were talked to 1.9 min/h in the NICU and 1.5 min/h in the postintensive care (nonsignificant difference)

Gottfried et al23: 12/42

To explore the amount, quality, and pattern of stimulation received by neonates in the neonatal nursery
Los Angeles County University of Southern California Medical Center Women's Hospital

Convenience sample of infants in the NICU and convalescent units (number unclear) with 846 NICU social observations and 705 social observations in the convalescent unit
Study design: descriptive quantitative
  • Three 1-min observations every hour for 3 d

  • Coding of observations included a category called “talking when in contact with an infant” (p. 673)

  • Talking to infants was more common in the NICU vs convalescent unit (13% vs 7.9%; χ2 ≥ 8.56; P < .001)

  • No temporal pattern was seen for any stimulation type

  • Combined tactile and verbal stimulation was infrequent

  • Both talking alone and combined talking and gazing at infants occurred mostly during handling (9.6% and 4.8%, respectively in the NICU; 7.2% and 5.5% in the postintensive care unit) and during procedural care in the NICU (7.9% and 3.8%, respectively)

Solberg and Morse24: 16/42

To describe comforting behaviors of neonatal staff during neonates' distress

Convenience sample of 4 male, full-term neonates who received major chest and caregivers of neonates (nurses, laboratory technicians, physicians, and parents)
Study design: descriptive quantitative
  • 40 h of videotaped interactions during routine postoperative care

  • Periods of distress and caregiver and infant interactions were described

  • Vocal comfort was given in 39% of comforting episodes

  • 34% of vocalizations occurred with touch

  • Vocalizations were usually brief, repeated words of sympathy, particularly during procedures

  • Nurses were more apt to engage in tactile, rather than vocal, comfort

  • Many nurses conversed with colleagues while providing physical comfort to infants who were silently crying

  • No proactive comforting occurred before painful experiences

Morse et al25: 22/42

To examine and describe the interaction between NICU caregivers and neonates
Large tertiary care hospital

Convenience sample of 4 male, full-term neonates who received major chest and caregivers of neonates (nurses, laboratory technicians, physicians, and parents)
Study design: descriptive quantitative
  • 40 h of videotaped interactions during routine postoperative care

  • Types of interactions were categorized

  • Touch was the most common means of communication between nurses and neonates

  • 33.5% of touch was accompanied by vocalizations to soothe the infant

  • 46% of all types of interactions occurred because of the neonates' behavioral cues (eg, crying)

  • 81.2% of all interactions yielded no change in neonates' state, but 15.7% improved their state

  • Vocalizations from caregivers were infrequent

Pinheiro et al26: 26/42

To understand NICU nurses' experiences of interacting with newborns and their families and to describe communication techniques nurses use to promote baby–family bonding
Public hospital in Sao Paulo, Brazil with 32 beds

Theoretical sample of 3 nurses and 4 nursing auxiliaries
Study design: grounded theory
  • Video-recorded observations of interactions between staff and infants and family

  • Semistructured interviews with staff

  • Staff spoke to infants because of their dependent situation and because they liked caring for them

  • Staff described taking on maternal roles, including talking and playing with infants, when mothers were absent

  • Nonprocedural caring techniques were deemed important for infants' development

  • Nursing staff were motivated to help parents bond with their infants by teaching them how to interact

  • When work was more stressful, staff described cutting back on interactions, only caring for physical needs

Saito et al27: 25/42

To assess for differences in neonatal cerebral blood flow when exposed to female nurses' voices versus mothers' voices
Prefectural Hospital of Hiroshima in Hiroshima, Japan

Convenience sample of 26 Japanese infants (13 boys, 13 girls) without neurologic injury who were 18-81 days' old with mean gestational age of 30.42 wk and mean birth weight of 1385.42 g
Study design: quasiexperimental
  • Recorded samples of mother's and primary nurses' voices given 3 times repeatedly over 10 s

  • Near-infrared spectroscopy brain imaging used to determine infant responses

  • Mothers' and nurses' voices both activated the left frontal lobe, but activation by mothers was greater

  • Nurses' voices activated the right frontal lobe

Efe et al28: 22/42

To describe Turkish nurses' understanding of neonatal pain assessment and use of nonpharmacologic pain management techniques
15 university hospitals in Turkey

