PLoS Med plos plosmed PLOS Medicine 1549-1277 1549-1676 Public Library of Science San Francisco, CA USA 10.1371/journal.pmed.1004040 PMEDICINE-D-21-05026 Research Article Biology and life sciencesPopulation biologyPopulation metricsDeath rates People and placesPopulation groupingsAge groupsChildrenInfants People and placesPopulation groupingsFamiliesChildrenInfants Biology and life sciencesDevelopmental biologyNeonates Medicine and health sciencesWomen's healthMaternal healthMaternal mortality People and placesPopulation groupingsEthnicitiesHispanic people Medicine and health sciencesWomen's healthMaternal healthBirth Medicine and health sciencesWomen's healthObstetrics and gynecologyBirth Medicine and health sciencesEpidemiologyEthnic epidemiology People and placesGeographical locationsNorth AmericaUnited States Associations between low Apgar scores and mortality by race in the United States: A cohort study of 6,809,653 infants Racial differences in the association of low apgar scores and mortality in the United States https://orcid.org/0000-0003-2719-5138 Gillette Emma Conceptualization Data curation Formal analysis Methodology Project administration Resources Software Writing – original draft Writing – review & editing 1 2 * https://orcid.org/0000-0003-3904-8960 Boardman James P. Investigation Methodology Supervision Validation Writing – review & editing 3 4 https://orcid.org/0000-0003-3272-1040 Calvert Clara Data curation Formal analysis Investigation Validation Writing – review & editing 2 5 https://orcid.org/0000-0001-9651-4193 John Jeeva Validation Writing – review & editing 4 https://orcid.org/0000-0003-4308-856X Stock Sarah J. Conceptualization Data curation Investigation Methodology Project administration Supervision Writing – review & editing 2 3 Arnhold Institute for Global Health at Icahn School of Medicine at Mount Sinai, New York, New York, United States of America University of Edinburgh Usher Institute, NINE Edinburgh BioQuarter, Edinburgh, United Kingdom MRC Centre for Reproductive Health, Queen’s Medical Research Institute, Edinburgh, United Kingdom Centre for Clinical Brain Sciences, Chancellor’s Building, Edinburgh, United Kingdom Department of Population Health, London School of Hygiene and Tropical Medicine, London, United Kingdom Tomlinson Mark Academic Editor Stellenbosch University, SOUTH AFRICA

I have read the journal’s policy and the authors of this manuscript have the following competing interests:SJS has received grant funding, paid to her institution, from the Wellcome Trust, The National Institute of Healthcare Research, The Chief Scientist Office Scotland and Tommy’s Charity. SJS is an academic Editor for PLOS Medicine.

 * E-mail: Emma.f.gillette@gmail.com 12 7 2022 7 2022 19 7 e1004040 7 12 2021 1 6 2022 2022 Gillette et al This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Background

Apgar scores measure newborn health and are strongly associated with infant outcomes, but their performance has largely been determined in primarily white populations. Given the majority of the global population is not white, we aim to assess whether the association between low Apgar score and mortality in infants varies across racial groups.

 Methods and findings

Population-based cohort study using 2016 to 2017 United States National Vital Statistics System data. The study included singleton infants born between 37+0 and 44+6 weeks to mothers over 15 years, without congenital abnormalities. We looked at 3 different mortality outcomes: (1) early neonatal mortality; (2) overall neonatal mortality; and (3) infant mortality. We used logistic regression to assess the association between Apgar score (categorized as low, intermediate, and normal) and each mortality outcome, and adjusted for gestational age, sex, maternal BMI, education, age, previous number of live births, and smoking status, and stratified these models by maternal race group (as self-reported on birth certificates). The cohort consisted of 6,809,653 infants (52.8% non-Hispanic white, 23.7% Hispanic, 13.8% non-Hispanic black, 6.6% non-Hispanic Asian, and 3.1% non-Hispanic other). A total of 6,728,829 (98.8%) infants had normal scores, 63,467 (0.9%) had intermediate scores, and 17,357 (0.3%) had low Apgar scores. Compared to infants with normal scores, low-scoring infants had increased odds of infant mortality. There was strong evidence that this association varied by race (p < 0.001) with adjusted odds ratios (AORs) of 54.4 (95% confidence interval [CI] 49.9 to 59.4) in non-Hispanic white, 70.02 (95% CI 60.8 to 80.7) in Hispanic, 23.3 (95% CI 20.3 to 26.8) in non-Hispanic black, 100.4 (95% CI 74.5 to 135.4) in non-Hispanic Asian, and 26.8 (95% CI 19.8 to 36.3) in non-Hispanic other infants. The main limitation was missing data for some variables, due to using routinely collected data.

 Conclusions

The association between Apgar scores and mortality varies across racial groups. Low Apgar scores are associated with mortality across racial groups captured by United States (US) records, but are worse at discriminating infants at risk of mortality for black and non-Hispanic non-Asian infants than for white infants. Apgar scores are useful clinical indicators and epidemiological tools; caution is required regarding racial differences in their applicability.

Emma Gillette and co-authors assess the associations between low Apgar scores and mortality in infants by race in the United States.

 Author summary Why was this study done?

Apgar scores are commonly used indicators of infants’ well-being at birth and as predictors of mortality and long-term disability.

Apgar scores have been validated in predominantly white populations.

The impact of race on the relationship between Apgar score and early neonatal (death within 0 to 6 days of birth), overall neonatal (death within 0 to 27 days), and infant mortality (death within 1 year) is unknown.

 What did the researchers do and find?

We conducted a cohort study of all singleton, term-birth infants born to mothers over 15 years old, without congenital abnormalities, in the United States (US) in 2017 and 2018.

Our analyses illustrated that race was associated with the assignment of the Apgar score category and with all categories (early neonatal, overall neonatal, and infant mortality) of mortality.

There was a stronger association between low Apgar score and all categories of mortality among non-Hispanic white, non-Hispanic Asian and Hispanic infants than in non-Hispanic black and non-Hispanic other infants.

 What do these findings mean?

These findings suggest that Apgar scores are less useful for estimating the odds of mortality for non-Hispanic black and non-Hispanic non-Asian infants than for non-Hispanic white, non-Hispanic other infants.

Apgar scores are useful predictors of morbidity and mortality; however, their association with mortality is influenced by infant race. Further work to understand which components of the score explain differential associations is needed for developing a scoring system that performs equally well across racial groups.

 http://dx.doi.org/10.13039/100004440 Wellcome Trust 209560/Z/17/Z https://orcid.org/0000-0003-4308-856X Stock Sarah J. SJS is funded by a Wellcome Trust Clinical Career Development Fellowship 209560/Z/17/Z (https://wellcome.org). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Data Availability Data for this manuscript was sourced from the CDC National Center for Health Statistics Vital Statistics Online Data Portal (https://www.cdc.gov/nchs/data_access/vitalstatsonline.htm). All data and associated data dictionaries are accessible online. Introduction

The Apgar score has been used for nearly 70 years to measure infant health and physical well-being immediately after birth and as a predictor of mortality and indicator of an infant’s response to resuscitative efforts [1,2]. Apgar scores are widely used in epidemiological studies for providing population-level information about infants’ status at birth, predicting neurodevelopmental outcomes and infant mortality and as surrogate markers of morbidity [3–5]. In these contexts, Apgar scores are applied across populations, but the tool was developed and validated in predominantly white populations.

The score is composed of 5 variables, each with a value of 0, 1, or 2 [6]. The variables are: heart rate, respiratory effort, muscle tone, reflex response, and skin coloring, each assessed at 1, 5 and 10 minutes after birth [6]. Of these, the 5-minute score is regarded as the best predictor of infant mortality and is commonly used in epidemiological studies and trials [6,7]. The overall Apgar score ranges from 0 to 10 and is frequently categorized as low (0 to 3), intermediate (4 to 6), or normal (7 to 10) [6]. These categorizations of the score were originally suggested by Dr. Virginia Apgar and allow comparison between similar studies [4,7,8].

Few studies validating the use of Apgar scores to assess infants’ status at birth or to look at the association between Apgar scores and adverse infant outcomes have considered race and ethnicity and none have specifically considered white, black, Asian, Hispanic, and other racial groups individually. Two studies have reported that black infants were assigned lower Apgar scores than white infants [9,10]. Some studies have speculated that this may be due to differential interpretation of the skin color variable in infants that are not of white ethnicity [8–10]. Additionally, a study reported that 1-minute Apgar scores were the strongest predictors of infant mortality for Mexican American infants, the worst for black infants, with an intermediate ability for white infants [8].

The lack of understanding around the application of Apgar scores across different race groups is surprising given the wide racial disparities that exist in birth outcomes across many settings including the United States (US) [8,11–15]. Black infants have an infant mortality rate of more than twice that of white infants in the US and have consistently been found to have higher incidences of adverse birth outcomes such as preterm birth, low birth weight, and being small for gestational age [11,16]. Drivers of poor pregnancy outcomes in some race groups are complex and are likely to reflect the interplay of multiple impacts of structural racism including socioeconomic inequalities, access to quality healthcare, and discrimination [17–19].

