Effect of Delayed Newborn Bath

The authors report no con- flict of interest or relevant financial relationships.

Correspondence Susan Warren, MSc, 5 Alderdice Place, St. John’s, Newfoundland and Labrador, Canada A1B2P8. [email protected]

Keywords bathing breastfeeding breastfeeding initiation delayed bath exclusive breastfeeding hypoglycemia hypothermia neonatal newborn newborn feeding

Susan Warren, MSc, is a student in the Department of Nursing, Memorial University of Newfoundland and Labrador, St. John’s, NL, Canada.

William K. Midodzi, PhD, is an assistant professor in the Department of Medicine, Memorial University of Newfoundland and Labrador, St. John’s, NL, Canada.

(Continued)

http://jognn.org

R E S E A R C H

Effects of Delayed Newborn Bathing on Breastfeeding, Hypothermia, and Hypoglycemia Susan Warren, William K. Midodzi, Leigh-Anne Allwood Newhook, Phil Murphy, and Laurie Twells

ABSTRACT

Objective: To determine whether delaying the newborn bath by 24 hours increases the prevalence of breastfeeding

initiation and exclusive breastfeeding at discharge in healthy full-term and late preterm newborns (34 0/7–36 6/7 weeks

gestation) and to examine the effect of delayed newborn bathing on the incidences of hypothermia and hypoglycemia.

Design: Pre–post implementation, retrospective, cohort study.

Setting: Provincial children’s hospital with an average of 2,500 births per year.

Participants: Healthy newborns (N ¼ 1,225) born at 34 0/7 weeks or more gestation who were admitted to the

mother–baby unit.

Methods: We compared newborns who were bathed before 24 hours (n ¼ 680, preimplementation group) to new-

borns who were bathed after 24 hours (n ¼ 545, postimplementation group).

Results: After adjustment for confounders, the odds of exclusive breastfeeding at discharge were 33% greater in the

postimplementation group than in the preimplementation group (adjusted odds ratio ¼ 1.334; 95% confidence interval

[1.049,1.698]; p ¼ .019). Delayed bathing was associated with decreased incidence of hypothermia and hypoglycemia

(p ¼ .007 and p ¼ .003, respectively). We observed no difference in breastfeeding initiation between groups.

Conclusion: Delaying the newborn bath for 24 hours was associated with an increased likelihood of exclusive

breastfeeding at discharge and a decreased incidence of hypothermia and hypoglycemia in healthy newborns. The

implementation of a delayed bathing policy has the potential to improve breastfeeding rates and reduce the incidence

of hypothermia and hypoglycemia.

JOGNN, 49, 181–189; 2020. https://doi.org/10.1016/j.jogn.2019.12.004

Accepted December 2019

irthing room and postpartum hospital rou-

Btines and practices may significantly disrupt

a newborn’s adjustment to the extrauterine envi-

ronment and early maternal-newborn in-

teractions, including breastfeeding. One such

practice is the timing of the newborn bath.

Newborn bathing imposes a physical separation

of the mother and newborn and impedes skin-to-

skin contact. Bathing also removes biological

substances such as amniotic fluid and vernix

caseosa, which may play a role in the newborn’s

adaptation to the extrauterine environment.

Newborns have been shown to respond positively

to the odor of their own amniotic fluid (Loos,

Reger, & Schaal, 2019). Radhakrishnan (2012)

described how newborns use the taste and

smell of amniotic fluid to guide them to their

ª 2020 AWHONN, the Association of Women’s Health, Obstetri

Published by Elsevier Inc. All rights reserved.

mothers’ nipples. Vernix caseosa has been

shown to decrease postnatal heat loss (Rissmann

et al., 2008), and researchers investigated the

possibility that vernix caseosa contains phero-

mones that contribute to olfactory cues that pro-

mote bonding and breastfeeding (Hoath,

Pickens, & Visscher, 2006). Whether vernix

caseosa is a source of olfactory pheromones

similar to amniotic fluid and breastmilk is un-

known, but this idea is consistent with the fact that

many pheromones are derived from glandular

skin secretions (Doty, 2014). The processes of

gestation and birth deposit biological substances

on the skin of newborns. Some of these sub-

stances have been shown to aid in temperature

regulation and breastfeeding, and thus the

removal of these substances during bathing may

hinder naturally occurring processes.

c and Neonatal Nurses. 181

Leigh-Anne Allwood Newhook, MD, FRCPC, is an associate professor in the Department of Pediatrics, Memorial University of Newfoundland and Labrador, St. John’s, NL, Canada.

Phil Murphy, MSc, is a data consultant/analyst in the Perinatal Program Newfoundland and Labrador, The Janeway Children’s Health and Rehabilitation Centre, St. John’s, NL, Canada.