Convenience sample of 111 pediatric surgical nurses
Study design: descriptive quantitative
  • Questionnaire to assess knowledge of neonatal pain assessment and techniques for relieving pain

  • Only 52.3% of nurses identified infants' behavioral cues (eg, crying and grimacing) to pain

  • 50.5% used both behavioral cues and physiologic indicators to assess for pain

  • Avoiding loud talking near the infant was one means of providing pain management

Guillaume et al29: 28/42

To explore the experiences and expectations of parents of NICU infants in terms of bonding with their infants and caregiver behavior
3 tertiary hospitals in Paris, France

Convenience sample of 30 mothers and 30 fathers (10 from each site)
Study design: descriptive qualitative
  • Semistructured interviews

  • Discourse analysis

  • Nurse behaviors of talking to and caring affectionately for infants helped parents trust and increased parents' ability to bond with their infant

Abbreviations: NICU, neonatal intensive care unit; NR, not reported.

Data Analysis

The Matrix Method20 was utilized in the initial analysis of data. Comparing and contrasting data found in columns provided beginning classification and themes. The second author identified the emerging themes, and these were refined and validated by the first author. The literature matrix, including key findings, was organized by the authors for ease of review and reference (Table 2).


Overview of Studies

In total, 8 studies conducted over a 33-year period (1980-2013) were analyzed (Figure 1). Solberg and Morse24 and Morse et al25 completed 2 studies utilizing the same observational data. The earliest articles were most relevant to the research question.22–25 Later articles focused more on infant–family bonding and how nurses play a role in facilitating this bond.26,27,29 Efe et al28 examined verbal communication to infants in the context of pain management. Only Pinheiro et al26 utilized a theoretical framework: symbolic interactionism.

Studies of verbal communication between HCPs, including RNs and neonates in the NICU, spanned the globe with 1 study each from Brazil,26 Japan,27 France,29 Turkey,28 Canada,22 and the United States.23 Two studies did not specify the geographical location.24,25 Most research was conducted at a single hospital, but Efe et al28 and Guillaume et al29 examined samples from multiple hospitals.

Sample sizes were generally small and ranged from 4 to 60 people, though Gottfried et al23 used the number of observations (n = 846) as the sample; the number of participants in their study was unclear. Sample composition varied significantly between studies. Three studies utilized infants as the research participants.22,23,27 Samples for Pinheiro et al26 and Efe et al28 consisted of nursing staff, while Guillaume et al29 conducted research with parents to report caregiver behavior including language use. Morse et al25 and Solberg and Morse24 utilized a variety of participants, including infants, NICU nurses, laboratory technicians, physicians, and parents.

Studies were overwhelmingly descriptive, most with quantitative methods. Only Saito et al27 utilized a quasiexperimental design. Early studies gathered data via observations only, either in-person or with a video camera.22–25 Pinheiro et al26 combined recorded observations of infant–nursing interactions with semistructured interviews with NICU nursing staff. Guillaume et al29 used semistructured interviews with parents.

Seven out of 8 studies employed convenience samples, and half of the qualitative studies had no explicit methodology. However, Pinheiro et al's26 study was an exception with both theoretical sampling and grounded theory methodology. Additionally, Morse et al25 and Solberg and Morse24 conducted their studies via the lens of ethology. Quantitative analysis occurred through descriptive,24,25,28 parametric,23 nonparametric,22,23 and inferential statistics.2,25,27 Morse et al25 also utilized regression analyses. Qualitative analysis ranged from coding26 to discourse analysis.29

Quality of Included Studies

To conduct a rigorous review of current evidence, researchers must appraise the quality of evidence cited. There are numerous tools available to assess quality, but many are best suited for specific research designs, such as qualitative or quantitative methodologies. Tools that enable researchers to evaluate evidence from mixed-methods studies or a variety of qualitative and quantitative designs are less prevalent. For this evidence-based review, which encompassed both quantitative and qualitative studies with various designs, the authors selected the QATSDD21 to assess the quality of each article included. This 16-item quality assessment tool was developed at the University of Leeds in the United Kingdom and has been embraced by researchers in the social sciences for ease of use, content validity, and interrater reliability (κ= 0.698-0.901).21 The authors scored each article individually, discussed their scores together, and developed a consensus (Table 2).

