Mother’s Milk Donation to a Human Milk Bank During Bereavement: Circumstances Associated with Completing the Donation Process

Abstract

Background:

Bereaved mothers describe positive experiences donating breast milk and negative experiences when not informed of opportunities to donate. Predictors of whether mothers donate milk are unknown, impairing efforts to optimize support in completing donation.

Objective:

To define circumstances associated with completing mother’s milk (MM) donation during bereavement.

Methods:

A retrospective cohort study included dyads of bereaved mothers and their deceased children if a child’s death occurred on-site at a quaternary care children’s hospital during 2016–2020, the child had documentation of MM availability, and age at death <24 months. The primary outcome was the completion of MM donation to the milk bank. Multivariate logistic regression measured associations between clinical variables and odds of completion.

Results:

Of 124 deceased children with documented MM exposure, 34 mothers (28%) of 35 of those children completed MM donation, donating a mean of 13.7 liters (SD 16.8). The child’s race/ethnicity documented in the medical record was White for 25 (71%), Black/African American (AA) for 1 (3%), Asian for 1 (3%), and Hispanic/Latino for 8 (23%). Referenced to mothers of White children, being a mother of an AA [OR 0.05 (95% CI: 0.01–0.43)] or Asian [0.08 (0.01–0.75)] child was associated with lower odds of donation. Referenced to mothers delivering full term (≥37 weeks’), mothers delivering <34 weeks showed higher odds [5.0 (1.5–17.5)] of donation.

Conclusion:

Relatively few bereaved mothers of children with indicators of MM exposure completed donation. The results suggest an opportunity to ensure bereaved mothers are uniformly informed and supported in donating.

Introduction

The circumstances for most women donating breast milk include having an excess supply of frozen and stored milk in the context of having healthy, older infants who would likely not need the full supply for their own nutrition.¹ However, lactating women who experience the death of a child are also eligible to donate to a milk bank accredited by the Human Milk Banking Association of North America (HMBANA). Bereaved mothers may undergo a staged screening to minimize burden during the donation process when possible. Regardless of medical history and screening outcome, a bereaved mother’s donation is accepted by HMBANA banks for nutrition or research purposes.

Bereaved mothers express positive experiences resulting from the chance to donate, describing it as helpful with their grieving process.^2–7 Commonly reported benefits of donation during bereavement include pride in doing something positive and finding meaning amid loss.^2,3,6 The process may also provide an opportunity to memorialize a child through ritual or shared artistic efforts with the partnering milk bank, such as a memorial wall.^3–5 In that context, some women provide negative feedback if not provided with sufficient education about milk donation, wishing they had more information at the time of their child’s death.⁶ Despite the potential benefits of donation, the clinical circumstances associated with the completion of milk donation by bereaved parents remain unknown.

Our medical center, a quaternary care children’s hospital, developed a hospital-wide program for informed decision-making regarding milk donation during bereavement. Anecdotally, it was identified that some mothers initiated the donation process by filling out the screening information yet did not complete the donation process. Therefore, the main objective of this study was to report the clinical characteristics associated with milk donation completion to an HMBANA-accredited milk bank. Understanding these characteristics will inform clinicians in their discussions and support of women and families during and after the bereavement period.

Materials and Methods

A retrospective cohort study enrolled dyads of bereaved mothers and their deceased children (IRB#2018-1530). Dyads were included if a child’s death occurred on-site at a Chicago quaternary care children’s hospital during 2016–2020, the electronic medical record (EMR) indicated the child’s mother was expressing breast milk, and the age at death was <24 months. Determining eligibility utilized indicators of mother’s milk exposure through either of the following conditions: 1. during the admission in which death occurred, the child’s EMR showed orders for bottle labels for mother’s milk, and 2. the local, HMBANA-accredited milk bank documented a mother initiating donation. Dyads were excluded if a child’s death occurred outside the children’s hospital.

Clinical setting and program for bereaved donors

At this children’s hospital, bereaved lactating mothers were informed of the donation process available through the local milk bank at the time of an infant’s death, along with the memory-making process for the infants. If the death was anticipated, discussions could have been initiated before the child’s death. There were also serial check-ins to the family provided by the hospital family service team for several weeks following the infant’s death, in which parents were offered resources to help support them through the grieving process. During these communications, families were reminded about the breast milk donation program.

