OBJECTIVES

Evaluate the association between maternal social factors and maternal time spent in the NICU for very preterm infants admitted to 4 level III and IV NICUs.

METHODS

In this prospective observational cohort study, we enrolled mother–infant dyads whose infants were born <32 weeks' gestation. Enrollment occurred after 2 weeks of NICU exposure, when maternal social factors and demographic information was collected. Maternal time spent in the NICU was abstracted from the electronic medical record and was dichotomized into 0 to 6 days and ≥6 days per week. Demographic differences between the 2 groups were compared by using χ2 tests. Logistic regression was used to assess the independent association between maternal social factors and the average number of days per week spent in the NICU.

RESULTS

A total of 169 mother–infant dyads were analyzed. Maternal social factors associated with more time spent in the NICU included an annual household income of >$100 000, compared with those with an annual household income of <$50 000 (adjusted odds ratio [aOR]: 5.68; 95% confidence interval [CI] 1.77–18.19), a travel time <30 minutes to the NICU (compared with those who traveled >60 minutes [aOR: 7.85; 95% CI 2.81–21.96]), and the lack of other children in the household, compared with women with other children (aOR: 3.15; 95% CI 1.39–7.11).

CONCLUSIONS

Maternal time spent in the NICU during a prolonged birth hospitalization of a very preterm infant differed by socioeconomic status, travel time, and presence of other dependents. Strategies to better identify and reduce these disparities to optimize engagement and, subsequently, improve infant health outcomes is needed.

Over 10% of US-born infants are born preterm with birth hospitalizations lasting weeks to months.1,2  In contrast to mothers who deliver healthy term newborns, mothers of preterm infants have starkly different birth experiences, with limitations on the frequency and quantity of direct care they can provide to their preterm infants, given the need for specialized medical care for various conditions, particularly for very preterm infants, <32 weeks' gestation. Previous qualitative work from our group revealed several barriers to maternal engagement during birth hospitalization, including the stress and trauma of preterm delivery and subsequent NICU hospitalization, presence of many tubes and wires within the physical environment of the NICU, and lack of available paid leave.3,4  Conversely, facilitators to engagement included strong family social support, the privacy of single-family rooms within the NICU, and families establishing a trusted relationship with their medical provider team.

Although the initial birthing and infant care experience of families of preterm infants contrast greatly from their term counterparts, a growing body of literature has revealed that parental engagement in the medical and developmental care of their preterm infants is critically important. Family integrated care, defined as parental engagement in all aspects of their neonate’s care during hospitalization, is associated with several positive outcomes, including higher breastfeeding rates, shorter hospitalization, and lower parental anxiety scores.5  Similarly, the use of kangaroo care is associated with many physiologic and behavioral short- and long-term benefits for preterm infants.6,7  Although we have some greater understanding of the positive impact of integrating parents into NICU care, we know little about the barriers and facilitators mothers experience in being able to provide care of their infants during the prolonged hospitalization, particularly in regions of the United States where the distance traveled from a primary residence to the hospital may be high.

To address this gap in the literature, we sought to investigate a broad array of maternal social factors such as race and ethnicity, marital status, and presence of other dependents and their association with the average days per week spent in the NICU in 4 level III and IV NICUs during the birth hospitalization of their very preterm infants.

For this prospective observational cohort study, we recruited mother–infant dyads from 4 NICUs. A total of 3 sites are level III NICUs, and 1 site is a level IV NICU. One level III unit serves as the regional safety net hospital in an urban area, another is the primary academic delivery hospital for the state, and the third serves a small city ∼60 miles from the metro area. The level IV unit, situated in a suburban area, serves a broad multistate catchment area for the region. Combined, the units have a total of 191 beds and admit an average of 2900 infants each year. All sites include in-house labor and delivery services, including a high-risk delivery service for the level IV unit, which primarily admits outborn infants. A total of 3 of the 4 units provide private or semiprivate rooms, and all units provide sleeping accommodations for families either at the bedside or in close proximity to the NICU.

Maternal inclusion criteria included mothers with infants born at <32 weeks' gestation, a primary language of English or Spanish, no diagnosed psychiatric disorders, and no recorded or stated illicit substance use. Infant inclusion criteria included birth at <32 weeks' gestation and no congenital birth defects. Dyads were enrolled before 34 weeks postmenstrual age and had been admitted to the NICU for at least 2 weeks. Dyads were managed through to NICU discharge, including potential back transport within the 4 study sites.

At enrollment, mothers completed an initial questionnaire including self-reported social factors, such as race and ethnicity, marital status, highest completed level of education, insurance payer, mode and average time of travel to the NICU, and presence of other children in their household. All surveys were completed while in the NICU either via a Health Insurance Portability and Accountability Act of 1996–compliant data platform (research electronic data capture) accessed on a provided iPad or on paper, on the basis of participant preference.8  Surveys were available in English and Spanish.