Laurie Twells, PhD, is an associate professor in the Departments of Medicine and Pharmacy, Memorial University of Newfoundland and Labrador, St. John’s, NL, Canada.

The World Health Organization recommended delaying the newborn bath by at least 24 hours after birth despite a

dearth of clinical evidence on this practice.

Effects of Delayed Newborn BathingR E S E A R C H

182

The most commonly referenced guideline for

newborn bathing is “Recommendations on

Postnatal Care of the Mother and Newborn—

2013” from the World Health Organization

(WHO; 2014). In these guidelines, the WHO

recommended that bathing should be delayed

for 24 hours after birth. Most recommendations

in this report were graded as strong or weak

based on the quality of evidence reviewed by

the Guideline Development Group. However, no

grading was provided for the recommendation

to delay newborn bathing; instead, this recom-

mendation was based on extant WHO guide-

lines (WHO, 1993). We conducted an exhaustive

search of all extant WHO guidelines and reports

and found no cited evidence for this recom-

mendation. The earliest reference to delayed

newborn bathing appeared in a report entitled

“Thermal Control of the Newborn: A Practical

Guide” (WHO, 1993), but this report included no

cited evidence. Although some findings indicate

that the practice of delayed newborn bathing

has the potential to improve breastfeeding rates

(Preer, Pisegna, Cook, Henri, & Philipp, 2013;

Suchy et al., 2018; Turney, Lowther, Pkya,

Mollon, & Fields, 2019), hypothermia (Brogan

& Rapkin, 2017; Chamberlain et al., 2019;

Gözen, Çaka, Bes‚ irik, & Perk, 2019; Kelly

et al., 2018; Suchy et al., 2018), and hypogly-

cemia (Chamberlain et al., 2019), there has

been limited evaluation of the recommendation

from the WHO to delay newborn bathing for at

least 24 hours (Chamberlain et al., 2019).

Previous researchers evaluated the effects of

delayed bathing on breastfeeding and reported

inconsistent results (Chamberlain et al., 2019;

Dicioccio, Ady, Bena, & Albert, 2019; Preer

et al., 2013; Suchy et al., 2018; Turney et al.,

2019). In the five studies we found, researchers

used pre–post retrospective chart reviews to

compare outcomes for newborns born before

and after implementation of a delayed bathing

policy. Preer et al. (2013) reported that delaying

the newborn bath by an average of 13.5 hours

was associated with a 166% increase in the odds

of breastfeeding initiation (adjusted odds ratio

[aOR] ¼ 2.66; 95% confidence interval [CI] [1.29,

5.46]) and a 39% increase in odds of exclusive,

in-hospital breastfeeding (aOR ¼ 1.39; 95% CI

JOGNN, 49, 181–189; 2020. https://doi.org/10.1016/j.jogn.2019.

[1.02, 1.91]). Dicioccio et al. (2019) reported that

delaying the newborn bath by a median of 17.9

hours was associated with a 49% increase in the

odds of in-hospital exclusive breastfeeding

(OR ¼ 1.49; 95% CI [1.14, 1.96]; p ¼ .004). In

contrast, Suchy et al. (2018), Turney et al. (2019),

and Chamberlain et al. (2019) reported no sig-

nificant changes in rates of exclusive breast-

feeding when bathing was delayed by 12, 13.7,

and 24 hours, respectively.

Similarly, inconsistent results with regard to the

effect of delayed bathing on newborn thermo-

regulation have been reported. A number of re-

searchers found no significant difference in

newborn temperatures when bathing was

delayed by up to 12 hours (Kelly et al., 2018;

Suchy et al., 2018). In contrast, Chamberlain

et al. (2019) and Brogan and Rapkin (2017) re-

ported the incidence of hypothermic events

significantly decreased (p < .001 and p ¼ .04,

respectively) when bathing was delayed by at

least 24 hours.

Decreases in newborn temperatures are often

associated with hypoglycemia because new-

borns increase their metabolic rates in an attempt

to increase body temperatures (Brogan & Rapkin,

2017). Although findings in published studies

suggested that delayed newborn bathing may

decrease the incidence of hypoglycemia (Lipka &

Schulz, 2012; McInerney & Gupta, 2015), there

are limited data to support this relationship. In

one study, Chamberlain et al. (2019) reported a

decrease in blood glucose levels equal to or less

than 45 (p ¼ .001) when bathing was delayed by

24 hours.

In accordance with the recommendation from the

WHO (2014), our study hospital in Newfoundland

and Labrador (NL), Canada, instituted a policy in

March 2015 to delay newborn bathing for 24

hours. The province of NL consistently reports the

lowest rates of breastfeeding initiation in Canada.