HCP–Infant Verbalizations Themes

Each study in this review provided a slightly different perspective on provider–infant verbal communication. In reviewing the results of each study, the frequency, context, perception, and inhibitors of verbal communication between NICU staff and infants became apparent. For ease of review, findings have been arranged by topical subheadings listed next.

Frequency of Infant-Directed Speech

Overall, descriptive findings show that NICU providers' infant-directed speech is limited and could arguably be classified as rare.22–25 Marton et al22 reported that infants had contact of any kind (eg, touch and voice) with neonatal staff only an average of 8 minutes every 1 hour. Infants in the NICU were spoken to 1.9 minutes per hour, and those in the postintensive care unit received only 1.5 minutes of directed speech per hour. Differences in overall contact and speech between the 2 units were not statistically significant.

Gottfried et al23 also found infant-directed speech to be uncommon. Over 92% of recorded observations in the NICU involved speech within earshot of the infant, but only 13% consisted of infant-directed speech. Differences in the convalescent unit were not quite as dramatic but were still sizable, with nearly 70% of observations involving talking, but only 7.9% of talking directed toward the infant. Morse et al25 also described vocalizations toward infants as infrequent. Instead, touch was a more commonly used means of comforting infants, with speech being combined with touch in only 33.5% of interactions. Speech alone accounted for only 3% of interactions.24

Gottfried et al,23 like Marton et al,22 found that infants in the NICU received more directed speech and procedural care than those in the convalescent care unit (13% vs 7.9% and 15.5% vs 3.8%, respectively), and these differences were significant in their study (χ2≥ 8.56; P < .001). They discovered that touch was more frequently employed in the convalescent unit (5.8% vs 0.02%). However, contrary to Marton et al,22 Morse et al,25 and Solberg and Morse,24 infant-directed speech alone was more prevalent than occasions of combined tactile and verbal stimulation (13% vs 1.7% in the NICU and 7.9% vs 3.7% in the convalescent unit). Talking to infants was also more common than combined visual (ie, trying to make eye contact with the infant) and verbal stimulation (13% vs 7.1% in the NICU and 7.9% vs 5.4% in the convalescent unit).

Context of Infant-Directed Speech

As mentioned earlier, the infant's location significantly determined the amount and types of contact infants had with their providers. Gottfried et al23 reported that infants in the NICU were spoken to in 13% of observations, and they described verbal communication as occurring predominantly during infant handling (9.6% of observations) and during procedures (7.9%) compared to social touching (1.7%), bottle feeding (1.1%), and rocking (0.1%) in the NICU. In the convalescent unit, however, less infant-directed speech was observed overall (only 7.9% of observations), but talking was seen most frequently during handling (7.2%), social touching (3.7%), and bottle feeding (2.8%). Infant-directed speech during procedural care and rocking occurred in only 2.9% of observations combined.23

Some researchers found that infant-directed speech occurred more commonly in combination with other forms of communication, particularly touch. Solberg and Morse24 found that caregiver touch and speech occurred together 34% of the time. Yet, they also noted that vocalizations by staff were oftentimes directed to other staff in the room, even as comforting touch was provided to the infant. Counter to Solberg and Morse's24 findings, Gottfried et al23 observed that speech alone was more common than speech combined with touch. However, these 2 studies had different contexts. Solberg and Morse24 observed caregiver interactions with infants who had just undergone surgery, while Gottfried et al's23 participants were a heterogeneous mix from the NICU and convalescent units. Solberg and Morse24 surmised that postoperative care of intubated infants may have rendered providers more likely to use touch rather than vocal comforting behaviors, since infants were themselves unable to vocalize in an audible manner.

Pain was also a significant contextual factor for infant-directed speech. Marton et al22 reported that infants in the postintensive care unit experienced more soothing interactions with staff, while those in the NICU received care more often related to painful procedures. Morse et al25 noticed that nearly half (46%) of postoperative interactions of any type were triggered by infants' behavioral cues, such as crying. Solberg and Morse,24 examining the same recorded observations, found that infant-directed speech was often very brief and repetitious, usually an expression of sympathy such as “I'm sorry, sweetie” (p. 106). Efe et al28 also described caring interactions in relation to infants' painful experiences. In this study, little more than half (52.3%) of nurses were able to correctly identify infants' pain cues, and only 50.5% stated they use both these behavioral cues and physiologic indicators to assess for pain. No mention of infant-directed speech was given, but nurses did state that they frequently avoided loud talking near the infant and held infants only minimally as means of providing nonpharmacological pain relief.