Throughout the NICU hospitalization, bedside nursing staff documented maternal presence at the bedside and other family members’ presence in the infant’s electronic medical record (EMR). For all admitted infants, charting was completed hourly at the level IV NICU and during scheduled infant care times in the other 3 NICUs. These data were subsequently extracted by Health Data Compass program, which serves as a health data warehouse and provides EMR data for 3 of the participating hospitals.9  The Compass data warehouse integrates EMR and hospital billing data and, through advanced algorithms, can link these clinical data to other state and public data sources. Independent chart abstraction was used to extract maternal time spent in the NICU for the fourth hospital. In addition, mothers completed weekly time-tracking diaries, noting when they were present at the bedside. For this analysis, EMR documentation was used because of inconsistencies and the overall incompleteness of maternal documentation within the weekly self-reported time-tracking diaries.10 

We defined maternal time spent in the NICU as the average number of days spent in the NICU per week of infant birth hospitalization, as documented in the EMR by bedside nursing staff. Days spent in the NICU was dichotomized into average days per week as 0 to 6 days and ≥6 days.

Covariates in our analysis included the maternal demographics of race and ethnicity, marital status, highest completed level of education, mode of delivery, insurance status, mode and duration of travel time to the NICU, and presence of other children in the home.

We calculated the average days per week that the mother was present in the NICU during birth hospitalization and dichotomized, on the basis of natural distribution, the average days per week as 0 to 6 days and ≥6 days. Maternal demographic differences between the 2 groups and selected maternal and infant characteristics by hospital of admission were compared by using χ2 tests. The following maternal characteristics were included a priori in the adjusted model: maternal race and ethnicity, average travel time to the NICU, and presence of other children. All other variables that were significant at ≤0.1 were considered for inclusion in the model. Logistic regression was used to assess the independent association between these maternal demographics and average number of days per week spent in the NICU during the birth hospitalization. We used forward regression modeling, comparing models’ Akaike information criterion values to assess for fit. Additional variables were included in the model if they decreased the Akaike information criterion by ≥2 points to achieve the best fit and most parsimonious model. Analyses were conducted by using SAS version 9.4 (SAS Institute, Inc, Cary, NC). This study was approved by the university’s institutional review board.

A total of 203 women were consented for study enrollment between June 2017 and December 2019 (Fig 1). Supplemental Figure 3 reveals the geographic catchment area for the enrolled cohort, encompassing a total of 10 states. After accounting for infant death, transfer, and study noncompliance, a final cohort of 169 infants were included in the analysis. Among these preterm infants, 20% (n = 34) were admitted to the level IV NICU, 7% (n = 13) to the regional safety net hospital level III NICU, and 15% (n = 27) to the NICU outside of the greater metro area, with the remainder of the cohort admitted to the metro area level III academic NICU. Overall, 62% (n = 105) of women were between ages of 25 and 35 years, 57% (n = 95) identified as non-Hispanic white, and the majority (82% [n = 136] were married or living with a partner (Table 1).

FIGURE 1

Flow diagram of study participants.

FIGURE 1

Flow diagram of study participants.

Close modal
TABLE 1

Maternal Sociodemographic Characteristics by Average Days per Week Spent in the NICU During Birth Hospitalization