In light of the widespread use of Apgar scores in clinical and epidemiological settings, the lack of research on racial differences in the applicability of the scores and the prevalence of racial disparities in birth outcomes in the US and elsewhere, this study aims to report on the associations between maternal race, 5-minute Apgar scores, and infant mortality. The objectives of this study are to evaluate the association between maternal race and 5-minute Apgar score, the association between maternal race and mortality, and whether there is a differential association between 5-minute Apgar scores and mortality by race. We hypothesize that the associations between 5-minute Apgar scores and early neonatal, overall neonatal, and infant mortality differ by race.

 Methods Study design and setting

This population cohort study evaluated all infants born between January 1, 2016 and December 31, 2017 in the US (n = 7,820,866). This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (S1 Table).

 Study participants

Inclusion criteria were single births of infants between 37+0 and 44+6 weeks to mothers older than 15 years who were residents of the US Births were excluded if they had no recorded gestational age, 5-minute Apgar score, or maternal race. Missing values for maternal age were imputed by National Center for Health Statistics (NCHS) as the age of the mother from the previous birth record of the same race and birth order in 0.01% of births, and missing values for plurality were imputed by NCHS as singleton births in 0.004% of births. As congenital abnormalities can affect birth outcomes, infants were excluded if they were born with any major congenital abnormality, which were identified by NCHS to include: anencephaly, meningomyelocele/spina bifida, cyanotic congenital heart disease, congenital diaphragmatic hernia, omphalocele, gastroschisis, limb reduction defect, cleft palate, Down syndrome, suspected chromosomal disorder, and hypospadias. This analysis was restricted to term births because elements of the Apgar score, including tone, color, and reflex irritability are dependent on physiological maturity, and recommendations on its use in preterm populations vary [20].

 Data sources

All data were nonidentifiable and publicly accessible through the NCHS Division of Vital Statistics cohort-linked birth and death database, and complied with the NCHS, Centers for Disease Control and Prevention (CDC) Data User Agreement Terms and Conditions [21–23]. The database is composed of data collected directly from US Standard Birth and Death Certificates, including demographic information, and is commonly used in CDC Reports and national studies on infant mortality and neonatal outcomes [24,25]. NCHS linked birth certificate and death certificate data using linking identification numbers, resulting in 22,197 (99.6%) linked death records.

 Variables

The outcome of interest was mortality across the first year of life, subdivided into early neonatal mortality (death within 0 to 6 days of birth), overall neonatal mortality (death within 0 to 27 days of birth), and infant mortality (death within 1 year of birth).

The explanatory variables were 5-minute Apgar score and maternal race/ethnicity, as a surrogate for infant race due to a high frequency of missing data for paternal race. Apgar score was measured by a birth attendant 5 minutes after birth, recorded on the infant’s medical chart, and transcribed to the birth certificate by hospital staff using an NCHS facility worksheet [23]. The same worksheet was completed for births outside the hospital [23]. Scores were recorded as whole numbers in the birth certificates and were categorized for this analysis as low (0 to 3), intermediate (4 to 6) and normal (7 to 10). Maternal race was self-reported by the mother by choosing 1 or more of 15 race categories and 5 Hispanic origin categories from the NCHS facility worksheet [23]. The combined race/ethnicity variable used in this analysis was categorized based on the NCHS designations in order to analyze racial groups that can be compared across studies (S2 Table). This analysis considered Hispanic as a separate race/ethnicity category in accordance with NCHS guidelines for this dataset, which utilize single-race categorizations.

All multivariable regression models adjusted for the confounding effects of the following covariates: gestational age (continuous scale of whole numbered weeks), fetal sex (male or female), maternal educational attainment (≤eighth grade, ninth to 12th grade without diploma, High School/ General Educational Development (GED), Associate’s degree, Bachelor’s degree, Master’s degree, Doctorate or Professional degree, unknown), maternal body mass index (BMI) (underweight [<18.5 kg/m2], normal [18.5 to 24.9 kg/m2], overweight [25 to 29.9 kg/m2], obesity I [30 to 34.9 kg/m2], obesity II [35 to 39.9 kg/m2], extreme obesity III [≥40 kg/m2], unknown), maternal smoking status (smoker or nonsmoker), maternal age (15 to 19, 20 to 24, 25 to 29, 30 to 34, 35 to 39, and 40+), and number of previous live births (0, 1 to 2, 3 to 4, and 5+). Missing values for maternal education, birth weight, smoking status, previous number of live births, and maternal BMI were included as “unknown” categories.

 Statistical methods

The data were analyzed using R version 4.0.2. No overall prospective analysis plan was used. Demographic characteristics were derived for the cohort and each Apgar score group (low, intermediate, normal), with frequencies and percentages reported for categorical variables and means and standard deviations reported for continuous variables.

Univariate logistic regression models were used to quantify the association between each racial group and odds of being assigned a low Apgar score (low versus not low), being assigned an intermediate Apgar score (intermediate versus not intermediate) and to quantify the association between race group and each mortality outcome (early neonatal mortality, neonatal mortality, and infant mortality). Univariate logistic regression models were also used to assess the crude association between each covariate and the mortality outcomes, stratified by race group.

Multivariable logistic regression models were conducted to determine the association between Apgar score and each mortality outcome in the total population and stratified by race group. We formally assessed whether there was evidence that the association between Apgar score and mortality varied by race group by including an interaction term in the adjusted model in the total population. Additionally, we conducted a chi-squared test to determine whether there were trends between Apgar score category and early neonatal, overall neonatal, and infant mortality rates among each race group.

 Ethics committee approval

The study was sponsored by the University of Edinburgh (reference AC20095). Prior to commencement, the research was subject to the Usher Institute (University of Edinburgh) ethics and data protection oversight process. The ethics and data protection triage and overview self-audit of ethics and data protection issues (completed by EG) confirmed that the proposed research, being secondary analysis of a fully anonymized publicly accessible dataset, posed no foreseeable ethics or data protection risks. This indicated there was no requirement for proceeding to a full formal ethics and data protection review by the Usher Research Ethics Group.

 Results Descriptive cohort characteristics

The NCHS cohort-linked database recorded 7,820,866 live births between January 1, 2016 and December 31, 2017. As shown in Fig 1, 6,809,653 births were eligible for inclusion in this study. Data were missing for Apgar score (0.4%), maternal race (0.9%), and gestational age (0.8%), and these cases were excluded from analysis.

A total of 17,357 (0.3%) newborns had low Apgar scores, 63,467 (0.9%) had intermediate scores, and 6,728,829 (98.8%) had normal scores (Table 1). There were 2,115 (0.03%) early neonatal deaths, 3,811 (0.06%) overall neonatal deaths, and 12,436 (0.2%) infant deaths.

10.1371/journal.pmed.1004040.g001 Flowchart of exclusions from study population to analysis cohort.

1Exclusion criteria overlapped, these values represent the frequency of cases in the study population, not necessarily the frequency of excluded cases. 2Important variables included maternal race, maternal education, number of prenatal care visits, smoking status, and 5-minute Apgar score.