The provincial rate has increased over the past 30

years, from 35.3% in 1986 to 74.2% in 2015

(Perinatal Program Newfoundland and Labrador

[PPNL], 2016), but this continues to be less than

the national average. The 2016 breastfeeding

initiation rate in NL was 77.2% compared with

89.9% throughout Canada (Statistics Canada,

2018). The primary objective of our study was to

determine whether delaying the newborn bath by

24 hours increases the prevalence of breast-

feeding initiation and exclusive breastfeeding at

discharge in healthy full-term and late preterm

12.004 http://jognn.org

Warren, S., Midodzi, W. K., Allwood Newhook, L.-A., Murphy, P., and Twells, L. R E S E A R C H

(34 0/7–36 6/7 weeks gestation) newborns; ex-

amination of the effects of delayed newborn

bathing on the incidences of hypothermia and

hypoglycemia were secondary end points of this

study.

Methods Design and Setting We used a retrospective cohort design to

compare outcomes for newborns born before

and after implementation of the policy to delay

newborn bathing for 24 hours. Ethical approval

for our study was obtained from Memorial Uni-

versity’s Health Research Ethics Authority and

Eastern Health’s Research Proposal Approvals

Committee.

Participants The participants for this study were drawn from

the provincial children’s hospital in NL. The

combined 42-bed (8 birthing and 34 antepartum

and postpartum) Obstetrics and Gynecology unit

of this hospital provides maternity services and

newborn and pediatric health care to the

approximately 200,000 people living in the sur-

rounding area.

We used computer-generated, random sampling

to select newborns from the 6-month period of

June through November 2014 (pre-

implementation group) and from the 6-month

period of July through December 2015 (post-

implementation group). We did not select partic-

ipants during the 3 months immediately before or

after implementation to allow time for staff and

patients to adjust to the new policy and proced-

ure, to minimize the lack of adherence to the

policy, and to insure that the current study was an

accurate reflection of the effects of bath timing

and not the effects of the transition to the new

bathing policy.

On average, 7.9% of newborns born at the study

hospital are born late preterm (34 0/7–36 6/

7 weeks gestation; PPNL, 2016). To ensure that

our sample was reflective of the general popula-

tion, we conducted computer-generated, random

sampling to obtain a proportion of 7.5% late

preterm newborns (103/1,379). This was done to

prevent an overselection of late preterm new-

borns, who are at greater risk to develop

hypothermia and hypoglycemia.

Newborns born during the specified time periods

who were 34 0/7 weeks gestation and older and

JOGNN 2020; Vol. 49, Issue 2

whose mothers were subsequently admitted to

the maternity unit were eligible for participation.

Newborns from multiple gestations who met the

inclusion criteria were also included. We

excluded newborns for whom delaying the bath

or breastfeeding was contraindicated; newborns

at higher risk for the outcomes of interest; new-

borns of women using illicit drugs or on metha-

done treatment; and newborns born to women

with HIV, hepatitis B, hepatitis C, active herpes

simplex virus infection, and methicillin-resistant

Staphylococcus aureus. Newborns whose

mothers were admitted to the intensive care unit,

newborns who were directly admitted to the NICU

from the birthing room, and newborns who were

not bathed in the time frame directed by the

policy (i.e., born before the policy change but

bathed after 24 hours and born after the policy

change but bathed before 24 hours) were

excluded. The flow of study participants is illus-

trated in Figure 1.

Procedures Before implementation of the delayed bathing

policy, nurses transported newborns from the

birthing room to the nursery, bathed them in a tub,

and then wrapped them in warmed blankets or

placed them under a radiant warmer before

returning them to their mothers. On average,

bathingoccurredat 3.5 hoursof life. Subsequent to

the policy change, bathing was typically done by

parents in the mothers’ rooms under the guidance

of a nurse at an average of 30 hours of life. After

bathing, mothers were encouraged to place their

newborns skin to skin, or newborns were wrapped

and given to their mothers.Newborn temperatures

were recorded before bathing, routinely once per

day for healthy full-term newborns and every 4

hours for preterm newborns and those at high risk

of temperature instability.

We abstracted data from electronic medical re-

cords and checked data accuracy against the

PPNL Provincial Perinatal Registry, a computer-

ized database that is used to collect information on

pregnancy outcomes for the province. Quality

assurance and data quality are ensured through

the PPNL’s routine edit checking process on

extracted data, and results are regularly submitted

to the Canadian Institute for Health Information.

