Socio-demographic factors related to parent engagement in the NICU and the impact of the SENSE program

Abstract

Background:

Early parent engagement in the neonatal intensive care unit (NICU) is important for both parent and infant mental health and for improving developmental outcomes. It remains unclear how different programs, such as the Supporting and Enhancing NICU Sensory Experiences (SENSE) program, may empower parents from various socio-demographic groups to engage in the NICU. An improved understanding could aid in individualizing interventions for those at the highest risk for health disparities.

Aims:

This exploratory study, which was part of a larger study, sought to explore 1) socio-demographic factors related to parent presence and engagement in the NICU and 2) if the Supporting and Enhancing NICU Sensory Experiences (SENSE) program related to increased parent presence and engagement among different socio-demographic groups.

Methods:

Seventy mother-infant dyads (≤ 32 weeks gestation) were randomized to SENSE programming (parent education and age-appropriate, positive sensory interventions for parents to conduct with their infants every day of hospitalization) or standard care after admission to the NICU. The amount of parent presence and participation in sensory activities was tracked using bedside logs, nursing records, and research team documentation.

Results:

Being married (p=0.048; p=0.01), having private insurance (p<0.001; p=0.01), and having fewer children (p=0.004; p=0.03) related to more parent presence and engagement. Parents who were Black had less presence and engagement in the NICU (p=.04; p=0.02). Participation in the SENSE program was related to more parent presence and engagement among younger mothers (p=.002; p=<0.001) and among parents living farther distances from the hospital (p<0.001; p=0.004).

Conclusion:

Programming, such as SENSE, can improve engagement among high-risk groups.

Introduction:

Early parent engagement in the neonatal intensive care unit (NICU) is important and relates to improved neonatal and parent outcomes. Aside from supporting and being a source of comfort to the infant during painful procedures, parents of preterm infants can engage in developmentally appropriate activities, such as skin-to-skin care and providing language exposure. These activities can improve parent-infant attachment, reduce stress, and improve infant developmental outcomes [1]. Although the importance of parent involvement in the NICU is well-understood, parents can experience challenges engaging in care.

There have been reports in the United States demonstrating that parents often are not present each day of hospitalization [1, 2]. Research has established that parents often want to be involved in their infant’s care but require better communication and education at frequent intervals to effectively do so [3, 4]. Parents’ perceived level of communication with medical staff impacts visitation rates and engagement in activities like skin-to-skin care [5]. To address this issue, different approaches have been created to improve communication, education, and increase parent-infant interaction and engagement in the NICU [6–8]. In addition, a few sensory interventions have been described that specifically aim to engage parents in providing stimulating activities for their infant(s) in the NICU, with reports of improved infant neurodevelopmental and maternal mental health outcomes [9–12]. The Supporting and Enhancing NICU Sensory Experiences (SENSE) program combines approaches to specifically promote parent presence and engagement by educating and empowering families to provide sensory interventions for their hospitalized infants.

The SENSE program is a protocol of evidence-based, developmentally-appropriate, and positive multimodal sensory exposures designed for parents to conduct with their preterm infant(s) every day of hospitalization [13, 14]. Through initial and weekly education on how to engage with the preterm infant and a description of appropriate types and doses of sensory exposures, parents are tasked with providing positive sensory experiences, such as skin-to-skin holding and reading, for their infants in the NICU. The hallmarks of the SENSE program are: defining parents as the most important part of the infant’s life during hospitalization and beyond, providing tasks that engage the parent with the infant, and using specific and evidence-based sensory activities each day of hospitalization to optimize the infant’s sensory environment [14]. Research has identified associations between SENSE programming and improved parent confidence, better infant neurobehavior, and more favorable communication outcomes [15, 16]. The SENSE program was developed for families who are at high risk for low parent engagement [1], however, the impact of the program on different types of families has not been explored.

Previous studies have evaluated the role of socio-demographic factors such as maternal age, education, race, marital status, distance from hospital to home, and number of children at home on parent engagement in the NICU [1, 5, 17, 18]. More parental presence has been observed in White mothers and those who are married, have greater family support, are older, have more education, and have fewer children [1, 17]. Findings related to the distance of the family home from the hospital and how it relates to parent presence have been mixed [19, 20]. While certain socio-demographics have been related to increased or decreased parental engagement in the NICU, it remains unclear how different programs, such as the SENSE program, may empower parents from various socio-demographic groups to engage in the NICU. An improved understanding could aid in individualizing interventions for those at the highest risk for health disparities.