 10.1371/journal.pmed.1004040.t001 Descriptive characteristics of study cohort, stratified by 5-minute Apgar score.
Total Cohort Low (0–3) Intermediate (4–6) Normal (7–10) Imputed/Missing
n = 6,809,653 n = 17,357 n = 63,467 n = 6,728,829
Year of birth
2016 3,443,416 (50·6%) 8,825 (50·8%) 32,038 (50·5%) 3,402,553 (50·6%) -
2017 3,366,237 (49·4%) 8,532 (49·2%) 31,429 (49·5%) 3,326,276 (49·4%)
Sex
Male 3,466,945 (50·9%) 9,959 (57·4%) 35,666 (56·2%) 3,421,320 (50·8%) 331
Female 3,342,708 (49·1%) 7,398 (42·6%) 27,801 (43·8%) 3,307,509 (49·2%)
Maternal race
Non-Hispanic white 3,592,235 (52·8%) 8,863 (51·1%) 36,144 (56·9%) 3,547,228 (52·7%) -
Hispanic 1,614,579 (23·7%) 3,080 (17·7%) 10,124 (16·0%) 1,601,375 (23·8%)
Non-Hispanic black 938,878 (13·8%) 3,931 (22·6%) 11,816 (18·6%) 923,131 (13·7%)
Non-Hispanic Asian 451,546 (6·6%) 785 (4·5%) 2,919 (4·6%) 447,842 (6·7%)
Non-Hispanic other 212,415 (3·1%) 698 (4·2%) 2,464 (3·9%) 209,253 (3·1%)
Maternal age
15–19 355,849 (5·2%) 1,370 (7·9%) 4,633 (7·3%) 349,846 (5·2%) 1521
20–24 1,387,621 (20·4%) 4,119 (23·7%) 14,765 (23·3%) 1,368,737 (20·3%)
25–29 2,007,175 (29·5%) 4,885 (28·1%) 18,235 (28·7%) 1,984,055 (29·5%)
30–34 1,924,013 (28·3%) 4,226 (24·3%) 15,911 (25·1%) 1,903,876 (28·3%)
35–39 936,203 (13·7%) 2,225 (12·8%) 7,978 (12·6%) 926,000 (13·8%)
>40 198,792 (2·9%) 532 (3·1%) 1,945 (3·1%) 196,315 (2·9%)
Maternal education
<Eighth grade 223,993 (3·3%) 576 (3·3%) 1,851 (2·9%) 221,566 (3·3%) -
Ninth–12th grade, no diploma 679,548 (10·0%) 1,890 (10·9%) 6,585 (10·4%) 671,073 (10·0%)
High school/GED 1,704,066 (25·0%) 4,818 (27·8%) 16,764 (26·4%) 1,682,484 (25·0%)
Some college credit 1,389,787 (20·4%) 3,863 (22·3%) 14,099 (22·2%) 1,371,825 (20·4%)
Associate’s 559,879 (9·2%) 1,380 (8·0%) 5,494 (8·7%) 553,005 (8·2%)
Bachelor’s 1,386,594 (20·4%) 3,038 (17·5%) 11,742 (18·5%) 1,371,814 (20·4%)
Master’s 630,970 (9·3%) 1,241 (7·1%) 4,972 (7·8%) 624,757 (9·3%)
Doctorate/professional 180,718 (2·7%) 358 (2·1%) 1,468 (2·3%) 178,892 (2·7%)
Unknown 54,098 (0·8%) 193 (1·1%) 4,920 (0·8%) 53,413 (0·8%)
Infant birth weight (g)
<1500 1,184 (0·02%) 47 (0·3%) 82 (0·1%) 1,055 (0·02%) -
1,500–1,999 9,023 (0·1%) 171 (1·0%) 353 (0·6%) 8,499 (0·1%)
2,000–2,499 156,531 (2·3%) 857 (4·9%) 2,545 (4·0%) 153,129 (2·3%)
2,500–2,999 1,160,490 (17·0%) 3,462 (19·9%) 11,701 (18·4%) 1,145,327 (17·0%)
3,000–3,499 2,866,785 (41·1%) 6,343 (36·5%) 24,047 (37·9%) 2,836,395 (42·2%)
3,500–3,999 2,014,972 (29·6%) 4,580 (26·4%) 17,663 (27·8%) 1,992,729 (29·6%)
4,000–4,499 517,583 (7·6%) 1,409 (8·1%) 5,620 (8·9%) 510,554 (7·6%)
4,500–4,999 72,522 (1·1%) 370 (2·1%) 1,159 (1·8%) 70,993 (1·1%)
≥5,000 8,302 (0·1%) 77 (0·4%) 237 (0·4%) 7,988 (0·1%)
Unknown 2,261 (0·03%) 41 (0·2%) 60 (0·1%) 2,160 (0·03%)
Smoking status
No smoking 6,320,204 (92·8%) 15,670 (90·3%) 57,312 (90·3%) 6,247,222 (92·8%) -
Smoking 460,417 (6·8%) 1,566 (9·0%) 5,758 (9·1%) 453,093 (6·7%)
Unknown 29032 (0·4%) 121 (0·7%) 397 (0·6%) 28,514 (0·4%)
Maternal BMI
Underweight (<18·5) 225,679 (3·3%) 495 (2·9%) 1,786 (2·8%) 223,398 (3·3%) -
Normal (18·5–24·9) 2,942,814 (43·2%) 6,486 (37·4%) 24,716 (38·9%) 2,911,612 (43·3%)
Overweight (25–29·9) 1,744,211 (25·6%) 4,309 (24·8%) 15,948 (25·1%) 1,723,954 (25·6%)
Obesity I (30–34·9) 954,107 (14·0%) 2,685 (15·5%) 9,574 (15·1%) 941,848 (14·0%)
Obesity II (35–39·9) 459,520 (6·7%) 1,465 (8·4%) 5,284 (8·3%) 452,771 (6·7%)
Obesity III (≥40) 319,681 (4·7%) 1,342 (7·7%) 4,563 (7·2%) 313,776 (4·7%)
Unknown 163,641 (2·4%) 575 (3·3%) 1,596 (2·5%) 161,470 (2·4%)
Previous live births
0 2,611,902 (38·4%) 9,708 (55·9%) 35,295 (55·6%) 2,566,899 (38·1%)
1–2 3,359,993 (49·3%) 5,751 (33·1%) 21,667 (34·1%) 3,332,575 (49·5%)
3–4 672,440 (9·9%) 1,378 (7·9%) 4,948 (7·8%) 666,114 (9·9%)
5+ 146,832 (2·2%) 421 (2·4%) 1,293 (2·0%) 145,118 (2·2%)
Unknown 18,486 (0·3%) 99 (0·6%) 264 (0·4%) 18,123 (0·3%)
Gestational age
Mean (SD) 39·0 (1·1) 39·0 (1·2) 39·0 (1·2) 39·0 (1·1) -
Mortality category
Early neonatal death (0–6 days) N (Rate per 1,000 births) 2,115 (0·3) 1,029 (59·3) 362 (5·7) 724 (0·1) -
Neonatal death (0–27 days) N (Rate per 1,000 births) 3,811 (0·6) 1,190 (68·6) 546 (8·6) 2,075 (0·3) -
Infant death (0–365 days) N (Rate per 1,000 births) 12,436 (1·8) 1,333 (76·8) 823 (13·0) 10,280 (1·5) -

1Missing values imputed by NCHS.

BMI, body mass index; GED, General Educational Development; SD, standard deviation.

Table 1 describes the cohort characteristics. Overall, 49.1% infants were female, 52.8% were non-Hispanic white, 23.7% were Hispanic, 13.8% were non-Hispanic black, 6.6% were non-Hispanic Asian, 3.1% were non-Hispanic other, and 78.2% were born to mothers aged 20 to 34. The majority of mothers (86.0%) had a high school degree or higher, and the average gestational age was 39 weeks. Data were missing and included as an “unknown” category for maternal education (0.8%), birth weight (0.03%), smoking status (0.4%), previous live births (0.3%), and maternal BMI (2.4%).

 Association between maternal race and 5-minute Apgar score

Race group was associated with the assignment of Apgar score category (p < 0.001). Compared with non-Hispanic white infants, non-Hispanic black infants had 1.7 times the odds of being assigned a low Apgar score (95% confidence interval [CI] 1.6 to 1.8) and non-Hispanic other infants had 1.3 times the odds (95% CI 1.2 to 1.4) (Table 2). Non-Hispanic Asian and Hispanic infants had 23% (95% CI 0.7 to 0.8) and 30% (95% CI 0.6 to 0.8) lower odds of being assigned low scores than non-Hispanic white infants, respectively.

10.1371/journal.pmed.1004040.t002 Unadjusted odds ratio of being assigned a low (0–3) or intermediate (4–6) Apgar score by maternal race.
Race group Assigned low score Odds ratio of low Apgar score(95% CI) P value* Assigned intermediate score Odds ratio of intermediate score (95% CI) P value*
Non-Hispanic white (n = 3,592,235) 8,863 (0·3%) 1 (ref) - 36,144 (1·01%) 1 (ref) -
Hispanic (n = 1,614,579) 3,080 (0·2%) 0·77 (0·74–0·81) <0.001 10,124 (0·6%) 0·62 (0·61–0·63) <0.001
Non-Hispanic black (n = 938,878) 3,931 (0·4%) 1·70 (1·64–1·76) <0.001 11,816 (1·3%) 1·25 (1·23–1·28) <0.001
Non-Hispanic Asian (n = 451,546) 785 (0·2%) 0·70 (0·65–0·76) <0.001 2,919 (0·7%) 0·64 (0·62–0·66) <0.001
Non-Hispanic other (n = 212,415) 698 (0·3%) 1·33 (1·23–1·44) <0.001 2,464 (1·2%) 1·15 (1·11–1·20) <0.001

*Wald p-value

CI, confidence interval.

 Association between maternal race and mortality

There was strong evidence of an association between maternal race and mortality across the first year of life. Compared with non-Hispanic white infants, non-Hispanic black infants had higher odds of all categories of mortality (early neonatal mortality: odds ratio [OR] 1.3 [95% CI 1.2 to 1.5]; overall neonatal mortality: OR 1.5 [95% CI = 1.3 to 1.6]; infant mortality OR 1.9 [95% CI = 1.8 to 2.0]; S4 Table). Conversely, non-Hispanic Asian and Hispanic infants had lower odds for all mortality outcomes when compared with white infants. There was no evidence of a difference in early neonatal mortality between non-Hispanic other and white infants, but non-Hispanic other infants had higher odds of neonatal and infant mortality.

 Impact of race on the relationship between 5-minute Apgar score and mortality

The early neonatal, overall neonatal, and infant mortality rates decreased with increasing Apgar scores in all races (S3 Table). Across all races, low Apgar score was a strong risk factor for mortality across the first year of life. Low Apgar score was a stronger risk factor than an intermediate score, and there was a strong association between score category and all categories of mortality (p < 0.001; S3 Table).

There was strong evidence that the adjusted association between low Apgar score and mortality varied by race group (p < 0.001) for all mortality outcomes. The adjusted odds ratio (AOR), comparing the odds of infant mortality among infants with low Apgar scores to those with normal Apgar scores, were higher in non-Hispanic Asian (AOR 100.4 [95% CI 74.5 to 135.4]), non-Hispanic white (AOR 54.4 [95% CI 49.9 to 59.4]), and Hispanic (AOR 70.0 [95% CI 60.8 to 80.7]) groups than non-Hispanic black (AOR 23.3 [95% CI 20.3 to 26.8]) and non-Hispanic other (AOR 26.8 [95% CI 19.8 to 36.3]) groups (Table 3). Similar associations were present in the early neonatal and overall neonatal mortality categories, with non-Hispanic black and non-Hispanic other groups consistently having the lowest odds ratio for the association between low Apgar score and mortality (Fig 2).