This program was implemented with input and ongoing support from the Mothers’ Milk Bank of the Western Great Lakes (MMBWGL), the HMBANA-accredited milk bank that serves Illinois and Wisconsin. The milk bank processes and distributes pasteurized donor human milk (PDHM) for inpatient feedings when medically indicated. At this milk bank, the disposition of milk from bereaved donors is determined by staff through standardized medical screening. Milk disposition is designated as either of the following: (1) safe for pasteurization and feeding other infants and children or (2) not suitable for feeding, and the milk is used for research and milk bank system support, such as machine calibration.

Data sources for this study included the hospital EMR and the milk bank database. The following details were abstracted from those sources: donation status (not initiated after child’s death, initiated/not completed, or completed); milk volumes obtained by the milk bank; final determination of how milk was utilized (pasteurized for infant feedings, research utilization); and dyad clinical characteristics. Dyad clinical characteristics included demographic data, perinatal history, and details surrounding the infants’ deaths. Demographics included age, race, and ethnicity, as determined from the EMR of the deceased infant and/or through maternal screening forms in the milk bank database. Race/ethnicity documented in the EMR was recorded from questions asked of the parent (i.e., self-report). Perinatal information included maternal primigravid status, delivery mode, and gestational age at delivery. Details surrounding infant death included the hospital unit in which death occurred [neonatal intensive care unit (NICU), pediatric ICU, cardiac ICU, or emergency department]; whether a palliative care consult occurred during admission, as determined by the presence of a palliative care note in the EMR; and if cardiopulmonary resuscitation (CPR) was performed immediately preceding death, as documented in the EMR. Palliative care consultation and end-of-life practices were included due to being plausible indicators of discussions of milk donation.

The primary outcome was the status of having completed the milk donation process, as confirmed in the milk bank database. Secondary outcomes included milk volumes of donations and dispensation of donated milk (volumes ultimately used for infant feedings after processing or utilized for the MMBWGL research program).

Statistical analysis

Analyses compared distributions of maternal and infant characteristics based on milk donation status, whether milk donation was completed or not. Group comparisons utilized two-tailed t-tests or Chi-square to assess for differences between groups for maternal age, infant age at death, gestational age at delivery, as categorized by degree of prematurity (full term, delivery ≥37 weeks of gestation; late preterm, delivery 34–36 weeks’; and early preterm, delivery <34 weeks’), infant race, infant ethnicity, and location of death (comparing all inpatient units in reference to the NICU). A multivariate logistic regression model tested for associations between clinical variables and donation status (dependent categorical variable). The multivariable model included variables that were significantly different in group comparisons or variables with plausible relevance to donation, including palliative care consultation and performance of CPR. Analysis was completed using Stata SE 17 software.

Results

Of 124 deceased infants in the enrollment period, 34 mothers of 35 deceased infants completed the milk donation process. Most deaths occurred during infancy and within the NICU (Table 1). The milk bank received a total of 467.2 L from 34 mothers, with mothers donating a mean of 13.7 (SD 16.8) L. The milk from 19 mothers was utilized for research exclusively, while the milk from 13 mothers was utilized for research and feeding. In total, 268.9 L were used for research-only purposes, while 198.3 L were used for feeding other infants. Reasons for research designation included contraindications related to maternal medications, expiration dates, and labeling errors, among others.

Table 1.

Characteristics of Bereaved Mothers and Deceased Children Based on Completion of Milk Donation Process

	Completed milk donation	Did not complete milk donation	p value
	n = 34^a	n = 89	p value
Maternal age, years	32.1 ± 4.2	30.1 ± 6.1	0.08
Maternal primigravida	5 (15)	21 (24)	0.21
	n = 35^a	n = 89
Delivery method			0.08
Vaginal delivery	9 (26)	37 (42)
Cesarean delivery	25 (71)	47 (53)
Infant age at death, days	71 ± 82	94 ± 122	0.32
Gestational age at delivery, weeks	32.8 ± 5.8	34.9 ± 5.2	0.06
Gestational age category			0.019
Full term (≥37 weeks)	12 (34)	45 (51)
Late preterm (34–36 weeks)	3 (9)	17 (19)
Early Preterm (<34 weeks)	20 (57)	27 (30)
Infant race^b			<0.001
Black/African American	1 (3)	21 (24)
Asian	1 (3)	13 (15)
Other	6 (17)	18 (20)
White	25 (71)	30 (34)
Infant ethnicity^b			0.869
Not Hispanic/Latino	24 (69)	61 (69)
Hispanic/Latino	8 (23)	22 (25)
Inpatient unit of child’s death			0.306
Pediatric intensive care unit (ICU)	2 (6)	16 (18)
Cardiac ICU	11 (31)	22 (25)
Neonatal ICU	22 (63)	50 (56)
Emergency department	0 (0)	1 (1)
Palliative care consultation documented during admission	14 (40)	42 (47)	0.47
Cardiopulmonary resuscitation provided	4 (11)	17 (19)	0.31

