All-Cause Severe Maternal Morbidity (SMM) and Transfusion-Only SMM Are Independently Associated with a Lower Likelihood of Exclusive Breastfeeding

Abstract

Background:

The rates of severe maternal morbidity (SMM) including blood transfusions after delivery are rising, yet little is known about the impact of these experiences on breastfeeding.

Materials and Methods:

This is a single-institution retrospective cohort study examining breastfeeding rates at three time points for 1,857 first-time parents delivered at term between July 1, 2016 and June 30, 2019. Our exposure of interest was SMM, which was subdivided into SMM where transfusion was the only indicator (transfusion-only SMM) and SMM where another indicator (diagnostic or procedural) was met, which may also include transfusion (all-cause SMM). Association between transfusion-only SMM and all-cause SMM with feeding method was determined using multinomial regression modeling and adjusting for relevant sociodemographic characteristics.

Results:

The majority of those with uncomplicated deliveries were exclusively breastfeeding at the 2- to 4-week and 2- to 3-month time points (59.6% and 53.6%, respectively), in contrast to 46.3% and 42.0% of those who had experienced transfusion-only SMM, and 40.9% and 30% of those who had experienced all-cause SMM. In adjusted models, receipt of a blood transfusion was found to be associated with greater risk of exclusive formula feeding at all time points. Experience of all-cause SMM was significantly associated with increased likelihood of exclusive formula feeding at hospital discharge and the 2- to 3-month time point.

Conclusions:

We identified that experience of all-cause SMM and transfusion-only SMM are independently associated with a lower likelihood of exclusive breastfeeding after adjusting for sociodemographic factors. Perinatal clinicians should be aware of these risks and offer increased support to these couplets.

Background

Human breast milk is the optimal food for infants and the American College of Obstetricians and Gynecologists and the American Academy of Pediatrics both recommend exclusive breastfeeding for the first 6 months of an infant's life, with continued breastfeeding for at least 1 year with introduction of complementary foods.^1,2 The physiological and psychological benefits of breastfeeding for both mother and child are well established.^3–5 Although most pregnant individuals intend to breastfeed their newborn, both personal, medical, and larger societal factors pose barriers to optimal population rates of breastfeeding.⁶

Concurrent with the public health challenge of suboptimal breastfeeding rates are increasing rates of maternal morbidity and mortality in the United States.^7–9 The Centers for Disease Control and Prevention define Severe Maternal Morbidity (SMM) as “the unexpected outcomes of labor and delivery that result in short- or long-term consequences to a woman's health.”¹⁰ SMM includes conditions such as organ failure (respiratory, renal, and cardiac), eclampsia, sepsis, coagulation defects (disseminated intravascular coagulopathy), and venous thromboembolism as well as emergency procedures, including blood transfusion, use of mechanical ventilation, and hysterectomy. The Centers for Disease Control and Prevention (CDC) monitors rates of SMM in the population utilizing specific International Classification of Diseases diagnosis and procedural codes that when present in a hospitalization in which childbirth occurred constitute SMM.

Concerningly, the rate of SMM increased almost 200% during the period from 1993 to 2014 (from 49.5 per 10,000 in 1993 to 144 per 10,000 in 2014), an increase driven largely by blood transfusions. Nontransfusion SMM also increased by >20% during this same time period.¹⁰ Many factors likely contribute to rising rates of SMM, ranging from increasing population comorbidities to increased cesarean deliveries.¹⁰ The consequences of increasing rates of SMM are far-reaching and include increased medical costs, longer hospitalizations, and other downstream health effects for the individual. Some birth complications will exert a lasting physiological and psychological toll on a person's body, and it is plausible to conclude that one impact may be a reduced ability to breastfeed.

Complications arising during pregnancy and childbirth are an established risk factor for not breastfeeding. Kozhimannil et al. found that people with medically complex pregnancies (including diabetes, hypertension, or obesity), had >30% lower odds of exclusive breastfeeding at 1 week postpartum than those who had noncomplex pregnancies.¹¹ Brown et al. examined the effects of several peripartum complications, including cesarean section, dysfunctional labor, and postpartum hemorrhage, and found each of these to be associated with shorter breastfeeding duration.¹² There is, however, little published literature specifically examining the association between SMM, as defined by CDC, with breastfeeding outcomes. Transfusion is the single largest contributor to SMM and is thus often reported separately from the other indicators.