Perception of Infant-Directed Speech

Nursing staff interviewed in Pinheiro et al's26 study stated that they were motivated to speak to infants, play with them, and interact in other loving ways because of infants' dependent situations and because they genuinely like caring for them. Indeed, NICU staff referred to themselves as temporary mothers to the infants in their care, particularly when parents were not on the unit (such as when visiting hours were limited). These staff members described supporting the infants with activities consistent with “mothering” to “develop properly.” Additionally, nurses and nurse auxiliaries believed these nonprocedural interactions with infants would teach parents how to interact with their infants, increasing the parent–infant bond and making parents feel more comfortable in the unit.26

Saito et al27 examined premature infants' perceptions of their mother's voice compared to their primary nurse's voice and found that both voices activate the left frontal lobe, though the mother's voice provided greater stimulation. Interestingly, only the nurses' voices activated the right frontal lobe in addition to the left based on near-infrared spectroscopy (NIRS) readings. The authors suggest that this may indicate that infants receive positive cerebral stimulation to the right hemisphere from nurses, but language development (typically associated with the left hemisphere) may be better promoted by mothers' voices. The activation of the right frontal lobe by the primary nurse's voice, they reason, may be indicative of a stress response upon hearing the voice of the nurse, which may become associated with painful experiences.

Parents' perceptions of infant-directed speech were described by Guillaume et al.29 Parents stated that some nurses talked to their infants and displayed affectionate care, doing more than just going through the motions. Both mothers and fathers identified these loving actions by NICU nurses as helpful in increasing parents' trust of the NICU team and in facilitating parent–infant bonding.

Inhibitors of Infant-Directed Speech

Although Pinheiro et al26 described loving interactions by nursing staff and cited their willingness to take on a more maternal role for infants, they also described inhibitors of such care. Nurses and nurse auxiliaries stated plainly that they often became less interactive with infants and families when their workloads increased or during tense moments on the unit. “In this circumstance, they no longer relate with the newborn, with whom they had established bonding, but interact with him only to attend his physical needs” (p. 1016). The findings by Efe et al28 point to another inhibitor of infant-directed speech: the emphasis on noise reduction in NICU environments. They argue that maintaining a quiet culture may inadvertently inhibit vocalizations toward infants by NICU staff, as they may see such sounds as harmful, stressful, or even painful to the infant.

Importance of Infant-Directed Speech

Solberg and Morse24 found that proactive comforting, either via touch or speech, was not provided to infants prior to a painful procedure. The authors note the missed opportunity that comforting words and other communication can provide to potentially ease the stress experienced by infants in the NICU. Morse et al,25 using the same observational data, also noted that 15.7% of caregiver comforting behaviors appeared to improve the infant's state, while 81.2% of interactions yielded no change in the infant's state and 3.1% actually worsened the infant's disposition. They were surprised that caring touch of postoperative infants occurred most frequently when the infants were asleep (38.6% of comforting interactions) and least when infants were wide awake (only 9%). Mothers and fathers of premature infants interviewed by Guillaume et al29 described nurse–patient interactions and the importance of affectionate caregiving as a way for nurses to build trust with the parents and model and encourage bonding behaviors between parents and infants.


In this review of literature, the authors sought to uncover the characteristics of infant-directed speech by HCPs in the NICU. RNs, the most consistent HCP, were the primary focus of much of the research reviewed on this topic. Overall, findings reveal that interaction between NICU staff and infants is a topic that has been investigated by researchers worldwide and over several decades. Studies from North America,22,23 South America,26 Europe,29 West Asia,28 and East Asia27 demonstrate the international concern for this topic among neonatal scientists. Yet, despite this research topic's wide footprint, the authors were surprised by the dearth of evidence surrounding HCPs' infant-directed speech. Additionally, the importance of the topic has been underestimated. Empiric evidence demonstrates the correlation between early exposure to language and improved language outcomes in the preterm population.5,16,30,31 Studies that truly examine verbal interactions between HCPs and neonates are over 10 years old,26 with many more than 25 years old.22–25 Several recent studies have mentioned infant-directed speech merely as a side note in discussions on infant pain,28 parents' NICU experiences,29 and the unique role of maternal voice.27 Multimodal neurodevelopmental care in the NICU, which often involves light and noise reduction, touch, positioning, and speech, has proven beneficial.6 However, these interventions are most often studied in concert to determine outcome differences. Little examination has occurred to determine the isolated impact of language input to the neonate while hospitalized in the NICU.