10.1371/journal.pmed.1004040.g002 AORs and 95% CIs for early neonatal, overall neonatal, and infant mortality in relation to Apgar score, stratified by maternal race group (<italic>n =</italic> 6,809,653).

AORs of (1) early neonatal mortality; (2) overall neonatal mortality; and (3) infant mortality for infants with low (0 to 3) and intermediate (4 to 6) 5-minute Apgar scores referent to infants with normal (7 to 10) 5-minute Apgar scores. ORs were adjusted for infant sex, maternal age, maternal smoking status, infant birth weight, maternal education, maternal BMI, previous number of live births and gestational age. AOR, adjusted odds ratio; CI, confidence interval; Hisp, Hispanic; NHA, non-Hispanic Asian; NHB, non-Hispanic black, NHO, non-Hispanic other; NHW, non-Hispanic white.

 10.1371/journal.pmed.1004040.t003 AORs for early neonatal mortality, neonatal mortality, and infant mortality in relation to Apgar score category, stratified by maternal race group.
Early neonatal mortality (<7 days) Overall neonatal mortality (<28 days) Infant mortality (≤1 year)
Deaths (rate per 1,000 births) AOR (95% CI) p-value* Deaths (rate per 1,000 births) AOR (95% CI) p-value* Deaths (rate per 1,000 births) AOR (95% CI) p-value*
Overall
Normal (n = 6,728,829) 724 (0·1) 1 (ref) 2,075 (0·3) 1 (ref) 10,280 (1·5) 1 (ref)
Intermediate (n = 63,467) 362 (5·7) 45·9 (40·3–52·2) <0.001 546 (8·6) 23·8 (21·6–26·2) <0.001 823 (13·0) 7·6 (7·02–8·1) <0.001
Low (n = 17,357) 1,029 (59·3) 493·1 (445·9–545·4) <0.001 1,190 (68·6) 199·4 (184·6–215·3) <0.001 1,333 (76·8) 47·5 (44·6–50·5) <0.001
Non-Hispanic white
Normal (n = 3,547,228) 349 (0·1) 1 (ref) 1,047 (0·3) 1 (ref) 4,952 (1·4) 1 (ref)
Intermediate (n = 36,144) 183 (5·1) 44·4 (37·0–53·4) <0.001 278 (7·7) 22·04 (19·2–25·3) <0.001 396 (11·0) 6·9 (6·2–7·6) <0.001
Low (n = 8,863) 564 (63·6) 598·9 (519·9–690·0) <0.001 642 (72·4) 225·0 (202·4–250·2) <0.001 703 (79·3) 54·4 (49·9–59·4) <0.001
Hispanic
Normal (n = 1,601,375) 172 (0·1) 1 (ref) 431 (0·3) 1 (ref) 1,886 (1·2) 1 (ref)
Intermediate (n = 10,124) 74 (7·3) 59·0 (44·5–78·2) <0.001 107 (10·6) 33·2 (26·7–41·4) <0.001 161 (15·9) 12·4 (10·5–14·7) <0.001
Low (n = 3,080) 205 (66·6) 537·1 (431·8–668·2) <0.001 234 (76·0) 245·1 (205·8–291·9) <0.001 270 (87·7) 70·02 (60·8–80·7) <0.001
Non-Hispanic black
Normal (n = 923,131) 127 (0·1) 1 (ref) 411 (0·4) 1 (ref) 2,518 (2·7) 1 (ref)
Intermediate (n = 11,816) 80 (6·8) 41·4 (31·07–55·3) <0.001 123 (10·4) 20·2 (16·4–24·9) <0.001 199 (16·8) 5·8 (5·03–6·8) <0.001
Low (n = 3,931) 172 (43·8) 286·7 (225·6–364·3) <0.001 212 (53·9) 114·2 (95·7–136·1) <0.001 245 (62·3) 23·3 (20·3–26·8) <0.001
Non-Hispanic Asian
Normal (n = 447,842) 51 (0·1) 1 (ref) 98 (0·2) 1 (ref) 352 (0·8) 1 (ref)
Intermediate (n = 2,919) 16 (5·5) 45·8 (25·6–82·02) <0.001 22 (7·5) 32·5 (20·2–52·5) <0.001 38 (13·0) 16·3 (11·5–23·0) <0.001
Low (n = 785) 47 (59·9) 545·8 (354·6–840·07) <0.001 56 (71·3) 335·4 (234·1–480·4) <0.001 60 (76·4) 100·4 (74·5–135·4) <0.001
Non-Hispanic Other
Normal (n = 209,253) 25 (0·1) 1 (ref) 88 (0·4) 1 (ref) 572 (2·7) 1 (ref)
Intermediate (n = 2,464) 9 (3·7) 27·0 (12·4–58·5) <0.001 16 (6·5) 14·0 (8·1–24·1) <0.001 29 (11·8) 4·1 (2·8–6·0) <0.001
Low (n = 698) 41 (58·7) 414·1 (242·6–707·0) <0.001 46 (65·9) 133·02 (89·9–196·9) <0.001 55 (78·8) 26·8 (19·8–36·3) <0.001

*Wald p-value

AOR, adjusted odds ratios; BMI, body mass index; CI, associated 95% confidence intervals.

Odds ratios and 95% CIs were adjusted infant sex, maternal age, maternal smoking status, infant birth weight, maternal education, maternal BMI, previous number of live births, and gestational age.

Similar trends of lesser magnitudes persisted in the associations between intermediate scores and all categories of mortality (Table 3). The AORs for the associations between intermediate Apgar score and neonatal mortality varied across race groups (p < 0.001) with AORs of 22.0 (95% CI 19.2 to 25.3) for non-Hispanic white infants, 20.2 (95% CI 16.4 to 24.9) for non-Hispanic black infants, 32.5 (95% CI 20.2 to 52.5) for non-Hispanic Asian infants, 14.0 (95% CI 8.1 to 24.1) for non-Hispanic other infants, and 33.2 (95% CI 26.7 to 41.4) for Hispanic infants (p-value < 0.001). The differences between the race groups in these associations increased across the first year of life (Table 3). The final adjusted models, stratified by race group, are presented in S5–S9 Tables.

 Discussion

Overall, we find that low and intermediate Apgar scores are strongly associated with mortality across the first year of life in the US. These findings align with Dr. Apgar’s original use of the score to predict neonatal mortality and support the use of the 5-minute Apgar score in research [4,7,26]. These data also illustrate for the first time, to our knowledge, that these strong associations persist across racial groups captured by US birth records. We do, however, also find evidence that there is variation in the relationship between Apgar score and mortality across different racial groups. Non-Hispanic black and non-Hispanic other infants have higher odds of mortality across the first year of life, and are more likely to be assigned a low Apgar score, when compared with white infants; yet, multivariable regression models revealed that the association between low Apgar score and mortality is weakest in non-Hispanic black and non-Hispanic other groups. These findings indicate that low Apgar scores are differentially associated with mortality across race groups and, more specifically, suggest that low Apgar scores are less good at discriminating the risk of mortality in non-Hispanic black and non-Hispanic other infants compared to other race groups.

The findings are consistent with literature suggesting that Apgar scores are strongly associated with mortality across the first year of life [6,13,27,28]. The results also align with research demonstrating that Apgar scores were more predictive of mortality in white and Mexican-American infants than in black infants [8]. This study expanded on the findings of previous research by evaluating more racial groups in a larger and more representative study population. These results add to a body of literature that suggests that the performance of mortality prediction tools in neonatal groups can be influenced by race, as demonstrated by the inclusion of race in the use of an estimator tool for bronchopulmonary dysplasia in preterm infants [29]. The mortality rates in our study population were lower than nationally reported estimates for 2018 of 3.75 neonatal deaths per 1,000 births and a rate of 5.64 infant deaths per 1,000 births [24,30]. The difference in rates is likely due to restricting the study population to term infants born without congenital malformations.

We report that the 5-minute Apgar score has a strong association with infant mortality in a large multiracial population in a real-world setting. Therefore, the score can be used for informing prognosis and as a valid metric in epidemiological studies. However, there are differences in the strength of association between Apgar score and mortality between racial groups, which should be taken in to account in clinical practice and research studies.

Reduced strength of association between the 5-minute Apgar score and mortality in black and non-Hispanic non-Asian groups might be explained in part by systematic differences in assignment of score at birth. The potential differential assignment of the score by race therefore requires further investigation; we suggest that particular consideration be given to individual components of the score, particularly the “skin color” component. The “skin color” scheme relies on classifying infants as blue, pale, or pink and is unlikely to be equally efficacious across a range of skin tones, as demonstrated by a recent study stating that a majority of physicians do not agree that “pink all over” is an accurate description of vigorous African-American infants [8,31]. It is possible that refinement of the scoring system to capture circulatory status more reliably could improve its performance in identifying infants at high risk of mortality. However, the reasons for the attenuated association between Apgar score and mortality in black and Hispanic infants are certainly more complex than just differences in how the score is assigned at birth and are likely to be driven by the social drivers behind poor outcomes in certain race groups, which are not necessarily captured by clinical scores such as the Apgar. Further research is needed to understand the complex social and structural pathways that explain the differential associations between Apgar scores and mortality across racial groups in order to inform future use of the score.