Values are reported as mean ± SD or n (%) as appropriate. Percentages may not total to 100 due to missing values.

Numbers of maternal and infant participants reflect one mother having a multiple gestation pregnancy.

Infant race and ethnicity were self-reported by parent and recorded in infant’s medical record (n = 9 missing for both).

Mothers completing donations delivered at earlier gestational ages and infants were predominantly White, as documented in the EMR (Table 1). However, there was no significant difference in the completion of donation for maternal age, infant age at death, or inpatient unit of infant’s death. Furthermore, there was no significant difference in the completion of donation if palliative care was consulted during the infant’s admission or if CPR was performed before the infant’s death.

In the multivariable logistic regression model, the odds for completing donation were higher for mothers whose infants were born early preterm at <34 weeks of gestation with no association for those delivering late preterm (Table 2). The odds of completing donation were significantly lower for mothers of infants documented to be Black/African American and Asian compared to mothers of White infants. Neither the hospital unit where the infant’s death occurred nor palliative care consultation was associated with the odds of completing donation.

Table 2.

Multivariable Model Showing Odds Ratio for Completion of Milk Donation by Bereaved Mothers

	Odds ratio	95% confidence interval	p value
Maternal primigravid status
Not Primigravid	Reference	Reference	Reference
Primigravid	0.33	[0.09, 1.16]	0.09
Gestational age category
Full term (≥37 weeks)	Reference	Reference	Reference
Late preterm (34–36 weeks)	0.89	[0.18, 4.51]	0.89
Early Preterm (<34 weeks)	5.03	[1.45, 17.47]	0.01
Infant race
White	Reference	Reference	Reference
African American	0.049	[0.006, 0.428]	0.006
Asian	0.083	[0.009, 0.749]	0.027
Other	0.273	[0.073, 1.024]	0.054
Infant ethnicity
Not Hispanic/Latino	Reference	Reference	Reference
Hispanic/Latino	1.24	[0.26, 5.97]	0.79
Inpatient unit of child’s death
All other inpatient units	Reference	Reference	Reference
Neonatal intensive care unit	0.72	[0.22, 2.34]	0.59
Palliative care consult documented during admission
No	Reference	Reference	Reference
Yes	0.77	[0.26, 2.22]	0.63
End-of-life care practice
Cardiopulmonary resuscitation (CPR) not performed	Reference	Reference	Reference
CPR performed	0.46	[0.10, 2.10]	0.32

Multivariable logistic regression model includes infant race, maternal primigravid status, gestational age category, inpatient unit of child’s death, palliative care consultation, and end-of-life care practice.

Discussion

In this analysis of milk donation during bereavement, most mothers with indicators of ongoing milk expression in their deceased infant’s EMR did not complete donation to the HMBANA bank in closest proximity to the hospital in which the child’s death occurred. Birth of the infant at earlier gestational ages was strongly associated with donation. This may suggest that the parent appreciated the implications of human milk for preterm infant health, having directly experienced discussions with healthcare providers about the utilization of MM/PDHM for their preterm infant. Given the widespread expansion of PDHM use for preterm infant feedings, conversations on the health benefits are common in most NICUs.^8–10

The null finding for the association of bereavement donation with hospital unit could represent consistency across units in the process of informing bereaved mothers of the option to donate. The development of the hospital’s bereavement program included informational sessions for staff across multiple hospital units, bringing shared awareness of milk donation options. In addition, services that are available to all units, such as chaplaincy, palliative care, and family services, may serve to equalize education surrounding the opportunity to donate milk after loss. This result may also reflect that most deaths occurred primarily in the NICU and cardiac ICU.