Measures Breastfeeding initiation was defined as any action

of placing the newborn to the mother’s breast

during the hospital stay. Exclusive breastfeeding

at discharge meant that the newborn received no

183

Data abstracted from charts of 1409

mother/infant pairs

30 excluded

• Mother using illicit drugs or methadone (n = 16)

• Mother HIV positive (n = 3)

• Mother Hep B positive (n = 1)

• Mother Hep C positive (n = 1)

• Mother MRSA (n = 4)

• Mother with active herpes virus (n = 4)

• Infant born 32 weeks gestation (n = 1)

721 participants born before

policy change (June

through November 2014)

658 participants born after

policy change (July through

December 2015)

te m

t o

n airetirc

n ois

ulc nI

Total of 1379 newborns included in

intention to treat analysis

e g

na hc

ycil o

p ts

o p/er

s P

is yla

na

41 participants

bathed after

24 hours thus

removed

from primary

analysis

113 participants

bathed before

24 hours thus

removed

from primary

analysis

680 participants bathed before

24 hours of life

545 participants bathed after 24

hours of lifesis yla

na y ra

mir P Total of 1225 newborns included in

primary analysis

Figure 1. Flowchart of study participants.

Effects of Delayed Newborn BathingR E S E A R C H

184

formula, water, glucose, or any substance other

than breastmilk during the hospital stay. If

newborn feeding supplementation occurred or

the newborn was given any substance (e.g., for-

mula, glucose, or water) other than breastmilk

and prescribed medications, feeding was recor-

ded as mixed. If no breastfeeding occurred and

the newborn was fed only formula, this was

recorded as exclusively formula fed.

In accordance with the position statement from

the Canadian Pediatric Society (Aziz & Dancey,

2004), we defined newborn hypoglycemia as a

recorded glucose level less than 2.6 mmol/L

(47 mg/dl). Consistent with the WHO’s (1997)

JOGNN, 49, 181–189; 2020. https://doi.org/10.1016/j.jogn.2019.

guideline, we defined newborn hypothermia

as a recorded temperature less than 36.5 �C (97.7 �F).

Analysis Using a power of 80% and a significance level of

.05, the required sample size for the primary

outcomes (breastfeeding initiation and exclusive

breastfeeding at discharge) was determined to

be 700 participants, including 350 before and

350 after the policy change. We determined that

the original sample size was sufficient to examine

the secondary outcome of hypothermia but

insufficient to examine the outcome of hypogly-

cemia. Therefore, we increased the sample size

12.004 http://jognn.org

Table 1: Characteristics of the Study Sample and a Comparison of Newborns Bathed

Before and After 24 Hours of Life

Characteristic

Total (N ¼ 1,225)

Bathed Before

24 Hours

(n ¼ 680)

Bathed After 24

Hours (n ¼ 545)

p Value*n % n % n %

Maternal characteristics

Age in years, mean (SD) 30.31 (5.0) 30.36 (5.1) 30.29 (4.9) .804a

Primiparous 573 46.8 310 45.6 263 48.3 .453a

Prenatal care 1212 99.3 670 99.0 542 99.6 .176b

Preexisting diabetes 11 0.9 5 0.7 6 1.1 .500b

Gestational diabetes 59 4.8 35 5.1 24 4.4 .546b

Labor and birth

Cesarean 342 28.0 177 26.1 165 30.5 .087b

Induction or augmentation 535 43.7 316 46.5 219 40.2 .027b

Epidural 924 75.4 501 73.7 423 77.3 .112b

Newborn characteristics

Gestational age, mean (SD) 39 weeks (1.4) 39 weeks (1.4) 39 weeks (1.4) .399b

Late preterm (<37 weeks) 91 7.4 50 7.4 41 7.5 .913b

Small or large for gestational age 221 18.0 122 17.9 99 18.2 .919b

Breastfeeding supplemented with

formula, glucose or water

156 12.7 91 13.4 65 11.9 .442b

Note. SD ¼ standard deviation. aAnalysis of variance. bPearson chi-square test. *Significant at the p < .05 level.

Warren, S., Midodzi, W. K., Allwood Newhook, L.-A., Murphy, P., and Twells, L. R E S E A R C H

to 1,300. We based the increased sample size

calculations on a power of 80%, a significance

level of .05, and an effect size of 20% as per re-

sults observed by McInerney and Gupta (2015).

We used frequencies and percentages to

describe categoric variables, and we used

means and standard deviations to describe

continuous variables. We used chi-square tests to

determine the association between the predictor

variable (control [early bath] vs. intervention

[delayed bath]) and each outcome variable

(breastfeeding initiation, exclusive breastfeeding

at discharge, hypothermia, and hypoglycemia).

We conducted multivariate logistic regression

analyses to test for differences in odds of

breastfeeding before and after the policy change

while controlling for potential confounding factors

(i.e., gestational age, induction or augmentation,

large or small for gestational age, cesarean birth,

and epidural anesthesia/analgesia) as well as to

test for differences in odds of newborn

JOGNN 2020; Vol. 49, Issue 2

hypothermia or hypoglycemia before and after

the intervention. The level of statistical signifi-

cance was p < .05.