This study has a number of strengths. The large sample size derived from a population of all births in the US from 2016 to 2017 allowed for a study cohort that was representative of the US population and minimized selection bias. All of the data were derived from routinely collected NCHS data, minimizing recall and social desirability biases. The present study included 5 maternal racial groups in analysis, allowing for a more thorough analysis of racial differences than previous studies.

The study also has limitations, largely due to the nature of routinely collected data. There were missing values for some covariates, and due to the low frequency of missing values and lack of significant associations between them, these missing values were included as “unknown” categories. The analysis considered maternal race as the infant’s race due to a high proportion of missing data for paternal race, which will have results in some misclassification for infants where paternal race was different than maternal race. This analysis excluded preterm births due to concerns about the reliability of the Apgar score in this population; however, a recent study has demonstrated a strong association between low Apgar score and neonatal mortality in preterm infants [3,20]. Further analysis should be conducted to assess how the association between Apgar score and mortality varies by race group among preterm infants.

A further limitation may have been that models were not adjusted for all maternal comorbidities or medication use, because the data to do so reliably were not available.

 Conclusions

Overall, low 5-minute Apgar scores are strongly associated with early neonatal, overall neonatal, and infant mortality outcomes in a large multiracial population. There are strong associations between race and the 5-minute Apgar score, with black infants having higher odds of being assigned a low score than their white counterparts. Strong associations also exist between race and mortality within the first year of life, with black and non-Hispanic other infants having the highest odds of neonatal and infant mortality. Importantly, there are racial differences in the strength of the association between Apgar score and mortality. This suggests that while Apgar scores should continue to be used in clinical and research settings, practitioners and researchers should be aware that both the assignment and predictive ability of the Apgar score varies across racial groups.

 Supporting information STROBE checklist.

(DOCX)

 Maternal race recategorizations.

(DOCX)

 Mortality rates per 1,000 births, stratified by Apgar score and maternal race.

(DOCX)

 Unadjusted odds ratios for early neonatal, overall neonatal, and infant mortality by race group.

(DOCX)

 Unadjusted and adjusted odds ratios for mortality for multivariable models in non-Hispanic white cohort.

(DOCX)

 Unadjusted and adjusted odds ratios for mortality for multivariable models in Hispanic cohort.

(DOCX)

 Unadjusted and adjusted odds ratios for mortality for multivariable models in non-Hispanic black cohort.

(DOCX)

 Unadjusted and adjusted odds ratios for mortality for multivariable models in non-Hispanic Asian cohort.

(DOCX)

 Unadjusted and adjusted odds ratios for mortality for multivariable models in non-Hispanic other cohort.

(DOCX)

 Abbreviations: AOR

adjusted odds ratios

 BMI

body mass index

 CDC

Centers for Disease Control and Prevention

 CI

confidence interval

 GED

General Educational Development

 NCHS

National Center for Health Statistics

 OR

odds ratio

 References Degrandi Oliveira CR. The legacy of Virginia Apgar. Br J Anaesth. 2020;124(3):e185–e6. doi: 10.1016/j.bja.2019.12.017 31948680 Razaz N, Cnattingius S, Joseph KS. Association between Apgar scores of 7 to 9 and neonatal mortality and morbidity: population based cohort study of term infants in Sweden. BMJ. 2019;365:l1656–l. doi: 10.1136/bmj.l1656 31064770. Cnattingius S, Johansson S, Razaz N. Apgar score and risk of neonatal death among preterm infants. N Engl J Med. 2020;383(1):49–57. doi: 10.1056/NEJMoa1915075 32609981 Iliodromiti S, Mackay DF, Smith GCS, Pell JP, Nelson SM. Apgar score and the risk of cause-specific infant mortality: a population-based cohort study. Lancet. 2014;384(9956):1749–55. doi: 10.1016/S0140-6736(14)61135-1 25236409 Persson M, Razaz N, Tedroff K, Joseph KS, Cnattingius S. Five and 10 minute Apgar scores and risks of cerebral palsy and epilepsy: population based cohort study in Sweden. BMJ. 2018;360:k207. doi: 10.1136/bmj.k207 29437691 Cnattingius S, Norman M, Granath F, Petersson G, Stephansson O, Frisell T. Apgar Score Components at 5 Minutes: Risks and Prediction of Neonatal Mortality. Paediatr Perinat Epidemiol. 2017;31(4):328–37. doi: 10.1111/ppe.12360 28493508 Casey BM, McIntire DD, Leveno KJ. The Continuing Value of the Apgar Score for the Assessment of Newborn Infants. N Engl J Med. 2001;344(7):467–71. doi: 10.1056/NEJM200102153440701 11172187 Doyle JM, Echevarria S, Frisbie WP. Race/ethnicity, Apgar and infant mortality. Popul Res Policy Rev. 2003;22(1):41–64. Hegyi T, Carbone T, Anwar M, Ostfeld B, Hiatt M, Koons A, et al. The apgar score and its components in the preterm infant. Pediatrics. 1998;101(1 Pt 1):77–81. Epub 1998/01/17. doi: 10.1542/peds.101.1.77 9417155. Petrikovsky BM, Diana L, Baker DA. Race and Apgar scores. Anaesthesia. 1990;45(11):988–9. Epub 1990/11/01. doi: 10.1111/j.1365-2044.1990.tb14649.x 2252204. Bediako PT, BeLue R, Hillemeier MM. A Comparison of Birth Outcomes Among Black, Hispanic, and Black Hispanic Women. J Racial Ethn Health Disparities. 2015;2(4):573–82. doi: 10.1007/s40615-015-0110-2 26561541 Chambers BD, Erausquin JT, Tanner AE, Nichols TR, Brown-Jeffy S. Testing the Association Between Traditional and Novel Indicators of County-Level Structural Racism and Birth Outcomes among Black and White Women. J Racial Ethn Health Disparities. 2018;5(5):966–77. doi: 10.1007/s40615-017-0444-z 29218496 Collins JW Jr, David RJ, Handler A, Wall S, Andes S. Very low birthweight in African American infants: the role of maternal exposure to interpersonal racial discrimination. Am J Public Health. 2004;94(12):2132–8. doi: 10.2105/ajph.94.12.2132 15569965 Kothari CL, Paul R, Dormitorio B, Ospina F, James A, Lenz D, et al. The interplay of race, socioeconomic status and neighborhood residence upon birth outcomes in a high black infant mortality community. SSM Population Health. 2016;2:859–67. doi: 10.1016/j.ssmph.2016.09.011 29349194 Ro A, Goldberg RE, Kane JB. Racial and Ethnic Patterning of Low Birth Weight, Normal Birth Weight, and Macrosomia. Prev Med. 2019;118:196–204. doi: 10.1016/j.ypmed.2018.10.012 30342108 CDC. Infant Health. cdc.gov: Centers for Disease Control and Prevention. 2018. Matoba N, Collins JW. Racial disparity in infant mortality. Semin Perinatol. 2017;41(6):354–9. doi: 10.1053/j.semperi.2017.07.003 28864275 Vedam S, Stoll K, Taiwo TK, Rubashkin N, Cheyney M, Strauss N, et al. The Giving Voice to Mothers study: inequity and mistreatment during pregnancy and childbirth in the United States. Reprod Health. 2019;16(1):77. doi: 10.1186/s12978-019-0729-2 31182118 Walton E. Residential segregation and birth weight among racial and ethnic minorities in the United States. J Health Soc Behav. 2009;50(4):427–42. doi: 10.1177/002214650905000404 20099449 American Academy of Pediatrics Committee on Fetus and Newborn; American College of Obstetricians and Gynecologists Committee on Obstetric Practice. The Apgar Score. Pediatrics. 2015;136(4):819. doi: 10.1542/peds.2015-2651 26416932 Martin JA, Schwartz S. 2018 Natality Public Use File. In: Statistics CfDCaPNCfH, editor. Period/Cohort Linked Birth-Infant Death Data Files. Hyattsville, Maryland: Division of Vital Statistics; 2019. Statistics NCfH. National Center for Health Statistics Data Use Agreement (DUA) for Vital Statistics Data Files. 2018. NCHS. User Guide to the 2018 Period/2017 Cohort Linked Birth/Infant Death Public Use File. User Guide. cdc.gov: Centers for Disease Control and Prevention. 2019. Ely DM, Driscoll AK. Infant Mortality in the United States, 2018: Data from the Period Linked Birth/Infant Death File. National Center for Health Statistics, 2020 July 16, 2020. Report No.: Contract No: 7. 32730740 Ventura SJ. The U.S. National Vital Statistics System: Transitioning Into the 21st Century, 1990–2017. Vital Health Stat 1. 2018;(62):1–84. Epub 2018/09/25. 30248018. Apgar V, Holaday DA, James LS, Weisbrot IM, Berrien C. Evaluation of the Newborn Infant—Second Report. J Am Med Assoc. 1958;168(15):1985–8. doi: 10.1001/jama.1958.03000150027007 13598635 Chen H-Y, Blackwell SC, Chauhan SP. Association between apgar score at 5 minutes and adverse outcomes among Low-Risk pregnancies. J Matern Fetal Neonatal Med. 2020;1–8. doi: 10.1080/14767058.2020.1754789 32299290 Ehrenstein V. Association of Apgar scores with death and neurologic disability. Clin Epidemiol. 2009;1:45–53. doi: 10.2147/clep.s4782 PubMed 20865086. Laughon MM, Langer JC, Bose CL, Smith PB, Ambalavanan N, Kennedy KA, et al. Prediction of bronchopulmonary dysplasia by postnatal age in extremely premature infants. Am J Respir Crit Care Med. 2011;183(12):1715–22. Epub 2011/03/04. doi: 10.1164/rccm.201101-0055OC 21471086. Hug L, Sharrow D, You D. Levels & Trends in Child Mortality. New York, New York: United Nations Children’s Fund, 2019. Adams BN, Grunebaum A. Does “Pink All Over” Accurately Describe an Apgar Color Score of 2 in Newborns of Color? Obstet Gynecol. 2014;123. 10.1371/journal.pmed.1004040.r001 Decision Letter 0 Odeny Beryne Senior Editor 2022 Beryne Odeny This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Submission Version 0