For this study, we aimed to explore 1) socio-demographic factors related to parent presence and engagement in the NICU and 2) if the Supporting and Enhancing NICU Sensory Experiences (SENSE) program related to increased parent presence and engagement among different socio-demographic groups.

Methods:

This study was approved by the Human Research Protection Office of Washington University with a Ceded Review at the University of Southern California. Mothers provided written informed consent.

Study Design:

This study enrolled 70 very preterm infants born ≤32 weeks estimated gestational age (EGA), within the first 7 days of birth who were part of a randomized clinical trial on the impact of the SENSE program on infant and parent outcomes [16]. The original study was powered to investigate the effect of the SENSE program on infant outcomes (not to detect interactions between intervention groups and socio-demographic factors). We acknowledge that a larger sample size is required for investigations of interactions, compared to evaluation of a single factor, however have used this convenience sample to explore this important topic.

Consecutive admissions were recruited from the 125-bed level IV NICU at St. Louis Children’s Hospital (SLCH). Infants were excluded if they had a known or suspected congenital anomaly. Infants who were wards of the state, or became wards of the state after enrolling in the study, were also excluded. Infants who expired or were transferred to another NICU were also excluded. While the NICU at SLCH had beds in an open ward as well as within private rooms, only participants in private rooms were enrolled, due to the significant variation in sensory exposure among infants in the open ward and the confounded experience of being a parent in an open ward environment [21]. The study site NICU is open to parents being present 24 hours a day, and parents can stay with their infant overnight within the private NICU room. For infants who were part of a multiple birth, only one infant was included as part of the parent-infant dyad to prevent repetition of data for this study. Among multiples, the first infant born was included as part of the parent-infant dyad except when that infant expired or was too sick for sensory exposures during the study, in which case, the second infant born was included as part of the parent-infant dyad.

Infants were randomized into one of two groups, the SENSE group or the standard of care group. Parent presence in the NICU and parent engagement in providing sensory exposures to their infants were tracked with bedside logging sheets starting immediately after enrollment. Socio-demographic information was extracted from the electronic medical record, and parents also completed a questionnaire to document additional socio-demographic factors prior to NICU discharge.

SENSE Program:

Parents of infants in the SENSE group received education early in hospitalization, usually starting within the first 4 days of birth, with specific instruction on how and when to provide sensory exposures to their infant(s). Recommended doses of sensory exposures were updated at least weekly, with advancing post-menstrual age (PMA), based on the SENSE program. Parent education included how to individualize sensory exposures based on the infant’s cues. Previous research identified that parents received an average of 4.8 educational sessions, amounting to an average of approximately 1.25 total hours of education over the course of NICU hospitalization [22].

The SENSE program was developed based on a rigorous and systematic process that included an integrative review, combined with expert opinion and parent input [13, 14]. The SENSE program includes specifically-timed and defined amounts of auditory (reading/talking/singing to the infant), visual (dim environment, cycled lighting), tactile (skin-to-skin care, holding, massage, gentle human touch), vestibular (rocking), kinesthetic (free movement) exposures, and olfactory (scent cloth, close contact with parents) that are to be provided to the infant each day of hospitalization [14]. The manualized SENSE program consists of doses that are tailored to the PMA of the infant, with the amounts of each sensory exposure increasing appropriately across the trajectory from preterm birth until term equivalent age. At term equivalent age, a minimum of 3 hours of tactile, 3 hours of auditory, 16 minutes of kinesthetic, 7 minutes of vestibular, and tailored visual and olfactory experiences are recommended [14]. Specific amounts and timing of interventions are tailored to the current medical status and preferences of the family, and parents are educated on how to read and respond to infant cues when using the program. The SENSE program is intended to be conducted by parents, when available, and by volunteers when the parents are unable to be present in the NICU. Parents in the SENSE group were provided weekly education and daily support to engage them in providing sensory-based interventions to their infants across the length of hospitalization. A sensory support team, consisting of occupational therapy graduate students, provided sensory exposures when the parents were unable to.