Days per Week
Total, n (%)0 to <6, n (%)≥6, n (%)POR (95% CI)
Maternal age    .36  
 <25 y 34 (20) 13 (38) 21 (62)  1.1 (0.5–2.5) 
 25–35 y 105 (62) 43 (41) 62 (59)  Reference 
 >35 y 30 (18) 8 (27) 22 (73)  1.9 (0.8–4.7) 
Race and ethnicity    .05  
 Non-Hispanic white 95 (57) 30 (32) 64 (68)  Reference 
 Other 70 (43) 33 (47) 37 (53)  0.4 (0.3–1.0) 
Annual household income    .08  
 <$50 000 71 (44) 32 (45) 39 (55)  Reference 
 $50 0000–$99 000 51 (32) 17 (33) 34 (67)  1.6 (0.8–3.5) 
 ≥$100 000 38 (24) 9 (24) 29 (76)  2.6 (1.1–6.4) 
Insurance status    <.01  
 Private 91 (55) 25 (27) 66 (73)  Reference 
 Public or uninsured 74 (45) 38 (51) 36 (49)  0.4 (0.2–0.7) 
Marital status    .87  
 Married or living with partner 136 (82) 52 (38) 84 (62)  1.1 (0.5–2.4) 
 Other status 30 (18) 11 (37) 19 (63)  Reference 
Highest completed education    .13  
 High school degree or less 47 (28) 22 (47) 25 (53)  0.6 (0.3–1.4) 
 Some college or associate’s degree 44 (27) 19 (43) 25 (57)  0.7 (0.3–1.6) 
 Bachelor’s degree 49 (30) 17 (35) 32 (65)  Reference 
 Master’s degree or higher 25 (15) 5 (20) 20 (80)  2.1 (0.7–6.7) 
Mode of travel to NICU    .49  
 Personal car 146 (88) 54 (37) 92 (63)  Reference 
 Other mode 20 (12) 9 (45) 11 (55)  0.7 (0.3–1.8) 
Travel time to NICU    <.01  
 <30 min 87 (51) 25 (29) 62 (71)  Reference 
 30–60 min 57 (34) 20 (35) 37 (65)  0.7 (0.4–1.5) 
 >60 min 25 (15) 19 (76) 6 (24)  0.1 (0.1–0.4) 
Pre-NICU employment    <.01  
 Working full-time 102 (62) 30 (29) 72 (71)  Reference 
 Other status 62 (38) 32 (52) 30 (48)  0.4 (0.2–0.8) 
Other children    <.01  
 Yes 98 (58) 46 (47) 52 (53)  Reference 
 No 71 (42) 18 (25) 53 (75)  2.6 (1.3–5.1) 
Days per Week
Total, n (%)0 to <6, n (%)≥6, n (%)POR (95% CI)
Maternal age    .36  
 <25 y 34 (20) 13 (38) 21 (62)  1.1 (0.5–2.5) 
 25–35 y 105 (62) 43 (41) 62 (59)  Reference 
 >35 y 30 (18) 8 (27) 22 (73)  1.9 (0.8–4.7) 
Race and ethnicity    .05  
 Non-Hispanic white 95 (57) 30 (32) 64 (68)  Reference 
 Other 70 (43) 33 (47) 37 (53)  0.4 (0.3–1.0) 
Annual household income    .08  
 <$50 000 71 (44) 32 (45) 39 (55)  Reference 
 $50 0000–$99 000 51 (32) 17 (33) 34 (67)  1.6 (0.8–3.5) 
 ≥$100 000 38 (24) 9 (24) 29 (76)  2.6 (1.1–6.4) 
Insurance status    <.01  
 Private 91 (55) 25 (27) 66 (73)  Reference 
 Public or uninsured 74 (45) 38 (51) 36 (49)  0.4 (0.2–0.7) 
Marital status    .87  
 Married or living with partner 136 (82) 52 (38) 84 (62)  1.1 (0.5–2.4) 
 Other status 30 (18) 11 (37) 19 (63)  Reference 
Highest completed education    .13  
 High school degree or less 47 (28) 22 (47) 25 (53)  0.6 (0.3–1.4) 
 Some college or associate’s degree 44 (27) 19 (43) 25 (57)  0.7 (0.3–1.6) 
 Bachelor’s degree 49 (30) 17 (35) 32 (65)  Reference 
 Master’s degree or higher 25 (15) 5 (20) 20 (80)  2.1 (0.7–6.7) 
Mode of travel to NICU    .49  
 Personal car 146 (88) 54 (37) 92 (63)  Reference 
 Other mode 20 (12) 9 (45) 11 (55)  0.7 (0.3–1.8) 
Travel time to NICU    <.01  
 <30 min 87 (51) 25 (29) 62 (71)  Reference 
 30–60 min 57 (34) 20 (35) 37 (65)  0.7 (0.4–1.5) 
 >60 min 25 (15) 19 (76) 6 (24)  0.1 (0.1–0.4) 
Pre-NICU employment    <.01  
 Working full-time 102 (62) 30 (29) 72 (71)  Reference 
 Other status 62 (38) 32 (52) 30 (48)  0.4 (0.2–0.8) 
Other children    <.01  
 Yes 98 (58) 46 (47) 52 (53)  Reference 
 No 71 (42) 18 (25) 53 (75)  2.6 (1.3–5.1) 

OR, odds ratio; —, not applicable.

In the bivariate analysis evaluating maternal characteristics by time in the NICU, mothers with a higher annual income, full-time pre-NICU employment, and shorter travel time to the NICU were more likely to spend ≥6 days per week in the NICU (Table 1). Conversely, women with other children present in the home were less likely to spend ≥6 days per week in the NICU. There was no significant difference in maternal time spent in the NICU by maternal age, race and/or ethnicity, marital status, educational attainment, or mode of travel to the NICU. Maternal time in the NICU decreased throughout birth hospitalization. In Fig 2, we outline the mean visitation, in days per week, and represent the initial 7 weeks of NICU admission, a period in which 80% of the study cohort remained hospitalized, and the week before discharge for all infants.

FIGURE 2

Mean maternal time in the NICU in days per week.

FIGURE 2

Mean maternal time in the NICU in days per week.