10 Dec 2021

Dear Dr Stock,

Thank you for submitting your manuscript entitled "Racial Differences in the Association of Low Apgar Scores and Mortality in the United States: a cohort study of 6,809,653 infants" for consideration by PLOS Medicine.

Your manuscript has now been evaluated by the PLOS Medicine editorial staff and I am writing to let you know that we would like to send your submission out for external peer review.

However, before we can send your manuscript to reviewers, we need you to complete your submission by providing the metadata that is required for full assessment. To this end, please login to Editorial Manager where you will find the paper in the 'Submissions Needing Revisions' folder on your homepage. Please click 'Revise Submission' from the Action Links and complete all additional questions in the submission questionnaire.

Please re-submit your manuscript within two working days, i.e. by Dec 14 2021 11:59PM.

Login to Editorial Manager here: https://www.editorialmanager.com/pmedicine

Once your full submission is complete, your paper will undergo a series of checks in preparation for peer review. Once your manuscript has passed all checks it will be sent out for review.

Feel free to email us at plosmedicine@plos.org if you have any queries relating to your submission.

Kind regards,

Beryne Odeny

PLOS Medicine

 10.1371/journal.pmed.1004040.r002 Decision Letter 1 Odeny Beryne Senior Editor 2022 Beryne Odeny This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Submission Version 1

10 Mar 2022

Dear Dr. Stock,

Thank you very much for submitting your manuscript "Racial Differences in the Association of Low Apgar Scores and Mortality in the United States: a cohort study of 6,809,653 infants" (PMEDICINE-D-21-05026R1) for consideration at PLOS Medicine.

Your paper was evaluated by a senior editor and discussed among all the editors here. It was also sent to independent reviewers, including a statistical reviewer. The reviews are appended at the bottom of this email and any accompanying reviewer attachments can be seen via the link below:

Considering these reviews, we would be grateful if you could please revise your manuscript to respond to comments raised by reviewers. We would strongly recommend that you pay special attention to the reviewer #2's comments regarding the premise of your research question and your approach. Please note that further consideration is dependent on the submission of a manuscript that addresses all reviewer concerns.

[LINK]

In light of these reviews, I am afraid that we will not be able to accept the manuscript for publication in the journal in its current form, but we would like to consider a revised version that addresses the reviewers' and editors' comments. Obviously we cannot make any decision about publication until we have seen the revised manuscript and your response, and we plan to seek re-review by one or more of the reviewers.

In revising the manuscript for further consideration, your revisions should address the specific points made by each reviewer and the editors. Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments, the changes you have made in the manuscript, and include either an excerpt of the revised text or the location (eg: page and line number) where each change can be found. Please submit a clean version of the paper as the main article file; a version with changes marked should be uploaded as a marked up manuscript.

In addition, we request that you upload any figures associated with your paper as individual TIF or EPS files with 300dpi resolution at resubmission; please read our figure guidelines for more information on our requirements: http://journals.plos.org/plosmedicine/s/figures. While revising your submission, please upload your figure files to the PACE digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at PLOSMedicine@plos.org.

We expect to receive your revised manuscript by Mar 31 2022 11:59PM. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

We ask every co-author listed on the manuscript to fill in a contributing author statement, making sure to declare all competing interests. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. If new competing interests are declared later in the revision process, this may also hold up the submission. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT. You can see our competing interests policy here: http://journals.plos.org/plosmedicine/s/competing-interests.

Please use the following link to submit the revised manuscript:

https://www.editorialmanager.com/pmedicine/

Your article can be found in the "Submissions Needing Revision" folder.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

We look forward to receiving your revised manuscript.

Sincerely,

Beryne Odeny,

PLOS Medicine

plosmedicine.org

-----------------------------------------------------------

Requests from the editors:

1) Please revise your title according to PLOS Medicine's style. Your title must be nondeclarative and not a question. It should begin with main concept if possible. Please place the study design in the subtitle (i.e., after a colon), e.g., a retrospective cohort study.

2) Please include line numbers in your next draft.

3) Is there a chance you can obtain more recent data from this setting?

4) Please conclude the “Introduction” with a clear description of the study question or hypothesis. The description of the database can be moved to the methods section.

5) Abstract:

a) Please ensure that all numbers presented in the abstract are present and identical to numbers presented in the main manuscript text.

b) Please quantify the main results (please present both 95% CIs and p values).

c) In the last sentence of the Abstract Methods and Findings section, please describe the main limitation(s) of the study's methodology.

6) At this stage, we ask that you include a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract. Please see our author guidelines for more information: https://journals.plos.org/plosmedicine/s/revising-your-manuscript#loc-author-summary

7) Did your study have a prospective protocol or analysis plan? Please state this (either way) early in the Methods section.

a) If a prospective analysis plan (from your funding proposal, IRB or other ethics committee submission, study protocol, or other planning document written before analyzing the data) was used in designing the study, please include the relevant prospectively written document with your revised manuscript as a Supporting Information file to be published alongside your study, and cite it in the Methods section. A legend for this file should be included at the end of your manuscript.

b) If no such document exists, please make sure that the Methods section transparently describes when analyses were planned, and when/why any data-driven changes to analyses took place.

c) In either case, changes in the analysis-- including those made in response to peer review comments-- should be identified as such in the Methods section of the paper, with rationale.

8) Thank you for providing the STROBE checklist as Supporting Information.

a) Please add the following statement, or similar, to the Methods: "This study is reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (S1 Checklist)."

b) When completing the checklist, please use section and paragraph numbers, rather than page numbers.

9) In your statistical analyses, please account for clustering of observations at hospital and county levels. Generalized Estimating Equations (GEE) or hierarchical/ multilevel models, among others, may be useful in this case.

10) Did you adjust for maternal comorbidity or medication use? If not, please consider this or acknowledge the lack of adjustment as a limitation

11) Please provide p values in addition to 95% CIs in the main text and tables

12) Please define the abbreviations in Tables and Figures e.g, GED, BMI, SD

13) Please remove the “Role of the funding source”, “Data sharing”, and “conflict of interest” statement in the methods section. This information is captured in the metadata obtained in the submission form

14) Please remove the “Competing Interests”, “Data availability,” and “Funding” statements at the end of the main text. This information is captured in the metadata obtained in the submission form.

15) References:

a) Please select the PLOS Medicine reference style in your citation manager. In-text reference call outs should be presented as follows noting the absence of spaces within the square brackets, e.g., "... services [1,2]."

b) References should have no more and no less than six names before et al. For references with more than six names, please ensure that et al., is inserted after six names

c) Please ensure that journal name abbreviations consistently match those found in the National Center for Biotechnology Information (NCBI) databases. https://journals.plos.org/plosmedicine/s/submission-guidelines#loc-references.

Comments from the reviewers:

Reviewer #1: Thank you for the opportunity to review this manuscript. This is overall a well written paper.

1. My comments are mainly related to the test of interactions. By using stratified analysis, the author claims that the association between Apgar scores and mortality varies across racial groups. However, whether the difference in the association between Apgar scores and mortality across racial groups is statistically significant or not has not been tested or presented. In order to acheive this, the author may need to used the adjusted model with interaction beteeen Apgar Score category and racial group. With one racial group as a reference group, ratio of AOR needs to be tested and presented.

2. The resoluiton of Figure 2 needs to be increased.

Reviewer #2: This study examines assess whether the association between low Apgar score and mortality in infants varies across racial groups.

This is a well written paper, and the analysis appears to be appropriate (having said this I would recommend that the paper is reviewed by a statistical reviewer).

The sample size is of course impressive - but I am afraid that this cannot remedy the other concerns I have with the paper.