A few studies have shown that people who require blood transfusions after childbirth, the vast majority of which are for severe hemorrhages, have lower rates of breastfeeding.^13,14 However, Furuta et al. found no association between experience of SMM, which included severe hemorrhage, and breastfeeding outcomes at 6–8 weeks postpartum except for those who had hypertensive disorders,¹⁵ which is contradictory to studies of transfusions alone. Additional studies are needed to provide an accurate estimate of the association between SMM and breastfeeding outcomes to inform clinical care and counseling of people who experience these complications and have a desire to breastfeed. Furthermore, no study, to our knowledge, has examined the independent contributions of SMM where transfusion was the only indicator (transfusion-only SMM) and SMM based on another indicator was met (all-cause SMM).

With increasing public health and health care attention on SMM, it is important to fully understand the short- and long-term impact of these events on the overall health of people who have given birth and are transitioning into new parenthood. The aim of our study was to examine the association between the experience of maternal morbidity further characterized as transfusion-only SMM and all-cause SMM and breastfeeding outcomes among first-time parents who delivered a singleton infant at term (37 weeks of gestation or greater). We divided our exposure of SMM into these subcategories given the rapid increase in transfusion-only SMM in the United States in recent years. Our hypothesis was that those who experienced SMM would have lower rates of breastfeeding as compared with those who did not experience SMM.

Methods

Study population

This retrospective cohort study was derived from the Iowa Perinatal Health Research Collaborative (PHRC) database, a resource composed of linked maternal and newborn electronic health record (EHR) data for deliveries occurring at a single Midwestern tertiary care unit. EHR data within this database were provided by the Institute for Clinical and Translational Science (ICTS) Bioinformatics Core at the University of Iowa. The retrospective study was approved for a waiver of consent by the University of Iowa Institutional Review Board (IRB no. 202001515). There were 6,829 deliveries between July 1, 2016 and June 30, 2019.

We excluded 298 multiple gestations, 4,077 multiparous birthing people, 450 preterm (<37 weeks) deliveries, and 139 deliveries where the infant was admitted to the neonatal intensive care unit (NICU) (Fig. 1). Breastfeeding information at discharge, the 2- to 4-week follow-up and the 2- to 3-month follow-up were abstracted by looking at discharge and after visit summaries for each of the three time points. The breastfeeding outcome was defined at each time point as exclusive breastfeeding, formula only, or a combination of breastfeeding and formula. Breastfeeding information was available for 1,857 infants at discharge, 1,367 infants at the 2- to 4-week well-child visit and 1,285 infants at the 2- to 3-month well-child visit.

FIG. 1.

Description of study population.

Our main exposure of interest was SMM, which we categorized into those whose only SMM indicator was receipt of a blood transfusion (transfusion-only SMM), those whose indicator was something other than transfusion (all-cause SMM), and those with no SMM indicator. All-cause SMM was identified following the CDC's diagnostic and procedural codes, cases further reviewed in accordance with the American College of Obstetricians and Gynecologist and Society for Maternal Fetal Medicine's joint obstetrical consensus statement on SMM¹⁶ to ensure accuracy of diagnostic coding. Transfusion-only SMM was based on if any units of blood products were transfused and no additional SMM indicator was present.

We defined multiple sociodemographic and clinical characteristics to include as covariates in the models. Race/ethnicity was grouped in the following categories: Non-Hispanic White, Black, Hispanic, Asian, including Native Hawaiian or Pacific Islander, Multiracial, and not reported. Of the 16 that were not reported the majority (88%) declined to provide racial/ethnic information and the remaining were missing this information in their medical record. The insurance status at time of birth was classified as private or public, which included seven individuals for whom insurance information was missing and were likely self-pay. Age of the birthing parent was categorized as ≤18 years, 19–24 years, 25–34 years, 35–39 years, and ≥40 years.