We conducted subgroup analysis on two groups

defined a priori. The high-risk subgroup included

newborns at high risk for hypoglycemia (i.e.,

those born less than or equal to 36 0/7 weeks

gestation, those small or large for gestational

age, and/or those born to women with diabetes).

The average risk subgroup included the total

sample minus the high-risk subgroup (i.e., new-

borns born greater than 36 0/7 weeks gestation,

those average size for gestational age, and/or

those born to women without diabetes).

Results A total of 1,225 newborns met the inclusion

criteria: 680 newborns born before implementa-

tion of the delayed bathing policy and 545 born

after implementation. We found no significant

differences between cohorts with regard to age,

parity, cesarean, epidural analgesia/anesthesia,

185

Table 2: Comparison of Changes in Breastfeeding Initiation and Exclusive Breastfeeding

at Discharge Between Newborns Bathed Before and After 24 Hours of Life Observed in

the Three Study Samples

Sample Outcome

Newborns

Bathed Before

24 Hours

Newborns

Bathed After 24

Hours Multivariate Analysis

n % n % aOR 95% CI p Value*

Total sample (N ¼ 1,225) BF initiation 538/680 79.1 449/545 82.4 1.275 [0.952, 1.708] .103

EBF Discharge 361/674 53.6 324/545 59.4 1.334 [1.049, 1.698] .019

Average riska (n ¼ 888) BF initiation 392/494 79.4 331/394 84 1.433 [1.008, 2.039] .045

EBF Discharge 292/491 59.5 263/394 66.8 1.401 [1.056, 1.857] .019

High riskb (n ¼ 337) BF initiation 146/186 78.5 118/151 78.1 0.948 [0.554, 1.620] .844

EBF discharge 69/183 37.7 61/151 40.4 1.176 [0.711, 1.944] .528

Note. Discrepancies in denominators are due to missing data in electronic medical records; charts with incomplete data were not included in analysis. aOR ¼ adjusted odds ratio controlling for gestational age, induction or augmentation of labor, large or small for gestational age, cesarean birth, and epidural analgesia/anesthesia; BF ¼ breastfeeding; CI ¼ confidence interval; EBF ¼ exclusive breastfeeding. aAverage-risk newborns are those who are born >36 weeks gestation, of average size for gestational age, to women without diabetes. bHigh-risk newborns include those who are born#36 weeks gestation, small or large for gestational age, and/or to women with diabetes. *Significant at the p < .05 level.

Effects of Delayed Newborn BathingR E S E A R C H

186

maternal diabetes, or newborn characteristics

(see Table 1). The proportion of women’s labors

that were induced or augmented with oxytocin in

the early bathing group (less than 24 hours) was

significantly greater (p < .05) than in the delayed

bathing group (more than 24 hours; see Table 1).

Primary End Points: Breastfeeding Initiation and Exclusive Breastfeeding at Discharge In the analysis of the total sample (N ¼ 1,225), in

which we compared 680 newborns bathed before

24 hours with 545 bathed after 24 hours, we

found no difference in breastfeeding initiation af-

ter the policy change (79.1% of 680 vs. 82.4% of

545, p ¼ .103). However, for newborns bathed

after 24 hours, the aOR of exclusive breastfeed-

ing at discharge was 33% greater than the aOR

for newborns bathed before 24 hours (aOR ¼ 1.334; 95% CI [1.049, 1.698]; p ¼ .019; see

Table 2).

The average-risk subgroup (newborns born >36

0/7 weeks gestation, newborns who were

average size for gestational age, and newborns

born to women without diabetes) included a total

of 888 newborns: 494 in the early bathing cohort

and 394 in the delayed bathing cohort. In this

subgroup, when bathing was delayed, the aOR

for breastfeeding initiation was 43% greater

(aOR ¼ 1.433; 95% CI [1.008, 2.039]; p ¼ .045),

JOGNN, 49, 181–189; 2020. https://doi.org/10.1016/j.jogn.2019.

and exclusive breastfeeding at discharge was

40% greater (aOR ¼ 1.401; 95% CI [1.056,

1.857]; p ¼ .019) than in the early bathing group

(see Table 2).

The high-risk subgroup (newborns born #36 0/

7 weeks gestation, small- or large-for-gestational-

age newborns, and/or newborns born to women

with diabetes) included a total of 337 newborns:

186 in the early bathing cohort and 151 in the

delayed bathing cohort. In this subgroup, there

were no significant differences in rates of

breastfeeding initiation or exclusive breastfeed-

ing at discharge with regard to the timing of the

bath (see Table 2).