A large part of the introduction is making the argument that while Apgar scores have been widely used and are highly predictive of infant mortality, the scores have largely been developed and validated in White populations. I suspect the authors are speaking about predictive validity - mortality. If so, this is of course an important issue. The authors also talk about validating the scores to assess infant status at birth. This is of course an entirely different matter and would involve a study comparing the gold standard judgement (presumably conducted by a physician) with the scores recorded on a birth chart/card. The authors even describe how Black infants have been shown to be assigned lower Apgar scores than White infants. This was reinforced by the later description of how there was a lack of research on the 'racial differences in the applicability of the scores..".

Reading this, I was inclined to conclude that validating Apgar scores (not just predictive validity) or trying to understand these differentials might be the focus of the study. But this is not the case. The paper, in my reading simply shows that in a country where Black infants are twice as likely to die as White infants (cited by the authors) that Apgar scores predict this and that Non-Hispanic Black and Non-Hispanic Other groups have higher odds of being assigned a low score. Without wishing to sound dismissive - how is this interesting? Is that not exactly what would be expected.

Had the paper been set up to show how systematic racism was implicated in the assigning of Apgar scores - i.e. higher or lower scores compared to gold standard and compared to other race groups - now that would be really interesting. If the study was able to show how Apgar scores differed by race from gold standard (irrespective of infant birth status) - that would be interesting.

As a result I do not think the paper passes muster in terms of novelty for a public health journal of the stature of PLOS Medicine. It should definitely be published but perhaps in a specialist journal such as an Obstetric journal.

Reviewer #3: Thank you for the opportunity to review this manuscript. In this paper, the authors examine the association of low Apgar scores and infant mortality across different racial groups using vital statistics data of infants born in the US. The paper finds that the five-minute Apgar score is strongly associated with infant mortality but identified differences in the strength of association between across different racial groups. Specifically, the strength of association between low Apgar score and mortality was reduced in Black and non-Hispanic non- Asian groups suggestive of reduced predictive ability of mortality in these groups.

The authors have examined an important topic that could make a significant contribution to the literature. The study has some significant implications for the interpretation and applicability of Apgar scores and their prediction of mortality across different racial groups.

Overall, this is a good paper but could be improved by being more in-depth in its descriptions and analysis of the findings and the meaning of the results as currently some sections are quite superficial. The methods section could be more detailed and the results section is currently very brief. The authors need to more specifically outline what the findings mean for the application and use of Apgar scores across different racial groups apart from stating that caution should be taken when interpreting the scores. I was expecting to see a section on why was the study done, what do these findings mean etc. to really highlight the key takeaway points - this would be helpful in drawing out the key findings of this paper as well.

More attention should be paid to detail as there were repeating text in the manuscript and parts that have not been updated after previous revisions (e.g STROBE statement page numbers)

I have made some further specific comments below for the authors consideration to revise the manuscript.

Specific comments

ABSTRACT

p2, Methods and findings: The detailed description in the results of how many infants of each race were included and excluded cases is not essential for the abstract. Perhaps you can keep the % and drop the n. I would suggest to review this and focus on presenting the most important results.

INTRODUCTION

p3, para 1-2: I would recommend the text in paragraph 2 of the introduction be moved up into paragraph 1 (after the first sentence) and the detail about the Apgar score from the second sentence be moved down into a separate paragraph. This is so the problem your study is investigating is upfront. The detail of the components of Apgar score is not so important that it needs to be in the first paragraph.

p4, para 2: missing word - "and" before "discrimination"

p5, para 1: The last sentence of the introduction ("The database consists of…") is not essential here and would be more appropriate in the methods section where a description of the data source is provided.

METHODS

The ethics sub-section can be placed towards the end of the methods section. The ethics statement is also repeated in the Study Population section at the beginning of page 4. This can be removed.

The methods section does not follow the STROBE checklist and the page numbers in the checklist do not match the current manuscript (e.g there is no Study Design section in the methods but the checklist in the Appendix indicates it is on pages 8-11 in the manuscript). This should be the first section or combined with study population. I would suggest the authors structure the methods section using similar headings to the strobe statement. The "Database" sub-heading should be "Data sources" for example.

Also, the current STROBE statement checklist is missing the first column that indicates which section of the manuscript the information can be found.

p7, para 2: Outcomes, exposures, covariates - this section needs to be better organised. I would suggest describing the outcome variable first followed by the explanatory variables.

p8, top paragraph: There is a bit of a mix up or description of analysis and variables in this section as well - the part on what was adjusted for should be in the analysis section but the definition of the variables should remain in this section. Handling of missing values also belongs in the analysis section.

Figure 1: please improve the quality of the flow diagram (font is currently blurred) and consider reducing amount of text and perhaps include the use of boxes.

RESULTS

p10, para 1: All of the numbers and % in the table do not need to be repeated in the text. Suggest to mention the % for race without the numbers in the text.

On the other hand, on p13 there is no description of the magnitude of the odds described at all. Instead of stating that there is only a lower or higher odds, the authors can be more specific - for example, "…non-Hispanic black infants had almost two times higher odds of being assigned a low Apgar score…" Otherwise the information in the text is not very informative.

p15-16. Please indicate which table the statement made in the third sentence in paragraph 1 is located.

There is no need to describe the chi-square test here - this belongs in the methods/analysis section.

The main results/tables of your study are in the Appendix - I would suggest to move (some) of these to the main manuscript. The unadjusted analyses I would think were less important and could be in the appendix.

Overall, the description of your results is very brief given the amount of results and tables included.

Tables S6-S8 in the Appendix need to be fit to the page as the last columns are cut off. Suggest to include the word documents with the original tables rather than an image.

DISCUSSION

p16, first para: The first paragraph of a discussion should highlight the most important findngs of the paper. However the authors describe the unadjusted results.

Please also avoid repeating results but rather explaining or interpreting the data in the discussion

Reviewer #4: Thanks for allowing me to review your very interesting manuscript. I have several comments, and some questions which may lead to some revisions.

1. I am curious as to why you excluded births to mothers under the age of 15? Whilst the numbers are likely to be small they are a high-risk group, more likely to have births of compromised babies due to fetal growth restriction and other problems of placental dysfunction. I note that you did include women who did not complete 8th grade at school and if this group is included then perhaps the very young should also be there. Could you elaborate on your reasons for their exclusion?

2. Similarly, the inclusion of births up to 44 + 6 weeks of gestation is unusual as this might imply a problem of data accuracy. Even without intervention, which is almost universal now before 42 + 6 weeks, very few pregnancies would progress beyond 43 weeks and certainly not beyond 44. Why such a high figure?

3. What is the origin of the list of congenital abnormalities that have been excluded? It contains many obvious severe conditions (e.g., anencephaly) but also some others which are quite common and not really associated with infant mortality (e.g., hypospadias).

4. I noted that the explanation about the exemption from full ethics review is repeated in the sections on Ethics and Study Population which is a duplication.

5.The overall smoking rates are much lower than I would have expected (6.8%). In my own country (Australia) we are still seeing rates of 10-15% overall, and these vary a lot between different ethnic groups. Can you comment on this, and if you can validate this figure against any other data source? As it is a risk factor for both neonatal and infant mortality it is important to get this right in any analysis, especially when calculation the adjusted OR.

6. Finally, I am curious as to the predictive value of the Apgar scores for both neonatal mortality (the original use) and later infant mortality. Whilst some causes of infant mortality are linked to conditions which have their origins in the perinatal period many others are not necessarily linked (e.g. trauma, drowning etc). Have you looked separately at the infant deaths which occur beyond 28 days (about 50% of the total cohort of deaths), rather than including all deaths in the infant mortality numbers, which includes the early and late neonatal ones

Reviewer #5: Thank you for the opportunity to review this manuscript. Overall the analyses are well conducted and the manuscript well written. I have a few minor edits to suggest:

1. The following text should be combined combined with the Inclusion and Exclusion Criteria to make a single continuous paragraph on study population, which also means that the gestational age criteria do not need to be stated twice:

"Study Population

This population cohort study evaluated singleton infants born between 37+0 and 44+6 weeks between January 1, 2016 and December 31, 2017 in the United States."

2. The 2nd sentence of the Study Population paragraph should be added to the Database paragraph: "All data were nonidentifiable and publicly available through the NCHS Division of Vital Statistics and complied with the NCHS, Centers for Disease Control and Prevention (CDC) Data User 6

Agreement Terms and Conditions.23,24"

3. The rest of the Study Population paragraph is an exact repeat from the Ethics paragraph and should be deleted, except the last sentence on Strobe reporting guidelines which can be moved to the end of the methods section.

4. I did not quite follow the logic for the following: "the findings suggest that the odds of mortality based on a low Apgar score are underestimated in these groups" and suggest this statement be deleted. The remainder of the discussion and conclusion are much clearer in their interpretations.

Any attachments provided with reviews can be seen via the following link:

[LINK]

 10.1371/journal.pmed.1004040.r003 Author response to Decision Letter 1 Submission Version 2

31 Mar 2022

Submitted filename: Response to reviewers.docx

 10.1371/journal.pmed.1004040.r004 Decision Letter 2 Odeny Beryne Senior Editor 2022 Beryne Odeny This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Submission Version 2

10 May 2022

Dear Dr. Stock,

Thank you very much for re-submitting your manuscript "Associations between low Apgar scores and mortality by race in the United States: a cohort study of 6,809,653 infants" (PMEDICINE-D-21-05026R2) for review by PLOS Medicine.