Interestingly, following the redesign of NICUs to single-family room (SFR) environments, researchers described poorer neonatal language outcomes, highlighting the lack of verbal stimulation by HCPs including nurses.15,30 A general lack of language exposure in the NICU was cited as a possible reason, with outcomes worse in infants with less family visitation.31 These study results mirrored the global lack of infant-directed language reported by Gottfried, Morse and Solberg.22–25 Pineda et al30 demonstrated the effect isolation can have on neonates in private rooms, as infants who received less language input experienced detrimental neurodevelopmental outcomes. Compared to infants cared for in NICUs with open-bay units, those in SFRs had reduced cerebral maturation identified by magnetic resonance imaging and electroencephalography. These infants also had worse language scores on the Bayley III test at 2 years of age, demonstrating possible long-term effects of SFRs. Pineda et al30 hypothesized that infants in SFRs without frequent parental involvement are at risk of social isolation and sensory deprivation.

Pineda et al30 also described the lack of verbalization by neonatal HCPs, including nursing staff, although this was outside the primary aim of the study. Efe et al28 touched on this as well when they reported that nurses often intentionally avoid talking near infants in the NICU as a means of pain relief. General lack of language exposure has significant impact to preterm infants, who are already known to be at risk of speech and language delays. Since the average length of stay in the NICU for the ELBW infant delivered at 24 weeks' gestation is about 23.4 weeks,1 the preterm infant may be missing out on as much as nearly 6 months of “language nutrition” (Table 1).5

The importance of family interactions on language exposure and functional auditory development cannot be understated, and 3 articles in this review highlighted the crucial role family plays in providing infant-directed speech.26,27,29 Nevertheless, as Guillaume et al29 described in their qualitative study, parents often have an expectation that HCPs will not only provide physical care to their patients but will also speak to them. Using NIRS technology, Saito et al27 demonstrated that infants respond to both mothers' and nurses' voices differently, evidence that cognitive pathways develop during NICU hospitalization.10,32 Findings of the preliminary studies of SFRs further underscore the influence of language input by hospital staff, particularly for infants missing interactive family experiences.30,31 HCPs, including nurses, can support an identified void of language exposure during this critical developmental period.9

Unfortunately, factors exist that work against infant-directed speech from HCPs. When NICU nursing staff are overwhelmed with a large workload or are otherwise stressed, they are less prone to spend the extra time and energy to speak with the infants in their charge.26 Additionally, significant emphasis on noise reduction in NICUs has dampened—perhaps unintentionally—HCP–infant vocalizations, as many staff believe that quiet is best for premature infants.28


Role modeling language behavior in the NICU may support improved patterns of infant-directed speech following discharge. Evidence shows infants who receive increased language exposure, termed language nutrition, during infancy and early childhood have higher cognitive and language scores on validated developmental testing.5 As frontline healthcare workers, bedside RNs are in a prime position to show parents and other caregivers how to provide infants with the language nutrition required for optimal development.5

Yet, role modeling for parents is not enough. Many families encounter significant barriers to visiting their preterm infant regularly in the NICU. As long as some infants are visited frequently and others rarely, disparities will exist in NICUs because of the tremendous variability in the number of words infants hear during their hospitalization.15,31 However, HCPs can intervene and potentially lessen this disparity by providing the touch, eye contact, and infant-directed speech that preterm infants need when their parents are unable to be present.19 Nurses, in particular, can employ this simple but powerful tool to begin the process of language nutrition for at-risk neonates.