Statistical analysis

The association between all-cause SMM, transfusion-only SMM, and breastfeeding outcome (defined in three categories) was determined using unadjusted and adjusted multinomial logistic regression models at discharge, the 2- to 4-week follow-up, and the 2- to 3-month follow-up. Covariates included in the adjusted model were parental age, race, and insurance. The relative risk ratios are presented for all analyses. Analyses were performed using Stata/SE (StataCorp LP, College Station, TX).

Results

Demographics for the study population are shown in Table 1. Most of the population was non-Hispanic White (73.5%) with an age between 25 and 34 years (65.5%). The mean age of the birthing parent was 28.2 years. 78.0% of our study population had private insurance coverage. Actual feeding modalities for the groups are shown in Figure 2. At hospital discharge, 40.9% of those with uncomplicated deliveries were exclusively breastfeeding compared with 29.2% who had transfusion-only SMM and 20.0% who had all-cause SMM. Among the group who experienced uncomplicated deliveries, the majority were exclusively breastfeeding at the 2- to 4-week and 2- to 3-month time points (59.6% and 53.6%, respectively). In contrast, only 46.3% and 42.0% of those who had experienced transfusion-only SMM were exclusively breastfeeding at the 2- to 4-week and 2- to 3-month time points, respectively. For those who had experienced all-cause SMM, 40.9% and 30% were exclusively breastfeeding at the 2- to 4-week and 2- to 3-month time points, respectively.

FIG. 2.

Unadjusted infant feeding modalities at three time points, hospital discharge, 2- to 4-week well-child visit, and 2- to 3-month well-child visit stratified by maternal experience of (a) all-cause SMM, (b) transfusion-only SMM, or (c) none. Exclusive BF describes couplets where the infant's diet is entirely maternal breast milk, which may include bottle feeding of expressed breast milk. Supplementation describes couplets where the infant's diet is partially maternal breast milk and partial supplementation, which could include infant formula or donor breast milk. Exclusive FF describes couplets where the infant's diet is entirely infant formula. BF, breastfeeding; FF, formula feeding; SMM, severe maternal morbidity.

Table 1.

Study Population Demographics and Presence of All-Cause Severe Maternal Morbidity (SMM) or Blood Transfusion-Only SMM

Group	n = 1,857	%
Race/ethnicity
White	1,365	73.5
African American/Black	154	8.3
Hispanic	124	6.7
Asian	163	8.8
Multiracial	35	1.9
Not reported	16	0.9
Age (years)
≤18	62	3.3
19–24	365	19.7
25–34	1,217	65.5
35–39	177	9.5
≥40	36	1.9
Insurance
Private	1,448	78.0
Public	409	22.0
Morbidity
No morbidity	1,755	94.5
All-cause SMM^a	30	1.6
Transfusion-only SMM^b	72	3.9

SMM as defined by the ACOG/SMFM OB consensus statement.¹⁶

Transfusion of any blood product during the delivery hospitalization without other SMM indicators.

SMM, severe maternal morbidity.

Figure 3 demonstrates the adjusted risk ratios for supplementing or exclusively formula feeding for those who experienced either all-cause SMM or transfusion-only SMM produced by multinomial logistic regression modeling accounting for maternal age, race, and insurance status. All-cause SMM was significantly associated with an increased risk of supplementation (aRR = 2.64, 95% confidence interval [CI] = 1.04–6.72) and exclusive formula feeding (adjusted risk ratio [aRR] = 3.61, 95% CI = 1.10–11.78) at hospital discharge. All-cause SMM was also significantly associated with exclusive formula feeding (aRR = 3.51, 95% CI = 1.22–10.12) at the 2- to 3-month follow-up visit. Positive associations between all-cause SMM and formula feeding at the 2- to 4-week time point and supplementation at the 2- to 4-week and 2- to 3-month time points were observed but did not achieve statistical significance.

FIG. 3.

Adjusted risk ratios for supplementation feeding (breast milk and supplement) and exclusive formula feeding as compared with exclusive breastfeeding at three time points for the maternal experience of (a) all-cause SMM or (b) transfusion-only SMM, as compared with couplets without experience of SMM.

Transfusion-only SMM was significantly associated with an increased risk of exclusive formula feeding at hospital discharge (aRR = 2.19, 95% CI = 1.04–4.62), the 2- to 4-week time point (aRR = 2.22, 95% CI = 1.07–4.61) and the 2- to 3-month time point (aRR = 2.63, 95% CI = 1.39–4.97). Transfusion-only SMM was not associated with supplementation at any of the time periods studied.