Secondary End Points: Incidence of Hypothermia or Hypoglycemia When the two bathing groups of the total sample

were compared, delaying newborn bathing by 24

hours was associated with a decrease in the

incidence of hypothermia (p ¼ .007). A similar

association was found in the average-risk sub-

group (p ¼ .045), although there was no associ-

ation of delayed bathing with hypothermia in the

high-risk subgroup (p ¼ .364; see Table 3).

Finally, delayed newborn bathing was associated

with a decrease in the incidence of hypoglycemia

in the total sample and in the high-risk subgroup

(p ¼ .03 and p ¼ .016, respectively; see Table 3).

12.004 http://jognn.org

Among our participants, delaying newborn bathing by 24 hours resulted in an increased prevalence of exclusive breastfeeding and better newborn thermoregulation and

Warren, S., Midodzi, W. K., Allwood Newhook, L.-A., Murphy, P., and Twells, L. R E S E A R C H

We observed no statistically significant difference

in the incidence of hypoglycemia in the average-

risk newborns with regard to the timing of the bath

(p ¼ .073).

glycemic control.

Discussion The WHO (2014) recommended delaying

newborn bathing until 24 hours after birth despite

a dearth of research evidence evaluating this

practice. Published research findings on the ef-

fects of newborn bath timing suggest that

delaying the newborn bath has the potential to

positively affect breastfeeding (Dicioccio et al.,

2019; Preer et al., 2013; Suchy et al., 2018;

Turney et al., 2019) and to decrease the inci-

dence of hypothermia and hypoglycemia (Brogan

& Rapkin, 2017; Chamberlain et al., 2019; Gözen

et al., 2019; Kelly et al., 2018; Suchy et al., 2018).

In our study, delaying newborn bathing until at

least 24 hours after birth was associated with an

increased prevalence of breastfeeding initiation

and exclusive breastfeeding at discharge from

the hospital for average-risk, healthy full-term

newborns. However, when only newborns

considered at high risk were compared, we

observed no association between delayed bath-

ing and breastfeeding initiation or exclusivity. This

result is not unexpected because preterm new-

borns commonly have breastfeeding difficulties

(Medoff-Cooper et al., 2012) and may require

more separation from their mothers to

Table 3: Comparison of Changes in Rates of

Between Newborns Bathed Before and After

Study Samples

Sample Outcome

Newb

Bathed Be

Hou

n

Total (N ¼ 1,225) Hypothermia 9/679

Hypoglycemia 14/677

Average riska (n ¼ 337) Hypothermia 5/491

Hypoglycemia 4/492

High riskb (n ¼ 337) Hypothermia 1/184

Hypoglycemia 10/185

Note. Discrepancies in denominators are due to missing data in el included in analysis. aAverage-risk newborns are those who are born >36 weeks gestation bHigh-risk newborns include those who are born#36 weeks gestation *Significant at the p < .05 level.

JOGNN 2020; Vol. 49, Issue 2

accommodate procedural and medical re-

quirements that interrupt breastfeeding. For all

newborns in our study, we found a statistically

insignificant increase in breastfeeding initiation

and a 33% increase in exclusive breastfeeding at

discharge from the hospital.

Interpretation of these results requires consider-

ation of the definition of breastfeeding initiation

used by the study site and as reflected in our

data. Any action of placing a newborn to the

mother’s breast was recorded as breastfeeding

initiation. This definition does not necessitate

adequate latch, milk transfer, or breastfeeding

success. The policy of the study hospital is to

allow uninterrupted skin-to-skin contact between

the mother and newborn for the first hour after

birth and to assist any mother who intends to

breastfeed to attempt a first breastfeeding during

that time. Because this first attempt to breastfeed

is part of hospital policy, it is likely that most

mothers who intend to breastfeed will have a re-

cord of a breastfeeding initiation attempt during

the first hour after birth, regardless of whether

their newborns were bathed before or after 24

hours. In addition, breastfeeding initiation

Hypothermia and Exclusive Hypoglycemia

24 Hours of Life Observed in the Three

orns

fore 24

rs

Newborns

Bathed After 24

Hours

Chi-square p Value*% n %

1.3 0/542 0 .007

2.1 1/544 0.2 .003

1.0 0/391 0 .045

0.8 0/394 0 .073

0.5 0/151 0 .364

5.4 1/150 0.7 .016

ectronic medical records; charts with incomplete data were not

, of average size for gestational age, to women without diabetes. , small or large for gestational age, and/or to women with diabetes.

187

Implementation of policies to delay newborn bathing by 24 hours may increase breastfeeding rates and reduce rates

of hypothermia and hypoglycemia.