Globally, there is a dearth of literature supporting or refuting the efficacy of infant-directed speech as an intervention strategy to improve neonatal outcomes. Additionally, the authors found that the articles available on this topic were missing important components of high-quality research, as evidenced by their QATSDD scores.21 High-quality research that specifically seeks to understand the early and frequent influence of language exposure in the NICU could help explain variability in language outcomes in this population. Future research using predictive modeling might support the use of directed language as part of a multimodal approach to neurodevelopmental care of the preterm infant.


This analysis is limited by the specific inclusion criteria necessary to respond to the clinical question about HCP communication to infants in the NICU. This eliminated research that studied multimodal approaches to NICU care. Both contemporary and seminal works were included in the review, but the findings of older studies may no longer be applicable given the significant changes in NICU environments and neonatal care. Additionally, the low-quality scores of some articles constrain the authors' ability to draw firm conclusions about the characteristics of providers' infant-directed speech.


Direct communication between neonatal HCPs and neonatal patients remains essential. Indeed, communication with those in their charge is part of the essence of nursing. Yet, evidence demonstrates this critical part of neonatal nursing care has not been the focus of empirical research and may be lacking in practice. Multimodal approaches, such as auditory, tactile, visual, and vestibular interventions, have been examined for their protective capabilities, but investigations exclusively examining infant-directed speech are sparse.6,33,34

The preterm infant is considered neuroatypical and is at major risk for developmental disability, specifically language delay.31 Delay in language is directly related to the degree of prematurity, with the youngest ELBWs at greatest risk.31 Infants need expressive and receptive language to connect with their parents, communicate their needs, and understand the world around them. Language delay has long-term consequences as well, as it has been directly linked to poor mental health outcomes and poor educational outcomes with implications to lower self-esteem.35

The NICU serves a specialized nonverbal population thought to be protected in the quiet, darkened, and often isolated bed spaces located in SFR environments. The redesign of NICUs to SFR configuration achieved several goals, including quieter rooms,36 increased privacy and spaces that supported family centered care,37 and improved breastfeeding38 and sepsis rates,37 though the latter was likely also influenced by concurrent evidence supporting human milk use in the neonate. One untoward consequence of the SFR configuration, however, is reduced neonatal exposure to language, demonstrating the lack of directed communication between HCPs and infants.15,30,31 This is particularly problematic for infants who are visited infrequently or rarely by family, and it serves to magnify the disparity between those infants who receive frequent language input from visiting family and those who lack language input because of decreased visitation. Missing language input from both clinicians and visitors could impact short- and long-term language outcomes.15 To promote optimal neurodevelopment outcomes for all infants hospitalized in the NICU, caregivers have the opportunity to provide language exposure during a critical time of auditory development. Nurses, the most consistent member of the healthcare team, play an important role in offering infant-directed speech to their patients. Additionally, nurses who role model this type of care to parents may positively influence language use in the home, affecting the long-term outcomes of children born preterm.

What we know:
  • Preterm infants have higher rates of language delay.

  • Single-family rooms have been correlated with a decrease in language exposure to NICU patients.

  • Seminal research reported very little infant-directed speech from neonatal caregivers including nurses.

  • Infant-directed speech is protective and developmentally appropriate for NICU patients.

What needs to be studied:
  • The current prevalence of verbal interactions between healthcare providers and infants in the NICU.

  • The barriers of neonatal staff vocalizations toward infants.

  • Ways to encourage staff to speak to the infants in their care.

What we can do today:
  • Begin speaking to infants routinely when providing care.

  • Role model infant-directed language when appropriate to parents and other family members.

  • Demonstrate how and when to speak to preterm infants.

  • Talk with parents about the importance of language development in the NICU and post-discharge.


The authors would like to thank Melanie Dixson, MSIS, AHIP, Library Services, University of Tennessee, Knoxville, for her assistance with database searching.