I have discussed the paper with my colleagues and it was also seen again by three reviewers. I am pleased to say that provided the remaining editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosmedicine@plos.org.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.  

We look forward to receiving the revised manuscript by May 17 2022 11:59PM.   

Sincerely,

Beryne Odeny,

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

1) Abstract:

a) Please structure your abstract using the PLOS Medicine headings (Background, Methods and Findings, Conclusions).

b) Please combine the Methods and Findings sections into one section, “Methods and findings”

c) In the last sentence of the Abstract Methods and Findings section, please describe the main limitation(s) of the study's methodology

Comments from Reviewers:

Reviewer #1: Thank you for addressing reviewers' comments and improving the manuscript. My questions in the previous round has been suffuciently answered.

1. In Table 2 and Table 3, why there's only one p-value for multiple tests for interaction? For instance, in Table 2, there should be seperate p-values for Non-Hispanic white vs Hispanic, and Non-Hispanic white vs Non-hispanic black. In table 3, there should be p-value for normal vs intermedaite comparison, and p-value for normal vs low comparison.

2. According to the Office of Management and Budget Standars for the Classification of Federal Data on Race and Ethnicity, Hispanic ethnicity is recorded as distinct from race. In this paper, the author used different category and treated Hispanic as a mutually exclusive racial group from other racial groups. Explanation is needed for the selction of the categories for the analysis.

Reviewer #2: The authors have responded to reviewer comments

Reviewer #3: Thank you to the authors for their efforts to address the comments and revise the manuscript. The revisions made have improved the manuscript somewhat but I'm afraid there remain some substantial issues which need to be addressed particularly in relation to the methods section and discussion. I have outlined these below:

Minor revisions

ABSTRACT

Line 45 - Stating the limitations is not needed in the abstract

Line 49 (findings) - In the sentence, "A total of 6, 728,829 infants…" the percentages should be included after the numbers for easier interpretation. It's not clear why they were removed from the previous version.

Line 52 - abbreviations (AOR) are usually not used in an abstract - please check with journal requirements. As it is not used multiple times in the abstract it may not be necessary.

AUTHORS' SUMMARY

This is a very nice overall summary. Just one small suggestion:

Line 90 - please remove the "n" as it doesn't add anything here and it is stated clearly in the title of the paper.

Major revisions

METHODS

The methods section needs reorganisation - grouping study design, setting and participants is not following the STROBE guidelines and these are all very different topics, especially participants (study setting may not be needed). This section now also discusses the exclusion and inclusion criteria.

There are now two sections with the heading of "variables" one called "quantitative variables" - all Generally, there should be description of the variables - dependent and independent. The statistical analysis should be separate. Please closely review the STROBE guidelines and other similar papers for order and content of the methods section.

I do not have statistical expertise but is this multivariable regression not multivariate?

Line 213 (Statistical methods and Quantitative Variables) - Please separate out this section into paragraphs as it is too long as one piece of text.

Line 251 - this statement on STROBE should be at the beginning of the methods not in the Ethics statement.

RESULTS

All of the figures are still very low quality and not clear.

Table 1 - this table is very hard to read and has no lines - please format and add some spacing between the variables.

DISCUSSION

There seems to be a disconnect between the first paragraph of the discussion and the author's summary around the implications of the findings. The discussion is very positive and supportive of the Apgar score while the Authors' summary is more critical indicating that Apgar scores are less useful among non-White populations. The overall conclusions should be based on the results only and need to be consistently communicated throughout the paper.

In my previous review I indicated that the authors should avoid only restating the results in the discussion but instead elaborate on what their results mean. Instead of doing this the text was just removed. The discussion cannot consist only of a summary and implications. The results need to be discussed more adequately and critically.

Any attachments provided with reviews can be seen via the following link:

[LINK]

 10.1371/journal.pmed.1004040.r005 Author response to Decision Letter 2 Submission Version 3

16 May 2022

Submitted filename: PLOS Med Revision Responses II.docx

 10.1371/journal.pmed.1004040.r006 Decision Letter 3 Odeny Beryne Senior Editor 2022 Beryne Odeny This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Submission Version 3

31 May 2022

Dear Dr. Stock,

Thank you very much for re-submitting your manuscript "Associations between low Apgar scores and mortality by race in the United States: a cohort study of 6,809,653 infants" (PMEDICINE-D-21-05026R3) for review by PLOS Medicine.

I have looked at the paper and it was also seen again by one reviewer. I am pleased to say that provided the remaining minor editorial and production issues are dealt with we are planning to accept the paper for publication in the journal.

The remaining issues that need to be addressed are listed at the end of this email. Any accompanying reviewer attachments can be seen via the link below. Please take these into account before resubmitting your manuscript:

[LINK]

***Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.***

In revising the manuscript for further consideration here, please ensure you address the specific points made by each reviewer and the editors. In your rebuttal letter you should indicate your response to the reviewers' and editors' comments and the changes you have made in the manuscript. Please submit a clean version of the paper as the main article file. A version with changes marked must also be uploaded as a marked up manuscript file.

Please also check the guidelines for revised papers at http://journals.plos.org/plosmedicine/s/revising-your-manuscript for any that apply to your paper. If you haven't already, we ask that you provide a short, non-technical Author Summary of your research to make findings accessible to a wide audience that includes both scientists and non-scientists. The Author Summary should immediately follow the Abstract in your revised manuscript. This text is subject to editorial change and should be distinct from the scientific abstract.

We expect to receive your revised manuscript within 1 week. Please email us (plosmedicine@plos.org) if you have any questions or concerns.

We ask every co-author listed on the manuscript to fill in a contributing author statement. If any of the co-authors have not filled in the statement, we will remind them to do so when the paper is revised. If all statements are not completed in a timely fashion this could hold up the re-review process. Should there be a problem getting one of your co-authors to fill in a statement we will be in contact. YOU MUST NOT ADD OR REMOVE AUTHORS UNLESS YOU HAVE ALERTED THE EDITOR HANDLING THE MANUSCRIPT TO THE CHANGE AND THEY SPECIFICALLY HAVE AGREED TO IT.

Please ensure that the paper adheres to the PLOS Data Availability Policy (see http://journals.plos.org/plosmedicine/s/data-availability), which requires that all data underlying the study's findings be provided in a repository or as Supporting Information. For data residing with a third party, authors are required to provide instructions with contact information for obtaining the data. PLOS journals do not allow statements supported by "data not shown" or "unpublished results." For such statements, authors must provide supporting data or cite public sources that include it.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript.

Please note, when your manuscript is accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you've already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosmedicine@plos.org.

If you have any questions in the meantime, please contact me or the journal staff on plosmedicine@plos.org.  

We look forward to receiving the revised manuscript by Jun 07 2022 11:59PM.   

Sincerely,

Beryne Odeny,

Senior Editor 

PLOS Medicine

plosmedicine.org

------------------------------------------------------------

Requests from Editors:

1. Abstract: please revise the subheading "Interpretation" to "Conclusion" in line with PLOS Medicine's style

2. Discussion section: please delete the subheadings including "Findings in context", "Implications" and "Strengths and Limitations"

Comments from Reviewers:

Reviewer #3: Thank you to the authors for the revisions to the manuscript. The paper is now much clearer and flows well. I have no further comments on this version.

I look forward to seeing this published.

Any attachments provided with reviews can be seen via the following link:

[LINK]

 10.1371/journal.pmed.1004040.r007 Author response to Decision Letter 3 Submission Version 4

31 May 2022

Submitted filename: PLOS Med Revision Responses III.docx

 10.1371/journal.pmed.1004040.r008 Decision Letter 4 Odeny Beryne Senior Editor 2022 Beryne Odeny This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Submission Version 4

1 Jun 2022

Dear Dr Stock, 

On behalf of my colleagues and the Academic Editor, Dr. Mark Tomlinson, I am pleased to inform you that we have agreed to publish your manuscript "Associations between low Apgar scores and mortality by race in the United States: a cohort study of 6,809,653 infants" (PMEDICINE-D-21-05026R4) in PLOS Medicine.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Once you have received these formatting requests, please note that your manuscript will not be scheduled for publication until you have made the required changes.

In the meantime, please log into Editorial Manager at http://www.editorialmanager.com/pmedicine/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production process. 

PRESS

We frequently collaborate with press offices. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximise its impact. If the press office is planning to promote your findings, we would be grateful if they could coordinate with medicinepress@plos.org. If you have not yet opted out of the early version process, we ask that you notify us immediately of any press plans so that we may do so on your behalf.

We also ask that you take this opportunity to read our Embargo Policy regarding the discussion, promotion and media coverage of work that is yet to be published by PLOS. As your manuscript is not yet published, it is bound by the conditions of our Embargo Policy. Please be aware that this policy is in place both to ensure that any press coverage of your article is fully substantiated and to provide a direct link between such coverage and the published work. For full details of our Embargo Policy, please visit http://www.plos.org/about/media-inquiries/embargo-policy/.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Thank you again for submitting to PLOS Medicine. We look forward to publishing your paper. 

Sincerely, 

Beryne Odeny 

PLOS Medicine