Close modal

In evaluation of maternal and infant characteristics by hospital (Table 2), infants admitted to the nonmetro level III unit were more likely to be larger and older, with an average birth weight of 1588±362 g (P < .001) and gestational age of 30.3±1.4 weeks (P < .001). There was no difference by hospital of admission in infant sex (P = .6) or 5-minute Apgar score (P = .3). Women whose infants were admitted to the level IV regional unit were more likely to be non-Hispanic white, have a higher annual household income, and be married or living with a partner, compared with the other units. There was no difference by hospital of admission in overall maternal age, completed education, presence of other children, or mode of travel or time traveled to the NICU.

TABLE 2

Selected Infant and Maternal Demographics by Hospital

TotalAcademic Suburban Level IIISafety Net Urban Level IIINonmetro Level IIIRegional Level IVP
Total, n (%) 169 35 (21) 12 (7) 24 (14) 98 (58)  
Maternal characteristics, n (%)       
 Maternal age      .16 
  <25 y 34 (20) 7 (20) 5 (42) 7 (29) 15 (15)  
  25–35 y 105 (62) 21 (60) 6 (50) 16 (67) 62 (63)  
  >35 y 30 (18) 7 (20) 1 (8) 1 (4) 21 (21)  
 Race or ethnicity      <.01 
  Non-Hispanic white 94 (57) 24 (71) 1 (8) 13 (62) 56 (58)  
  Other 70 (43) 10 (29) 11 (92) 8 (38) 41 (42)  
 Annual household income      <.01 
  <$50 000 71 (44) 10 (30) 8 (80) 7 (33) 46 (48)  
  $50 0000–$99 000 51 (32) 8 (24) 2 (20) 9 (43) 32 (33)  
  ≥$100 000 38 (24) 15 (45) 0 (0) 5 (24) 18 (19)  
 Insurance status      <.01 
  Private 91 (55) 22 (63) 0 (0) 12 (62) 57 (58)  
  Public or uninsured 74 (45) 13 (37) 12 (100) 9 (43) 41 (42)  
 Marital status      .07 
  Married or living with partner 136 (82) 34 (97) 9 (75) 17 (81) 76 (78)  
  Other status 30 (18) 1 (3) 3 (25) 4 (19) 22 (22)  
 Highest completed education      .2 
  High school degree or less 47 (28) 7 (20) 7 (58) 5 (24) 28 (29)  
  Some college or associate’s degree 44 (27) 10 (29) 5 (42) 6 (29) 23 (24)  
  Bachelor’s degree 49 (30) 12 (34) 6 (29) 31 (32)  
  Master’s degree or higher 25 (15) 6 (17) 4 (19) 15 (15)  
 Mode of travel to NICU      .4 
  Personal car 146 (88) 31 (89) 9 (75) 20 (95) 86 (88)  
  Other mode 20 (12) 4 (11) 3 (25) 1 (5) 12 (12)  
 Travel time to NICU      .12 
  <30 min 87 (51) 12 (35) 8 (67) 14 (58) 53 (54)  
  30–60 min 57 (34) 12 (35) 4 (33) 8 (33) 32 (33)  
  >60 min 25 (25) 10 (29) 0 (0) 2 (8) 13 (13)  
 Pre-NICU employment      <.01 
  Working full-time 102 (62) 21 (60) 2 (17) 18 (86) 61 (64)  
  Other status 62 (38) 14 (40) 10 (83) 3 (14) 35 (36)  
 Other children      .58 
  Yes 98 (58) 20 (57) 9 (75) 15 (63) 54 (55)  
  No 71 (42) 15 (43) 3 (25) 9 (37) 44 (45)  
Infant characteristics       
 Birth wt, g, mean (SD) 1253 (435) 1291 (436) 1184 (363) 1564 (338) 1172 (432) <.01 
 Gestational age, wk, mean (SD), 28.6 (2.4) 28.5 (2.2) 28.3 (2.7) 30.2 (1.4) 28.2 (2.5) <.01 
 Sex, n (%)      .58 
  Male 95 (56) 21 (60) 7 (58) 16 (67) 51 (52)  
  Female 74 (44) 14 (40) 5 (42) 8 (33) 47 (48)  
 5-min Apgar, mean (SD) 7.5 (1.4) 7.7 (1.4) 7.4 (1.6) 7.9 (1.1) 7.4 (1.4) .34 
TotalAcademic Suburban Level IIISafety Net Urban Level IIINonmetro Level IIIRegional Level IVP
Total, n (%) 169 35 (21) 12 (7) 24 (14) 98 (58)  
Maternal characteristics, n (%)       
 Maternal age      .16 
  <25 y 34 (20) 7 (20) 5 (42) 7 (29) 15 (15)  
  25–35 y 105 (62) 21 (60) 6 (50) 16 (67) 62 (63)  
  >35 y 30 (18) 7 (20) 1 (8) 1 (4) 21 (21)  
 Race or ethnicity      <.01 
  Non-Hispanic white 94 (57) 24 (71) 1 (8) 13 (62) 56 (58)  
  Other 70 (43) 10 (29) 11 (92) 8 (38) 41 (42)  
 Annual household income      <.01 
  <$50 000 71 (44) 10 (30) 8 (80) 7 (33) 46 (48)  
  $50 0000–$99 000 51 (32) 8 (24) 2 (20) 9 (43) 32 (33)  
  ≥$100 000 38 (24) 15 (45) 0 (0) 5 (24) 18 (19)  
 Insurance status      <.01 
  Private 91 (55) 22 (63) 0 (0) 12 (62) 57 (58)  
  Public or uninsured 74 (45) 13 (37) 12 (100) 9 (43) 41 (42)  
 Marital status      .07 
  Married or living with partner 136 (82) 34 (97) 9 (75) 17 (81) 76 (78)  
  Other status 30 (18) 1 (3) 3 (25) 4 (19) 22 (22)  
 Highest completed education      .2 
  High school degree or less 47 (28) 7 (20) 7 (58) 5 (24) 28 (29)  
  Some college or associate’s degree 44 (27) 10 (29) 5 (42) 6 (29) 23 (24)  
  Bachelor’s degree 49 (30) 12 (34) 6 (29) 31 (32)  
  Master’s degree or higher 25 (15) 6 (17) 4 (19) 15 (15)  
 Mode of travel to NICU      .4 
  Personal car 146 (88) 31 (89) 9 (75) 20 (95) 86 (88)  
  Other mode 20 (12) 4 (11) 3 (25) 1 (5) 12 (12)  
 Travel time to NICU      .12 
  <30 min 87 (51) 12 (35) 8 (67) 14 (58) 53 (54)  
  30–60 min 57 (34) 12 (35) 4 (33) 8 (33) 32 (33)  
  >60 min 25 (25) 10 (29) 0 (0) 2 (8) 13 (13)  
 Pre-NICU employment      <.01 
  Working full-time 102 (62) 21 (60) 2 (17) 18 (86) 61 (64)  
  Other status 62 (38) 14 (40) 10 (83) 3 (14) 35 (36)  
 Other children      .58 
  Yes 98 (58) 20 (57) 9 (75) 15 (63) 54 (55)  
  No 71 (42) 15 (43) 3 (25) 9 (37) 44 (45)  
Infant characteristics       
 Birth wt, g, mean (SD) 1253 (435) 1291 (436) 1184 (363) 1564 (338) 1172 (432) <.01 
 Gestational age, wk, mean (SD), 28.6 (2.4) 28.5 (2.2) 28.3 (2.7) 30.2 (1.4) 28.2 (2.5) <.01 
 Sex, n (%)      .58 
  Male 95 (56) 21 (60) 7 (58) 16 (67) 51 (52)  
  Female 74 (44) 14 (40) 5 (42) 8 (33) 47 (48)  
 5-min Apgar, mean (SD) 7.5 (1.4) 7.7 (1.4) 7.4 (1.6) 7.9 (1.1) 7.4 (1.4) .34 