MMBWGL promotes widespread knowledge of donation options for healthcare providers and families. Milk bank efforts to support bereaved families in the hospital setting include a streamlined application process, coordinated milk pickups, and bereavement donation education. The milk bank invites bereaved families to participate in a memorial program and special events. These measures attend to each family’s individual need for managing stored milk, providing meaning to their loss, finding community, and honoring their infant’s legacy.

Of concern, associations between donation status and race documented in the EMR suggest the possibility of inconsistent support through the donation process. For preterm infants born at a very low birth weight, disparities in exposure to mother’s milk and donor milk occur in NICUs, specifically with lower exposure for infants born to Black/African American mothers.^11,12 Disparities related to donor milk utilization among well babies have also been found, identifying non-Hispanic Black women as the least likely to use PDHM.¹³ Feeding disparities are multifactorial, and contributors include unequal provision of lactation support to parents with a minoritized racial identity, neglect, stigmatization, systemic barriers, and provider bias.¹¹ With that background, our findings suggest the need for further interrogation of the system to ensure that the process is being explained to all families in a culturally congruent manner and that both hospitals and milk banks provide equitable support throughout the donation process.

MMBWGL implemented an equity program to reduce disparities in milk bank services, utilizing a ten-step framework that includes developing community partnerships, engaging in public health research, and partnering with safety net hospitals. These strategies are shared among all HMBANA member milk banks in a toolkit created to reduce inequities in milk banking.¹⁴ Milk banks and hospitals can utilize strategies in the toolkit, in addition to information gleaned from this study, to develop culturally congruent programming.

No matter what the basis for donation, an ongoing supply of milk donations to HMBANA-accredited milk banks is an important consideration. Public health guidelines recommend PDHM feedings for hospitalized preterm and critically ill infants when mother’s milk is not available.^8,10 To this end, inpatient prescribing of PDHM has significantly increased in the past years.^15–16 Supplementation of MM with PDHM allows an infant’s diet to be composed predominantly or entirely of human milk during high-risk periods. Over two-thirds of dispensed DM is used in NICUs.¹⁷ A mean volume of 13.7 L donated by a bereaved mother is quite substantial and has great potential to meaningfully contribute to research and feed other high-risk infants.

In considering study limitations, this report measured a single center’s experience, which impacts generalizability. Also, it is possible that donations by some of these bereaved mothers occurred through other HMBANA-accredited milk banks. This study did not address donation when a child’s death occurred outside of the hospital, such as in the home with palliative or hospice care management plans, as these circumstances may be managed by providers outside of our medical system.

Although many mothers have reported milk donation to be a positive experience that helps with the grieving process, donation is unlikely to be universally perceived as helpful by bereaved parents. There are no data available to suggest what an expected rate of donation is or should be. Similarly, there is room for a similar approach with lactation support following bereavement. Lactation suppression and breastfeeding cessation can be a challenging process for newly bereaved mothers, and it is important to provide lactation support to these mothers, irrespective of donation of milk.¹⁸ As a part of the process, it may be valuable for staff to include a personalized approach to discussing the options within lactation suppression and donation. This study did not include interviews with bereaved mothers to gain their perspectives on why they either chose or did not choose to donate. Future research in this area may help guide processes to support mothers through their lactation options after perinatal loss.

Conclusion

While bereaved mothers may find benefit in knowing about the option to donate breast milk as well as from the donation process itself, findings from a single-center cohort of bereaved mothers and their deceased children identify distinctions between mothers completing donation versus those who do not. These distinctions emphasize the need to ensure that bereavement programming is developed using culturally congruent strategies for equitable education and support. The establishment of a hospital-wide system in partnership with a nonprofit milk bank should ensure that all mothers, including mothers at risk for reduced lactation support, are informed of their options after loss. At the same time, sensitivity is required to understand that bereavement donation is a personal choice, and in some cases, donation does not further benefit the mother through the grieving process. Most importantly, understanding strategies to best enhance this process would likely be optimized with direct input from bereaved parents themselves.

Footnotes

Authors’ Contributions

M.P.: Methodology (equal), formal analysis (lead), and writing—original draft; G.K.: Conceptualization (equal), methodology (equal), and investigation (equal); N.H.: Conceptualization (equal); J.F.: Conceptualization (equal); S.K.: Conceptualization (equal) and methodology (equal); J.H.: Conceptualization (equal) and resources (lead); D.R.: Conceptualization (equal), methodology (equal), and supervision; and all authors: writing—review and editing (equal).