Standard of Care Group:

Parents in the standard of care group were supported by the medical team to engage in care and provide sensory exposures. However, they did not receive the same early, specialized education on how to facilitate sensory experiences. They also did not have the specific dosage recommendations across PMA that are part of the SENSE program.

Socio-demographics:

Socio-demographic information was extracted from the electronic medical record. In addition, 1-2 weeks prior to discharge or at 40 weeks PMA, depending on whichever came first, parents completed a questionnaire to document additional socio-demographic factors. The questionnaire took approximately 15-20 minutes to complete. Parents were encouraged to complete the questions online as a REDCap Survey, but a written version of the questionnaire was used when parents requested it or did not complete the online survey. The following maternal socio-demographic factors were documented: maternal age, maternal race (categorized as Black or not Black), insurance type (public or private), number of children outside the hospital, education level (college degree or no college degree), marital status (legally married or single), and distance of the family home from the hospital in minutes (calculated using Google Maps, based on parent-defined residence). For the purpose of obtaining consistent socio-demographic data, information from only the biological mother of the infant was collected.

Infant factors:

Infant factors that were extracted from the electronic medical record included infant sex, EGA at birth; APGAR scores at 1 minute and 5 minutes; days of endotracheal intubation and days on non-invasive mechanical ventilation; whether the infant had necrotizing enterocolitis, sepsis, patent ductus arteriosus, or cerebral injury (Grade III-IV intraventricular hemorrhage or periventricular leukomalacia defined by cranial ultrasound or MRI); and length of stay and PMA at discharge. These factors gave context to the severity of illness of the cohort and the context for parent presence and engagement.

Bedside Logging of Parent Presence and Engagement:

For both the SENSE group and standard of care group, log sheets were placed at the infant’s bedside at enrollment to enable documentation of type (e.g. auditory, tactile, visual) and timing (date, time, and duration) of sensory-based interventions across each day of hospitalization, as well as to document who the intervention was conducted by (parent, member of research team, sensory support team member, or other caregiver). Bedside logs were designed to be completed by parents as well as health care team members and members of the sensory support team. Bedside log sheets were supplemented by information from the research team and nursing documentation that defined parent engagement in sensory experiences with each infant. Presence and engagement were calculated based on either parent being present in the NICU (for presence) and all sensory stimulation that the infant received from either parent (for engagement).

Parent Presence:

Parent presence was calculated as the average number of days either parent were present per week (0-5), not including weekends, over the length of hospitalization. Weekend days were not included due to research staff not being available to ensure consistent documentation. For descriptive purposes, high presence was defined as either parent being present an average of 4 or 5 days per week, out of the 5 weekdays. Low presence was defined as either parent being present an average of 0 to 3 days per week, out of the 5 weekdays.

Parent Engagement:

Parent engagement was the primary outcome variable, and for the purposes of this study, parent engagement was calculated based on the parent’s (either parent) completion of auditory and tactile sensory exposures, captured using the parent logs contained in the infant’s room; research team documentation from observations made during the study; and nursing notes that detailed how long parents were present in the NICU and whether they engaged in providing positive sensory experiences to their infants during that time. High parent engagement was defined as parents administering 90% or more of the minimum recommended doses of auditory and tactile exposures for the infant’s PMA during each weekday, as defined in the SENSE protocol [14]. Low engagement was defined as conducting less than 90% of the defined auditory and tactile exposures at each PMA during each weekday.

Statistical Analyses

IBM SPSS Statistics 24 (IBM Corporation, Armonk, New York, 2016) was used for statistical analysis. The relationship between socio-demographic factors and parent presence and engagement, as well as the interaction of group (SENSE vs. standard of care) randomization on presence and engagement for different socio-demographic factors was investigated using a 2-Factor ANOVA with a factorial model. All analyses were run with α=0.05.

Results:

Seventy infants were enrolled in the overarching randomized controlled trial. There were 9 sets of living twins, 3 infants who expired, 5 infants who were withdrawn, and 2 infants who were transferred prior to 35 weeks PMA, leaving 51 parent-infant dyads. Of those dyads, 27 (53%) were Black and 35 (69%) had public insurance. 20 (39%) of the mothers in the sample were legally married. On average, families in this sample lived 64.2 (± 54.7) miles from the study site hospital. See Table 1 for infant characteristics.

Table 1.

Characteristics of the cohort.