Effects of Delayed Newborn BathingR E S E A R C H

188

typically occurs before early or late bathing and is

thus unlikely to be affected by bathing practices.

Delaying newborn bathing by 24 hours was

associated with a decreased incidence of

newborn hypothermia in the total sample and the

average-risk subgroup. Previous researchers

who studied a 24-hour bathing delay (Brogan &

Rapkin, 2017; Chamberlain et al., 2019) also

found significant decreases in hypothermia.

However, there were no significant differences in

the incidence of hypothermia when delays of

newborn bathing between 1 and 12 hours were

studied (Kelly et al., 2018; Suchy et al., 2018).

These data suggest that a bathing delay of at

least 24 hours may be preferable to decrease the

incidence of neonatal hypothermia. The inci-

dence of hypothermia in the high-risk newborns

in our study was very low regardless of the time of

bathing. A much larger cohort or alternative study

design such as a prospective case-control or

randomized study may be required to detect an

association between bathing and hypothermia in

newborns at higher risk.

Similar to Chamberlain et al. (2019), we found that

delaying newborn bathing by at least 24 hourswas

associated with a decreased incidence of hypo-

glycemia in the total sample and the high-risk

subgroup. Although the incidence of hypoglyce-

mia decreased in the average-risk subgroup, the

incidence was low, and the reduction was not

statistically significant. Larger studies of this

population may be required to determine if

delayed bathing after birth decreases the likeli-

hood of hypoglycemia in healthy term newborns.

Limitations As a retrospective cohort study, our findings indi-

cate an association between delayed newborn

bathing and an increased prevalence of exclusive

breastfeeding at discharge from the hospital and a

decreased incidence of hypothermia and hypo-

glycemia. However, it is not possible to attribute

the increases in breastfeeding exclusivity solely to

an effect of delayed bathing. Researchers sug-

gested that the effects of early newborn bathing

JOGNN, 49, 181–189; 2020. https://doi.org/10.1016/j.jogn.2019.

are not mutually exclusive but are interconnected

in newborn adjustment and potentially further

affected by mother–newborn skin-to-skin contact

and the removal of potentially protective biological

substances such as vernix (Hoath et al., 2006) and

amniotic fluid (Loos et al., 2019; Porter, 2004;

Radhakrishnan, 2012) that are involved in

mother–newborn signaling. The association be-

tween delayed newborn bathing and the

increased breastfeeding rates we observed may

be the result of the increased time spent in skin-to-

skin contact and the maintenance of biological

cues for breastfeeding that result when bathing is

delayed. Attempts were made to collect data

pertaining to skin-to-skin contact, but because of

the inadequate charting of such information, these

data were unavailable for analysis.

Similarly, data related to newborn temperature,

including ambient room temperature, bath water

temperature, specific clothing, and blanketing,

were not recorded and could not be assessed as

potential confounders. We relied on the accuracy

of in-hospital records and were thus limited to the

available information without the ability to verify

the accuracy of the data. Although no other policy

or procedural changes were implemented during

the time of data collection, it is possible that

increasing staff awareness of Baby-Friendly

practices influenced their support and promo-

tion of breastfeeding, which led to erroneous

attribution of breastfeeding increases to the

change in the bathing policy. Finally, differences

in temperature and glucose monitoring proced-

ures between full-term and late preterm new-

borns may have resulted in a potential detection

bias in incidences of hypothermia and hypogly-

cemia between these groups.

Conclusion Delaying newborn bathing by at least 24 hours

increased the prevalence of exclusive breast-

feeding at discharge and decreased the inci-

dence of newborn hypothermia and

hypoglycemia. Whether because of the decrease

in the separation of mother and newborn, the

maintenance of biological cues, or the decrease

in newborn stress, the practice of delaying the

newborn bath may play a role in providing the

best possible conditions for breastfeeding, ther-

moregulation, and glycemic control in healthy

newborns at late preterm and term gestations.

12.004 http://jognn.org

Warren, S., Midodzi, W. K., Allwood Newhook, L.-A., Murphy, P., and Twells, L. R E S E A R C H

Acknowledgment We acknowledge the Janeway Children’s Hospital

Foundation, Newfoundland and Labrador, Canada,

for providing funding for this study.

REFERENCES Aziz, K., & Dancey, P. (2004). Screening guidelines for newborns at risk

for low blood glucose. Paediatric Child Health, 9(10), 723–729.

Brogan, J., & Rapkin, G. (2017). Implementing evidence-based

neonatal skin care with parent-performed, delayed immersion

baths. Nursing for Women’s Health, 21(6), 442–450.