Discussion

The key finding from this study is that experience of all-cause SMM or transfusion-only SMM appears to be an independent risk factor for exclusive formula feeding. Most individuals who require a blood transfusion during their delivery hospitalization have experienced a severe hemorrhage. Across all groups, the rates of exclusive formula feeding increased from discharge to 2–4 weeks to 2–3 months. In all three groups, the rate of exclusive breastfeeding peaked at the 2- to 4-week time point, and then decreased by the 2- to 3-month time point. The rates of supplementation in all three groups decreased from discharge to 2–4 weeks to 2–3 months. Overall, there are similar feeding patterns as time progresses among all three subgroups. However, when comparing all-cause SMM and transfusion-only SMM with uncomplicated deliveries, there is increased likelihood of exclusive formula feeding.

These findings have important implications for both clinical care and public health. Clinicians should be aware of the heightened risk for failure to achieve breastfeeding goals for individuals who experience SMM and consider proactive steps to support these new parents, such as close follow-up with a lactation specialist. Newborn care providers should likewise be aware of these risks and provide appropriate follow-up of infants born to parents with complicated deliveries. Pregnant individuals who are at increased risk for morbidity should be counseled prenatally on these potential downstream impacts. Despite lower rates of exclusive breastfeeding at hospital discharge among those who experienced SMM, there was an increase observed at 2–3 weeks suggesting harms may be transient and recoverable for motivated individuals. Public health surveillance of breastfeeding rates often stratifies by sociodemographic characteristics, but perhaps should also consider medical conditions and peripartum morbidity experienced by the birthing parent.

There are several limitations to our study. We were unable to assess intention to breastfeed, as this was inconsistently documented in the EHR, although there is no reason to suspect feeding intentions would be significantly different between groups. A second limitation was loss to follow-up of a portion of our cohort who did not receive infant care at our institution. Thirty-one percent of our initial study population was missing from the latter time points of feeding assessment. Finally, although we attempted to control for maternal–infant separation through exclusion of newborns admitted to the NICU, it is not known if separation still occurred between parents and their newborns. Rooming-in is the standard of care at our institution, but a nursery is available on parental request, and it is plausible that parents who had experienced a complicated delivery would be more likely to request nursery care for their newborn.

There are also several strengths to our study. This was a large cohort study from a tertiary care center with adequate rates of the exposure of interest. Several known confounding factors were excluded within the study design such as preterm birth, multiple gestations, and NICU admission. The study examined first-time parents only to avoid bias from past breastfeeding experiences. The exposure of all-cause SMM was validated by clinician chart review and infant feeding modality was also hand abstracted to minimize errors arising from coding variance.

This study reveals opportunities for future studies to improve understanding of the short- and long-term impacts of maternal morbidity. Our study was underpowered to detect differences between specific categories of SMM (diagnostic indicators) and some conditions may pose a greater challenge to lactation. Replicating this type of analysis on a larger population with experience of SMM could reveal which morbidities pose the greatest risk to breastfeeding success.

This study was not designed to determine a causal pathway between maternal morbidity and lactation problems; however, there are plausible physiological and psychological challenges to lactation posed by the experience of SMM. Future qualitative studies could explore the postpartum experience of survivors of SMM and shed light on specific barriers and facilitators to lactation. Finally, maternal morbidity exists on a spectrum of severity. Although our study looked specifically at SMM, quantification of the impact of less-severe, but more common, complications of pregnancy and childbirth such as maternal hemorrhage without need for transfusion, hypertensive conditions complicating childbirth, puerperal infections, or medically necessary maternal–infant separation would be insightful.

Complications of pregnancy and childbirth are on the rise in the United States and have both immediate and long-lasting impacts on the well-being of the birthing parent. Our study found that one such impact is a reduced likelihood of breastfeeding among those who experienced all-cause SMM or transfusion-only SMM. Further study is needed to clarify the causal pathways, such as whether this is primarily a physiological or psychological challenge. Regardless, perinatal clinicians should be aware of this association and offer increased support to those birthing parents who experience SMM and who express a desire to breastfeed their infant.