N=51	N(%), Mean (STD) or Median (IQR)
Medical Factors
Female sex	32 (63%)
Caesarean section delivery (n=50)	32 (64%)
EGA, weeks	29.3 (2.6)
Apgar scores at 1 min (n=50)	4.2 (2.7)
Apgar scores at 5 min (n=50)	6.3 (2.2)
Endotracheal intubation, days (n=49)	1 (0-12)
NIMV days (n=47)	5 (1-11)
NEC (n=48)	3 (6%)
Sepsis (n=49)	7 (14%)
Length of stay, days (n=49)	57 (41-92)
PDA (n=49)	12 (25%)
Cerebral injury (n=48)	3 (6%)
PMA at discharge, weeks (n=49)	39.5 (9.7)

Abbreviations: EGA: estimated gestational age; NIMV: non-invasive mechanical ventilation, NEC: necrotizing enterocolitis; PDA: patent ductus arteriosus; Cerebral injury: Grade III-IV intraventricular hemorrhage or cystic periventricular leukomalacia; PMA: postmenstrual age;

Socio-demographic Factors and Relationships to Parent Presence:

See Table 2 for the relationships of socio-demographic factors with parent presence before and after accounting for the randomization group. More parent presence in the NICU was related to having private insurance (p=0.03), fewer children (p=0.004), and being married (p=0.048). Parents who were Black demonstrated less parent presence (p=.04). There were no other significant relationships between socio-demographic factors and parent presence.

Table 2.

Relationships between socio-demographic factors and parent presence. N=51

	N (%), Mean (STD) or Median (IQR)†		N (%), Mean (STD) or Median (IQR)†
	High Presence N=15	Low Presence N=7	High Presence N=18	Low Presence N=11
Maternal age	25.7 (5.6)	25.1 (4.9)	29.7 (7.4)	31.2 (6.3)	0.002
Maternal race					0.46	0.04
Black	7 (47%)	5 (71%)	6 (33%)	9 (82%)
Not Black	8 (53%)	2 (29%)	12 (67%)	2 (18%)
Education level					0.59	0.11
College degree	3 (25%)	1 (14%)	7 (54%)	0 (0%)
No college degree	9 (75%)	6 (86%)	6 (46%)	5 (100%)
Insurance type					0.25	<0.001
Public	9 (60%)	6 (86%)	9 (50%)	11 (100%)
Private	6 (40%)	1 (14%)	9 (50%)	0 (0%)
Children outside the hospital *	0 (0-1)	2 (1-3)	0 (0-1)	2 (1-4)	0.26	0.004
Marital status **					0.66	0.048
Legally married	4 (33%)	1 (14%)	9 (69%)	6 (83%)
Single	8 (67%)	6 (86%)	4 (31%)	1 (17%)
Distance from hospital in minutes ***	56 (18-152)	25 (20-66)	34 (14-113)	40 (16-120)	<0.001

^*N=37,

^**N=38,

^***N=50.

^†Median (IQR) was used when the data had a non-normal distribution.

^1.P-value is from investigations of an interaction effect of randomization group on parent presence among different socio-demographic groups using a 2-factor ANOVA with a factorial model, significance, p<0.05.

^2.P-value is from investigations of relationships between parent presence and sociodemographic groups using a 2-factor ANOVA with a factorial model, significance, p<0.05.

Note: Black boxes indicate that the interaction effect of randomization was significant, so the relationship between parent presence and socio-demographics is not valid.

-High presence was defined as either parent being present an average of 4 or 5 days per week, not including weekends. Low presence was defined as either parent being present an average of 0 to 3 days per week, not including weekends.

Socio-demographic Factors and Relationships to Parent Engagement:

See Table 3 for socio-demographic characteristics of the cohort along with relationships of socio-demographic factors with parent engagement before and after accounting for group randomization. More parent engagement was observed in parents with private insurance (p=0.01), who had fewer children (p=0.03), and were legally married (p=0.01). Parents who were Black demonstrated less parent engagement (p=0.02). There were no other significant relationships between socio-demographic factors and parent engagement.

Table 3.