Disclosure Statement

D.R. is an institutional principal investigator, with no salary funding, for a consortium database sponsored by Mead Johnson Nutrition, previously received research funds from Fresenius Kabi, separately was a member of the Data Safety Monitoring Board for a clinical investigation sponsored by Fresenius Kabi, and was a consultant and speaker for Baxter. The other authors have no disclosures to report.

Funding Information

No funding support was received for the development or conduct of this study or article writing.

References

Gutierrez Dos Santos

, Perrin

. What is known about human milk bank donors around the world: A systematic scoping review. Public Health Nutr, 2022; 25(2):312–322; doi: 10.1017/S1368980021003979

Ward

, Adair

, Doherty

, et al. Bereaved mothers' experience of expressing and donating breast milk: An interpretative phenomenological study. Matern Child Nutr, 2023; 19(3):e13473; doi: 10.1111/mcn.13473

Paraszczuk

, Candelaria

, Hylton-McGuire

, et al. The voice of mothers who continue to express milk after their infant’s death for donation to a milk bank. Breastfeed Med, 2022; 17(8):660–665; doi: 10.1089/bfm.2021.0326

Eidelman

. The ritual value of mothers expressing and donating breast milk after their infant’s death. Breastfeed Med, 2022; 17(8):631–632; doi: 10.1089/bfm.2022.29218.editorial

Carroll

, Lenne

, McEgan

, et al. Breast milk donation after neonatal death in Australia: A report. Int Breastfeed J, 2014; 9(1):23; doi: 10.1186/s13006-014-0023-4

Welborn

. The experience of expressing and donating breast milk following a perinatal loss. J Hum Lact, 2012; 28(4):506–510; doi: 10.1177/0890334412455459

Tully

. Donating human milk as part of the grieving process. J Hum Lact, 1999; 15(2):149–151; doi: 10.1177/089033449901500217

Meek

, Noble

, Section on Breastfeeding. Policy statement: Breastfeeding and the use of human milk. Pediatrics, 2022; 150(1):e2022057988; doi: 10.1542/peds.2022-057988

Quigley

, Embleton

, McGuire

. Formula versus donor breast milk for feeding preterm or low birth weight infants. Cochrane Database Syst Rev, 2019; 7(7):CD002971; doi: 10.1002/14651858.CD002971.pub5

10.

Parker

, Stellwagen

, Noble

, et al. Promoting human milk and breastfeeding for the very low birth weight infant. Pediatrics, 2021; 148(5):e2021054272; doi: 10.1542/peds.2021-054272

11.

Palmquist

AEL

, Asiodu

, Tucker

, et al. Racial disparities in donor human milk feedings: A study using electronic medical records. Health Equity, 2022; 6(1):798–808; doi: 10.1089/heq.2022.0085

12.

Boundy

, Perrine

, Nelson

, et al. Disparities in hospital-reported breast milk use in neonatal intensive care units—United States, 2015. MMWR Morb Mortal Wkly Rep, 2017; 66(48):1313–1317; doi: 10.15585/mmwr.mm6648a1

13.

Kair

, Nidey

, Marks

, et al. Disparities in donor human milk supplementation among well newborns. J Hum Lact, 2020; 36(1):74–80; doi: 10.1177/0890334419888163

14.

Human Milk Banking Association of North America. Member toolkit to advance the equitable distribution of donor milk. 2024. Available from: https://www.hmbana.org/file_download/inline/a06d0830-3fc3-44e9-b064-b02afde4cc66 [Last accessed: February 27, 2024].

15.

Updegrove

. Donor human milk banking: Growth, challenges, and the role of HMBANA. Breastfeed Med, 2013; 8(5):435–437; doi: 10.1089/bfm.2013.0079

16.

Perrine

, Scanlon

. Prevalence of use of human milk in US advanced care neonatal units. Pediatrics, 2013; 131(6):1066–1071; doi: 10.1542/peds.2012-3823

17.

Parker

, Barrero-Castillero

, Corwin

, et al. Pasteurized human donor milk use among US level 3 neonatal intensive care units. J Hum Lact, 2013; 29(3):381–389; doi: 10.1177/0890334413492909

18.

Carroll

, Noble-Carr

, Sweeney

, et al. The “Lactation After Infant Death (AID) Framework”: A guide for online health information provision about lactation after stillbirth and infant death. J Hum Lact, 2020; 36(3):480–491; doi: 10.1177/0890334420926946