—, not applicable.

In the adjusted analysis (Table 3), maternal social factors associated with a higher time spent in the NICU each week revealed that women with an annual household income of ≥$100 000 were more likely to spend ≥6 days a week in the NICU, compared with those with an annual household income of <$50 000 (adjusted odds ration [aOR]: 5.68; 95% confidence interval [CI] 1.77–18.19). Compared with women with other children, those with no other children in the household had a 3.15 times higher odds of spending ≥6 days a week in the NICU (aOR: 3.15; 95% CI 1.39–7.11). The time spent in the NICU differed by travel time: compared with women who traveled >60 minutes from home to the NICU, those who traveled <30 minutes were much more likely to spend ≥6 days a week in the NICU (aOR: 7.85; 95% CI 2.81–21.96). Evaluation of maternal time spent in the NICU revealed no significant difference by maternal reported race and/or ethnicity.

TABLE 3

Odds of Spending ≥6 d per Week in the NICU During Birth Hospitalization

aOR (95% CI)
Race and ethnicity  
 Non-Hispanic white Reference 
 Other 0.71 (0.32–1.56) 
Annual household income  
 <$50 000 Reference 
 $50 000–$99 999 1.74 (0.72–4.24) 
 ≥$100 000 5.68 (1.77–18.19) 
Travel time to NICU  
 <30 min 7.85 (2.81–21.96) 
 30–60 min 5.86 (2.02–17.02) 
 >60 min Reference 
Other children  
 Yes Reference 
 No 3.15 (1.39–7.11) 
aOR (95% CI)
Race and ethnicity  
 Non-Hispanic white Reference 
 Other 0.71 (0.32–1.56) 
Annual household income  
 <$50 000 Reference 
 $50 000–$99 999 1.74 (0.72–4.24) 
 ≥$100 000 5.68 (1.77–18.19) 
Travel time to NICU  
 <30 min 7.85 (2.81–21.96) 
 30–60 min 5.86 (2.02–17.02) 
 >60 min Reference 
Other children  
 Yes Reference 
 No 3.15 (1.39–7.11) 

Adjusted for maternal race and ethnicity, annual household income, and time travel to NICU and other children.