1. March of Dimes. 2019 March of Dimes Report Card. https://www.marchofdimes.org/materials/US_REPORTCARD_FINAL.pdf. Published 2019. Accessed June 30, 2020.
2. Kobaly K, Schluchter M, Minich N, et al. Outcomes of extremely preterm or extremely low birth weight children. Pediatrics. 2008;121(1):73–81.
3. Glass H, Costarino A, Stayer S, Brett C, Cladis F, Davis P. Outcomes for extremely premature infants. Anesth Analg. 2015;120(6):1337–1351. doi:10.1213/ANE.0000000000000705.
4. Hintz SR, Kendrick DE, Hohn BR, Poole WK, Higgins RD. Changes in neurodevelopmental outcomes at 18-22 months' corrected age among infants of less than 25 weeks' gestational age born in 1993-1999. Pediatrics. 2005;115(6):1645–1651.
5. Mahony AD, Zauche LH, Hallowell S, Weldon A. Leveraging the skills of nurses and the power of language nutrition to ensure a better future for children. Adv Neonatal Care. 2017;17(1):45–52. doi:10.1097/ANC.0000000000000373.
6. White-Traut RC, Nelson MN, Silvestri JM, et al. Developmental patterns of physiological response to a multisensory intervention in extremely premature and high-risk infants. J Obstet Gynecol Neonatal Nurs. 2004;33(2):266–275. doi:10.1177/0884217504263289.
7. Symington A, Pinelli J. Developmental care for promoting development and preventing morbidity in preterm infants. Cochrane Database Syst Rev. 2003;(4):CD001814.
8. Cristobal R, Oghalai J. Hearing loss in children with very low birth weight: current review of epidemiology and pathophysiology. Arch Dis Child Fetal Neonatal Ed. 2008;93(6):F462–F468. doi:10.1136/adc.2007.124214.
9. Graven SN. Sound and the developing infant in the NICU: conclusions and recommendations for care. J Perinatol. 2000;20(8, pt 2):S88–S93.
10. Graven SN, Browne JV. Auditory development in the fetus and infant. Newborn Infant Nurs Rev. 2008;8(4):187–193. doi:10.1053/j.nainr.2008.10.010.
11. Kisilevsky BS, Hains SMJ, Lee K, et al. Effects of experience on fetal voice recognition. Psychol Sci. 2003;14(3):220–224. doi:10.1111/1467-9280. 02435.
12. Guarini A, Sansavini A, Fabbri C, Alessandroni R, Faldella G, Karmiloff-Smith A. Reconsidering the impact of preterm birth on language outcome. Early Hum Dev. 2009;85(10):639–645. doi:10.1016/j.earlhumdev.2009.08.061.
13. Webb AR, Heller HT, Benson CB, Lahav A. Mother's voice and heartbeat sounds elicit auditory plasticity in the human brain before full gestation. Proc Natl Acad Sci USA. 2015;112(10):3152–3157. doi:10.1073/pnas.1414924112.
14. Moore JK, Linthicum FH Jr. The human auditory system: a timeline of development. Int J Audiol. 2007;46(9):460–478.
15. Caskey M, Stephens B, Tucker R, Vohr B. Adult talk in the NICU with preterm infants and developmental outcomes. Pediatrics. 2014;133(3):e578–e584. doi:10.1542/peds.2013.0104.
16. Filippa M, Devouche E, Arioni C, Imberty M, Gratier M. Live maternal speech and singing have beneficial effects on hospitalized preterm infants. Acta Paediatr. 2013;102:1017–1020. doi:10.1111/apa.12356.
17. Filippa M, Panza C, Ferrari F, et al. Systematic review of maternal voice interventions demonstrates increased stability in preterm infants. Acta Paediatr. 2017;106(8):1220–1229. doi:10.1111/apa.13832.
18. Moher D, Liberati A, Tetzlaff J, Altman DG, the PRISMA Group. Preferred Reporting Items for Systematic Reviews and Meta-Analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097. doi:10.1371/journal.pmed.1000097.
19. Weber A, Harrison TM. Reducing toxic stress in the neonatal intensive care unit to improve infant outcomes. Nurs Outlook. 2019;67(2):169–189. doi:10.1016/j.outlook.2018.11.002.
20. Garrard J. Health Sciences Literature Review Made Easy: The Matrix Method. 5th ed. Burlington, MA: Jones & Bartlett Learning; 2017.
21. Sirriyeh R, Lawton R, Gardner P, Armitage G. Reviewing studies with diverse designs: the development and evaluation of a new tool. J Eval Clin Pract. 2011;18(4):746–752. doi:10.1111/j.1365-2753.2011.