Relationships between sociodemographic factors and parent engagement. N=51

						P value for Engagement2
	N (%), Mean (STD) or Median (IQR)†		N (%), Mean (STD) or Median (IQR)†
	High Engagement N=12	Low Engagement N=10	High Engagement N=15	Low Engagement N=14
Maternal age	26 (5.3)	25.4(5.5)	30.1 (7.0)	29.2(7.1)	0.004
Maternal race					0.73	0.02
Black	5(42%)	7 (70%)	5 (33%)	10 (71%)
Not Black	7(58%)	3 (30%)	10 (67%)	4 (29%)
Education level *					0.68	0.07
College degree	3 (30%)	1 (11%)	6 (55%)	6 (86%)
No college degree	7 (70%)	8 (89%)	5 (45%)	1 (14%)
Insurance type					0.28	0.01
Public	7 (58%)	8 (80%)	7 (47%)	13 (93%)
Private	5 (42%)	2 (20%)	8 (53%)	1 (7%)
Children outside the hospital *	0 (0-1)	1.5 (0-3)	0 (0-1)	1.5 (.3-3)	0.46	0.03
Marital status **					0.95	0.01
Legally married	4 (44%)	1 (10%)	8 (73%)	6 (75%)
Single	5(56%)	9 (90%)	3 (27%)	2 (25%)
Distance from hospital in minutes ***	82 (19-148)	24 (16-83)	30 (14-119)	41 (16-113)	<0.001

^*N=37,

^**N=38,

^***N=50.

^†Median (IQR) was used when the data had a non-normal distribution.

^1.P value is from investigations of an interaction effect of randomization to receive the SENSE program on parent engagement among different socio-demographic groups using a 2-factor ANOVA with a factorial model, significance, ^‡p<0.05.

^2.P-value is from investigations of relationships between parent engagement and sociodemographic groups using a 2-factor ANOVA with a factorial model, significance, ^‡p<0.05.

Note: Black boxes indicate that the interaction effect of randomization was significant, so the relationship between parent engagement and socio-demographics is not valid.

-High parent engagement was defined as parents administering 90% or more of the recommended doses of auditory and tactile exposures for the infant’s postmenstrual age during each weekday, as defined in the SENSE protocol. Low engagement was defined as conducting less than 90% of the defined auditory and tactile exposures at each postmenstrual age during each weekday.

Impact of SENSE:

Participation in the SENSE program was related to more parent presence in the NICU (p=0.002) and parent engagement in the NICU (p<0.001) for younger mothers. Participation in the SENSE program was also related to more parent presence in the NICU (p<0.001) and more parent engagement in the NICU (p=0.004) for parents living farther distances from the hospital. There were no other significant relationships.

Discussion:

The key findings were that mothers with private insurance, who were not Black, who had fewer children, and who were married were more likely to be present in the NICU and were more likely to engage in providing sensory interventions to their infants. Participation in the SENSE program was related to improved parent presence and engagement for younger mothers and those who live farther away from the hospital.

Our findings are consistent with other studies that have demonstrated that unmarried mothers, those with more children and those on public insurance are less likely to be present in the NICU and participate in care [1, 23]. These characteristics could impact the ability of parents to engage in the NICU for many reasons. Unmarried mothers may have less family support and more responsibilities outside of the hospital. Also, with more children at home, it may be difficult to coordinate childcare so mothers can visit the NICU [1]. Public insurance is a proxy for low socioeconomic status, which has been shown to be predictive of low engagement in pediatric care [24]. Our findings of parents who are Black having less presence and engagement in the NICU are consistent with other studies. People who are Black often experience discrimination in receiving prenatal care and experience poorer communication and responsiveness from providers in the NICU [25–28], which could lead to parents withdrawing from the NICU environment.

Relationships between race and presence and engagement, in relation to the SENSE program, were not observed. However, this was an exploratory study that sets the stage for further inquiry. Of note, 33% of parents who were Black in the standard of care group had high presence in the NICU, compared to 47% of parents who were Black in the SENSE group. Similarly, 33% of parents who were Black in the standard of care group had high engagement in the NICU, compared to 53% of parents who were Black in the SENSE group. For the average number of days parents were present (parent presence), a sample of n=51 provides 80% power to detect an interaction when the true interaction size is 2.3. For the proportion of parents who were engaged in the NICU (parent engagement), a sample of n=51 provides 80% power to detect an interaction when the true interaction size is 52.8%. These effect sizes are well above levels considered clinically meaningful, which indicates that a rather large sample size would be needed to detect the impact of the SENSE program on different racial groups, if it exists.