In this prospective cohort study of mothers of infants born <32 weeks from 4 NICUs, we found that a higher annual household income, a shorter travel time to the NICU, and the absence of other children in the home were independently associated with more days each week spent in the NICU during birth hospitalization.

Our results call attention to several issues and barriers that families may encounter during a prolonged birth hospitalization of a very preterm infant. They include the disparities in the availability or lack thereof of paid family leave and the broader financial burden for families with limited financial resources. Indeed, our cohort demonstrated decreased maternal time in the NICU throughout birth hospitalization, highlighting the likely several competing interests a mother faces during this time, including potential return to the workplace. These issues highlight the impact of social determinants of health (SDOHs), the environmental and economic conditions in which individuals are born and live, on both the hospitalized preterm infant and the family unit.11  SDOHs broadly contribute to health inequities, such as preterm birth, with known disparities by factors including maternal race and/or ethnicity, education, and marital status.12,13 

In our study, conducted before to the onset of the coronavirus disease 2019 pandemic, participating hospitals had varied sibling visitation polices. Policies ranged from siblings under the age of 12 years being provided a playroom outside the unit but being prohibited from entering the patient areas to siblings of all ages having unrestricted access with supervision in single patient rooms, along with the availability of drop-in child care on-site. Despite these seemingly generous accommodations, participating women in our study without other dependents at home had a >3 times higher odds of spending ≥6 days a week in the NICU, compared with women with other dependents at home. Undoubtedly, the coronavirus disease 2019 pandemic has worsened this disparity, with universal stricter visitation policies and the exclusion of sibling visitation throughout hospitalization.

Similar to previous studies in which researchers evaluated the distance from home to hospital, we found that increased travel time to the NICU was significantly associated with less maternal time spent in the NICU.14  Our cohort included infants admitted from a wide geographic catchment area of the Rocky Mountain west and beyond, resulting in a significant distance from home to hospital for several families. Access to acute obstetrical and neonatal care for nonurban dwelling women is becoming an increasing concern; low access to perinatal facilities in the noncoastal western region ranges from 75.5% to 81.7% of the population residing within 60 minutes of such facility.15  This contrasts starkly to the Northeast and California, where local perinatal care access rates are >97%. The rural–urban gap to accessing care is widening; between 2002 and 2013, an increased percentage of nonurban women experienced longer travel distances to access both obstetrical and advanced neonatal care.16  Loss of these rural obstetrical services has been associated with subsequent increases in nonhospital births and preterm births.17  With decreased access to perinatal care close to home, antenatal transfer of high-risk pregnant women and postnatal transfer of preterm infants is often necessary.

Although perinatal transfer may be unavoidable, opportunities for back transport closer to maternal residence, once an infant is deemed medically stable, should be considered. Evaluation of back transport within a geographically similar cohort within the Rocky Mountain region revealed that overall prevalence of back transport was low and medically stable infants (defined as low flow nasal cannula and full enteral feedings) remained admitted to a higher level of neonatal care for 4 additional weeks.18  Although regional variation in the prevalence of back transport exists, this represents a potential opportunity for greater family engagement closer to home and without a significant difference in the length of stay, whereas pursuing back transport must be balanced with resource availability, parental agreement, cost, care quality, and the risk of requiring readmission to a higher level of care.1921 

In this work, we highlight the need and importance of standardized screening for and subsequently addressing SDOHs during birth hospitalization.22  The NICU hospitalization affords a unique opportunity for this work, given the presence of preterm birth disparities and the often prolonged hospitalization of weeks to months; however, in previous work, researchers have found that screening for basic needs, aside from employment status, are inconsistently assessed.23  These basic needs, including housing and food insecurity, child care, transportation, and utility necessities, were assessed <40% of the time.23  Adoption of SDOH screening within the outpatient pediatric community has been successful, with the leverage of existing clinic infrastructure and social work support, a method that has not yet been widely adopted by inpatient providers.24 

For effective integration of SDOH screening into routine clinical care in the NICU, recognizing and addressing the disparate experiences of certain families will be essential. For instance, despite the typical availability and involvement of social workers within the NICU, existing gaps were highlighted in a recent qualitative study including families of color and with low socioeconomic status.25  Many families reported being unaware of the scope of social work’s role or available services, placing an undue burden on families to seek out this information on their own. In addition, families with limited English-language proficiency will require special attention. Families who primarily speak Spanish report being less comfortable asking questions to their medical team and have been associated with a decreased ability to correctly identify their infant’s diagnosis, factors that may contribute to overall decreased parental engagement in the NICU.26,27  Additional factors that need to be considered include substance use disorders, mental health conditions, and other experiences of trauma among some families. SDOH screening in the NICU will need to occur with approaches that integrate principles of trauma-informed care and stigma and bias reduction.