22. Marton PL, Dawson H, Minde K. The interaction of ward personnel with infants in the premature nursery. Infant Behav Dev. 1980;3:307–313. doi:10.1016/S0163-6383(80)80039-7.
23. Gottfried AW, Wallace-Lande P, Sherman-Brown S, King J, Coen C, Hodgman JE. Physical and social environment of newborn infants in special care units. Science. 1981;214(4521):673–675. doi:10.1126/science.7197393.
24. Solberg S, Morse JM. The comforting behaviors of caregivers toward distressed postoperative neonates. Issues Compr Pediatr Nurs. 1991;14(1):77–92.
25. Morse JM, Solberg SM, Edwards J. Caregiver-infant interaction: comforting post-operative neonates. Scand J Caring Sci. 1993;7(2):105–111. doi:10.1111/j.1471-6712.1993.tb00178.x.
26. Pinheiro EM, da Silva MJP, Angelo M, Ribeiro CA. The meaning of interaction between nursing professionals and newborns/families in a hospital setting. Rev Latino-am Enfermagem. 2008;16(6):1012–1018. doi:10.1590/S0104-11692008000600012.
27. Saito Y, Fukuhara R, Aoyama S, Toshima T. Frontal brain activation in premature infants' response to auditory stimuli in neonatal intensive care unit. Early Hum Dev. 2009;85(7):471–474. doi:10.1016/j.earlhumdev.2009.04.004.
28. Efe E, Dikmen S, Atlas N, Boneval C. Turkish pediatric surgical nurses' knowledge and attitudes regarding pain assessment and nonpharmacological and environmental methods in newborns' pain relief. Pain Manag Nurs. 2013;14(4):343–350. doi:10.1016/j.pmn.2011.08.003.
29. Guillaume S, Michelin N, Amrani E, et al. Parents' expectations of staff in the early bonding process with their premature babies in the intensive care setting: a qualitative multicenter study with 60 parents. BMC Pediatr. 2013;13:18. doi:10.1186/1471-2431-13-18.
30. Pineda RG, Neil J, Dieker D, et al. Alterations in brain structure and neurodevelopmental outcome in preterm infants hospitalized in different neonatal intensive care unit environments. J Pediatr. 2014;164(1):52–60. doi:10.1016/j.jpeds.2013.08.047.
31. Vohr B. Speech and language outcomes of very preterm infants. Semin Fetal Neonatal Med. 2014;19(2):78–83. doi:10.1016/j.siny.2013.10.007.
32. Lahav A, Skoe E. An acoustic gap between the NICU and womb: a potential risk for compromised neuroplasticity of the auditory system in preterm infants. Front Neurosci. 2014;8:381. doi:10.3389/fnins.2013.00381.
33. White-Traut RC, Nelson MN, Silvestri JM, et al. Developmental intervention for preterm infants diagnosed with periventricular leukomalacia. Res Nurs Health. 1999;22(2):131–143. doi:10.1002/(sici)1098-240x(199904)22:2<131::aid-nur5>3.0.co;2-e.
34. White-Traut RC, Nelson MN, Silvestri JM, et al. Effect of auditory, tactile, visual, and vestibular intervention on length of stay, alertness, and feeding progression in preterm infants. Dev Med Child Neurol. 2002;44(2):91–97. doi:10.1017/s0012162201001736.
35. Behrman RE, Butler AS, eds. Preterm Birth: Causes, Consequences, and Prevention. Institute of Medicine (US) Committee on Understanding Premature Birth and Assuring Healthy Outcomes. Washington DC: National Academies Press; 2007.
36. Shahheidari M, Homer C. Impact of the design of neonatal intensive care units on neonates, staff, and families. J Perinat Neonat Nurs. 2012;26(3):260–266. doi:10.1097/JPN.0b013e318261ca1d.
37. Lester BM, Hawes K, Abar B, et al. Single-family room care and neurobehavioral and medical outcomes in preterm infants. Pediatrics. 2014;134(4):754–760. doi:10.1542/peds.2013-4252.
38. Domanico R, Davis DK, Coleman R, Davis BO. Documenting the NICU design dilemma: comparative patient progress in open-ward and single-family room units. J Perinatol. 2011;31(4):281–288. doi:10.1038/jp.2010.120.

infant development; language acquisition; neonatal intensive care unit; preterm infant; spoken language

© 2021 by The National Association of Neonatal Nurses.