We were not able to demonstrate a relationship (p=.07) of a college degree and parent engagement in the NICU. In a study of 11 European countries, mothers with a higher level of education were found to be present in the NICU more and participate in more skin-to-skin contact [17]. It is possible that more educated mothers have higher health literacy and are likely to have better reproductive health knowledge and engage in postnatal care [29]. Our findings related to education may be confounded by culture in the United States. Unlike most European countries, parents often must return to work while their infant is in the hospital to save their maternity/paternity leave for after their infant is discharged, which could have impacted the results. This study will enable power calculations for future trials to investigate this relationship further.

Participation in the SENSE program was related to increased parent presence and engagement for parents living farther away from the hospital. Other reports have demonstrated mixed results on whether distance from the hospital impacts engagement in the NICU [20, 23]. Factors that could impact the results related to distance could be available housing, such as a Ronald McDonald House, or having specific activities expected of them in the NICU, as in the SENSE program [19]. While making the commitment to travel to the hospital can be more challenging for parents having a significant distance to travel, by providing SENSE education which emphasizes the parental role in their infant’s development and defines specific tasks for parents to complete with their infant in the hospital each day, parents were more likely to be present and engage.

Participation in the SENSE program was also related to increased presence and engagement among younger mothers. Teenage pregnancies are considered high-risk for various health outcomes such as preterm birth, low birth weight and perinatal mortality [30]. Younger mothers are also less likely to engage in care in the NICU [1]. This could be due to additional stressors or concerns about judgement or acceptance. The combination of a high-risk pregnancy and low engagement during the perinatal period could compound the effects on neonatal outcomes. However, the impact of SENSE participation on young mothers is promising, with younger mothers who were randomized to the SENSE program being more likely to be present and engage in the NICU.

Limitations of this study include using a sample from a randomized clinical trial aimed at evaluating outcomes related to the SENSE program. Therefore, power analyses were not conducted to ensure adequate sample size for evaluating socio-demographic factors that impact engagement and for evaluating if the SENSE program can increase parent presence and engagement. However, the interaction estimates in this study are exploratory and provide insight for future studies targeting these relationships. Given the complexity and co-linearity of many of the socio-demographic factors and their impact on health, further research should examine the relationships between multiple socio-demographic factors on parent engagement using properly powered samples.

The study results cannot be generalized, as the sample was limited to an urban population with high rates of social disadvantage. There is also a methodological limitation in that when parent-infant dyads were enrolled, they were informed that it was a study about sensory exposures in the NICU, and study procedures tracked what parents did with their infants in both groups. It remains unclear if this could have been suggestive to families in the control group and increased sensory exposure above what would have occurred normally. Finally, this study was limited by capturing sensory exposures only during weekdays and did not capture them on weekends, and this study could have been impacted by inaccuracies in documentation of sensory experiences in the NICU. Further, accurately capturing parent engagement and presence in the NICU is challenging, with no well-defined and valid methodology. The method that was used for this study supplemented parent documentation with nursing and research staff documentation, which allowed for differentiation in how data was collected based on the individual collecting the data. Parent presence was measured by days the parent visited, but did not capture the exact amount of time the parent spent in the NICU during each visit.

This study isolated socio-demographic factors that are related to parent presence and engagement in the NICU, which can aid targeted interventions to empower participation in the NICU. In addition, the SENSE program, as a parent-delivered, positive sensory program, shows promise for improving parent engagement among young mothers and parents living farther distance from the hospital, which in turn has the potential to improve outcomes. Further study of health equity among racial and ethnic minorities in the NICU are needed.

Highlights:

• More parent presence and engagement were observed among married, privately insured parents who had fewer children and were not Black.

• SENSE programming increased parent presence and engagement in younger mothers and parents living farther distances from the hospital.

• SENSE, as a parent-delivered, positive sensory program, shows promise for improving parent engagement among Black mothers, young mothers and families living farther away from the hospital, which in turn has the potential to improve outcomes for these populations.

Abbreviations:

SENSE

Supporting and Enhancing NICU Sensory Experiences program

NICU

Neonatal Intensive Care Unit

SLCH

St. Louis Children’s Hospital

PMA

Postmenstrual Age

EGA

Estimated Gestational Age