The strengths of this study include using an interdisciplinary (medicine, nursing, and social work) approach to study design and recruiting mother–infant dyads from multiple sites that represent a broad population reflective of the state’s demographics.28  In this study, we also included a wide-ranging look at the social factors impacting time in the NICU beyond race and ethnicity and socioeconomic status, as typically defined by only income and/or insurance status. Limitations to this study include the use of cross-sectional data at study enrollment, which has the potential to change over time during a prolonged birth hospitalization, depending on individual circumstances. Although this study recruited both English- and Spanish-speaking families, the overall number of Spanish-speaking mothers enrolled was small, resulting in the inability to analyze data by primary language. Potential enrollment bias, favoring English- and Spanish-speaking families who were available for consent either in-person or by telephone must also be acknowledged. Although English and Spanish are the primary 2 languages spoken by families served by the participating hospitals, evaluation of other languages, especially those without as readily available interpreters and translated written materials, are crucial to reduce language discordance.

Maternal time spent in the NICU during a prolonged birth hospitalization of a very preterm infant differed by socioeconomic status, travel time, and presence of other dependents. Strategies to better identify through standardized SDOH inpatient screening and reduce these disparities to optimize engagement and, subsequently, improve infant health outcomes is needed.

We dedicate this article to Dr Annie Hall, a beloved colleague, neonatologist, patient advocate, dear friend, and mother.

FUNDING: Supported by a National Institutes of Health National Center for Advancing Translational Sciences Colorado Clinical and Translational Science Award (grant number UL1 TR002535). The contents are the authors’ sole responsibility and do not necessarily represent official National Institutes of Health views. Funded by the National Institutes of Health (NIH).

Dr Bourque provided statistical analysis oversight and drafted and edited the manuscript; Ms Weikel completed the statistical analysis and edited the manuscript; Drs Palau and Hall, Ms Scott, Ms Shah, and Ms Roybal edited the manuscript; Drs Greenfield, Klawetter, and Neu designed the study and reviewed the manuscript; Dr Hwang designed the study, provided statistical analysis oversight, and reviewed and edited the manuscript; and all authors approved the final manuscript as submitted.