Footnotes

Acknowledgments

We thank the Institute of Clinical and Translational Science at the University of Iowa for their support in developing the IPHRC data set (UL1TR002537). We also thank Nancy Weathers at the University of Iowa for project management support.

Authors' Contributions

S.G., K.K.R., and S.R. designed the study and drafted the initial article. K.K.R. performed statistical analysis. S.G. and E.A. collected data and prepared data for analysis. S.G., K.K.R., E.A., and S.R. provided critical input on the intellectual content of the article and reviewed and revised the article. All authors approved the final article as submitted and agreed to be accountable for all aspects of the study.

Disclosure Statement

No competing financial interests exist.

Funding Information

This publication was supported by the Health Resources and Services Administration (HRSA) of the U.S. Department of Health and Human Services (HHS) as part of an award totaling $10,361,110.00. The contents are those of the author(s) and do not necessarily represent the official views of, nor an endorsement, by HRSA, HHS, or the U.S. Government.

References

ACOG Committee opinion no. 756: Optimizing support for breastfeeding as part of obstetric practice. Obstet Gynecol, 2018; 132:e187–e196.

American Academy of Pediatrics. Committee on Nutrition, Barness LA. Pediatric Nutrition Handbook, 6th ed. Elk Grove Village, IL: American Academy of Pediatrics, 2009, xlix, 1470 p.

Del Ciampo

, Del Ciampo

IRL

. Breastfeeding and the benefits of lactation for women's health. Rev Bras Ginecol Obstet, 2018; 40:354–359.

Krol

, Grossmann

. Psychological effects of breastfeeding on children and mothers. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz, 2018; 61:977–985.

Brahm

, Valdes

. [The benefits of breastfeeding and associated risks of replacement with baby formulas]. Rev Chil Pediatr, 2017; 88:7–14.

McGuire

. U.S. Dept. of Health and Human Services. The surgeon general's call to action to support breastfeeding. U.S. Dept. of Health and Human Services, Office of the Surgeon General. 2011. Adv Nutr, 2011; 2:523–524.

Creanga

. Maternal mortality in the United States: A review of contemporary data and their limitations. Clin Obstet Gynecol, 2018; 61:296–306.

Petersen

, Davis

, Goodman

, et al. Vital signs: Pregnancy-related deaths, United States, 2011–2015, and strategies for prevention, 13 states, 2013–2017. MMWR Morb Mortal Wkly Rep, 2019; 68:423–429.

Gibson

, Rohan

, Gillespie

. Severe maternal morbidity during delivery hospitalizations. WMJ, 2017; 116:215–220.

10.

Division of Reproductive Health. Severe Maternal Morbidity in the United States 2021. Updated February 2, 2021. Available at www.cdc.gov/reproductivehealth/maternalinfanthealth/severematernalmorbidity.html (accessed December 1, 2021).

11.

Kozhimannil

, Jou

, Attanasio

, et al. Medically complex pregnancies and early breastfeeding behaviors: A retrospective analysis. PLoS One, 2014; 9:e104820.

12.

Brown

, Jordan

. Impact of birth complications on breastfeeding duration: An internet survey. J Adv Nurs, 2013; 69:828–839.

13.

Chessman

, Patterson

, Nippita

, et al. Haemoglobin concentration following postpartum haemorrhage and the association between blood transfusion and breastfeeding: A retrospective cohort study. BMC Res Notes, 2018; 11:686.

14.

Drayton

, Patterson

, Nippita

, et al. Red blood cell transfusion after postpartum haemorrhage and breastmilk feeding at discharge: A population-based study. Aust N Z J Obstet Gynaecol, 2016; 56:591–598.

15.

Furuta

, Sandall

, Cooper

, et al. Severe maternal morbidity and breastfeeding outcomes in the early post-natal period: A prospective cohort study from one English maternity unit. Matern Child Nutr, 2016; 12:808–825.

16.

The American College of Obstetricians and

Gynecologists

, the Society for Maternal-Fetal

Medicine

, Kilpatrick

, Ecker

. Severe maternal morbidity: Screening and review. Am J Obstet Gynecol, 2016; 215:B17–B22.