Chamberlain, J., McCarty, S., Sorce, J., Leesman, B., Schmidt, S., Meyr-

ick, E.,… Coultas, L. (2019). Impact on delayed newborn bathing

on exclusive breastfeeding rates, glucose and temperature sta-

bility, and weight loss. Journal of Neonatal Nursing, 25(2), 74–77.

Dicioccio, H., Ady, C., Bena, J., & Albert, N. (2019). Initiative to improve

exclusive breastfeeding by delaying the newborn bath. Journal

of Obstetric, Gynecologic, & Neonatal Nursing, 48(2), 189–196.

Doty, R. L. (2014). Human pheromones: Do they exist? In C. Mucignat-

Caretta (Ed.), Neurobiology of chemical communication (pp.

535–556). Boca Raton. FL: CRC Press.

Gözen, D., Çaka, S., Bes‚ irik, S., & Perk, Y. (2019). First bathing time of

newborn infants after birth: A comparative analysis. Journal for

Specialists in Pediatric Nursing, 24(2). e12239.

Hoath, S. B., Pickens, W. L., & Visscher, M. O. (2006). The biology of

vernix caseosa. International Journal of Cosmetic Science, 28,

319–333.

Kelly, P., Classen, K., Crandall, C., Crenshaw, J., Schafer, S., Wade, D.,

… Fossee, K. (2018). Effect of timing of the first bath on a

healthy newborn’s temperature. Journal of Obstetric, Gyneco-

logic, & Neonatal Nursing, 47(5), 608–619.

Lipka, D., & Schulz, M. (2012). “Wait for eight”: Improvement of

newborn outcomes by the implementation of newborn bath

delay. Journal of Obstetric, Gynecologic, & Neonatal Nursing,

41(S1), S46–S47.

Loos, H., Reger, D., & Schaal, B. (2019). The odour of human milk: Its

chemical variability and detection by newborns. Physiology &

Behavior, 199, 88–99.

JOGNN 2020; Vol. 49, Issue 2

McInerney, C. M., & Gupta, A. (2015). Delaying the first bath de-

creases the incidence of neonatal hypoglycemia. Journal of

Obstetric, Gynecologic, & Neonatal Nursing, 44, S73–S74.

https://doi.org/10.1111/1552-6909.12650

Medoff-Cooper, B., Holditch-Davis, D., Verklan, M., Fraser-Askin, D.,

Lamp, J., Santa-Donato, A., … Bingham, D. (2012). Newborn

clinical outcomes of the AWHONN late preterm infant research-

based practice project. Journal of Obstetric, Gynecologic, &

Neonatal Nursing, 41(6), 774–785.

Perinatal Program Newfoundland and Labrador. (2016). Provincial

neonatal screening program. St. John’s, Newfoundland and

Labrador, Canada: Eastern Health.

Porter, R. H. (2004). The biological significance of skin-to-skin contact

and maternal odours. Acta Paediatrica, 93, 1560–1562.

Preer, G., Pisegna, J. M., Cook, J. T., Henri, A.-M., & Philipp, B. L.

(2013). Delaying the bath and in-hospital breastfeeding rates.

Breastfeeding Medicine, 8(6), 485–490.

Radhakrishnan, S. A. (2012). Breast crawl. Asian Journal of Nursing

Education and Research, 2(1), 21–24.

Rissmann, R., Groenink, H., Gooris, G., Oudshoorn, M., Hennink, W.,

Ponec,M., &Bouwstra, J. (2008). Temperature-inducedchanges

in structural and physicochemical properties of vernix caseosa.

Journal of Investigative Dermatology, 128(2), 292–299.

Statistics Canada. (2018). Canadian Community Health Survey

(CCHS). Retrieved from https://www150.statcan.gc.ca/n1/en/

catalogue/82M0024X

Suchy, C., Morton, C., Ramos, R., Ehrgott, A., Quental, M., Burridge,

A., & Rutledge, D. (2018). Does changing newborn bath pro-

cedure alter newborn temperatures and exclusive breastfeed-

ing? Neonatal Network, 37(1), 4–10.

Turney, J., Lowther, A., Pkya, J., Mollon, D., & Fields, W. (2019).

Delayed newborn first bath and exclusive breastfeeding rates.

Nursing for Women’s Health, 23(1), 31–37.

World Health Organization. (1993). Thermal control of the newborn, a

practical guide. Geneva, Switzerland: Author.

World Health Organization. (1997). Thermal protection of the newborn:

A practical guide. Geneva, Switzerland: Author.

World Health Organization. (2014). Recommendations on postnatal

care of the mother and newborn—2013. Retrieved from https://

apps.who.int/iris/bitstream/handle/10665/97603/978924150664

9_eng.pdf;jsessionid¼0D02140AF065B70198F35DCCFF6F433

F?sequence¼1

189