1.
March of Dimes
.
2020 March of Dimes report card
.
2.
Lee
HC
,
Bennett
MV
,
Schulman
J
,
Gould
JB
,
Profit
J
.
Estimating length of stay by patient type in the neonatal intensive care unit
.
Am J Perinatol
.
2016
;
33
(
8
):
751
757
3.
Klawetter
S
,
Neu
M
,
Roybal
KL
,
Greenfield
JC
,
Scott
J
,
Hwang
S
.
Mothering in the NICU: a qualitative exploration of maternal engagement
.
Soc Work Health Care
.
2019
;
58
(
8
):
746
763
4.
Neu
M
,
Klawetter
S
,
Greenfield
JC
,
Roybal
K
,
Scott
JL
,
Hwang
SS
.
Mothers’ experiences in the NICU before family-centered care and in NICUs where it is the standard of care
.
Adv Neonatal Care
.
2020
;
20
(
1
):
68
79
5.
O’Brien
K
,
Robson
K
,
Bracht
M
, et al
;
FICare Study Group and FICare Parent Advisory Board
.
Effectiveness of Family Integrated Care in neonatal intensive care units on infant and parent outcomes: a multicentre, multinational, cluster-randomised controlled trial [published correction appears in Lancet Child Adolesc Health. 2018;2(8):e20]
.
Lancet Child Adolesc Health
.
2018
;
2
(
4
):
245
254
6.
Campbell-Yeo
ML
,
Disher
TC
,
Benoit
BL
,
Johnston
CC
.
Understanding kangaroo care and its benefits to preterm infants
.
Pediatric Health Med Ther
.
2015
;
6
:
15
32
7.
Charpak
N
,
Tessier
R
,
Ruiz
JG
, et al
.
Twenty-year follow-up of kangaroo mother care versus traditional care
.
Pediatrics
.
2017
;
139
(
1
):
e20162063
8.
Harris
PA
,
Taylor
R
,
Thielke
R
,
Payne
J
,
Gonzalez
N
,
Conde
JG
.
Research electronic data capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support
.
J Biomed Inform
.
2009
;
42
(
2
):
377
381
9.
Health Data Compass
.
Health Data Compass
.
Available at: https://www.healthdatacompass.org/. Accessed January 25, 2021
10.
Powers
SA
,
Taylor
K
,
Tumin
D
,
Kohler
JA
 Sr
.
Measuring parental presence in the neonatal intensive care unit [published online ahead of print August 20, 2020]
.
Am J Perinatol
.
doi: 10.1055/s-0040-1715525
11.
US Department of Health and Human Services
.
Social determinants of health
.
12.
Thoma
ME
,
Drew
LB
,
Hirai
AH
,
Kim
TY
,
Fenelon
A
,
Shenassa
ED
.
Black-white disparities in preterm birth: geographic, social, and health determinants
.
Am J Prev Med
.
2019
;
57
(
5
):
675
686
13.
Purisch
SE
,
Gyamfi-Bannerman
C
.
Epidemiology of preterm birth
.
Semin Perinatol
.
2017
;
41
(
7
):
387
391
14.
Latva
R
,
Lehtonen
L
,
Salmelin
RK
,
Tamminen
T
.
Visits by the family to the neonatal intensive care unit
.
Acta Paediatr
.
2007
;
96
(
2
):
215
220
15.
Rayburn
WF
,
Richards
ME
,
Elwell
EC
.
Drive times to hospitals with perinatal care in the United States
.
Obstet Gynecol
.
2012
;
119
(
3
):
611
616
16.
Hung
P
,
Casey
MM
,
Kozhimannil
KB
,
Karaca-Mandic
P
,
Moscovice
IS
.
Rural-urban differences in access to hospital obstetric and neonatal care: how far is the closest one?
J Perinatol
.
2018
;
38
(
6
):
645
652
17.
Kozhimannil
KB
,
Hung
P
,
Henning-Smith
C
,
Casey
MM
,
Prasad
S
.
Association between loss of hospital-based obstetric services and birth outcomes in rural counties in the United States
.
JAMA
.
2018
;
319
(
12
):
1239
1247
18.
Bourque
SL
,
Levek
C
,
Melara
DL
,
Grover
TR
,
Hwang
SS
.
Prevalence and predictors of back-transport closer to maternal residence after acute neonatal care in a regional NICU
.
Matern Child Health J
.
2019
;
23
(
2
):
212
219
19.
Kunz
SN
,
Dukhovny
D
,
Profit
J
,
Mao
W
,
Miedema
D
,
Zupancic
JAF
.
Predicting Successful Neonatal Retro-Transfer to a Lower Level of Care
.
J Pediatr
.
2019
;
205
:
272
276.e1
20.
Attar
MA
,
Lang
SW
,
Gates
MR
,
Iatrow
AM
,
Bratton
SL
.
Back transport of neonates: effect on hospital length of stay
.
J Perinatol
.
2005
;
25
(
11
):
731
736
21.
Hanrahan
K
,
Gates
M
,
Attar
MA
,
Lang
SW
,
Frohna
A
,
Clark
SJ
.
Neonatal back transport: perspectives from parents of Medicaid-insured infants and providers
.
Neonatal Netw
.
2007
;
26
(
5
):
301
311
22.
Parker
MG
,
Garg
A
,
McConnell
MA
.
Addressing childhood poverty in pediatric clinical settings: the neonatal intensive care unit is a missed opportunity
.
JAMA Pediatr
.
2020
;
174
(
12
):
1135
1136
23.
Parker
MG
,
Garg
A
,
Brochier
A
, et al
.
Approaches to addressing social determinants of health in the NICU: a mixed methods study [published online ahead of print October 30, 2020]
.
J Perinatol
.
doi: 10.1038/s41372-020-00867-w
24.
Sokol
R
,
Austin
A
,
Chandler
C
, et al
.
Screening children for social determinants of health: a systematic review
.
Pediatrics
.
2019
;
144
(
4
):
e20191622
25.
Sigurdson
K
,
Profit
J
,
Dhurjati
R
, et al
.
Former NICU families describe gaps in family-centered care
.
Qual Health Res
.
2020
;
30
(
12
):
1861
1875
26.
Palau
MA
,
Meier
MR
,
Brinton
JT
,
Hwang
SS
,
Roosevelt
GE
,
Parker
TA
.
The impact of parental primary language on communication in the neonatal intensive care unit
.
J Perinatol
.
2019
;
39
(
2
):
307
313
27.
Miquel-Verges
F
,
Donohue
PK
,
Boss
RD
.
Discharge of infants from NICU to Latino families with limited English proficiency
.
J Immigr Minor Health
.
2011
;
13
(
2
):
309
314
28.
Colorado Department of Public Health and Environment Vital Statistics Program
.
Colorado Births and Deaths
.
Denver, CO
;
Colorado Department of Public Health
:
2019

Competing Interests

POTENTIAL CONFLICTS OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors declare no competing financial interests.