Mortality, morbidity and clinical care in a referral neonatal intensive care unit in Haiti.

BACKGROUND: Neonatal mortality rates in Haiti are among the highest in the Western hemisphere. Few mothers deliver with a skilled birth attendant present, and there is a significant lack of pediatricians. The neonatal intensive care unit (NICU) at St. Damien Pediatric Hospital, a national referral center, is one of only five neonatology departments in Haiti. In order to target limited resources toward improving outcomes, this study seeks to describe clinical care in the St. Damien NICU.
METHODS: A retrospective medical record review was performed on available medical records on all admissions to the NICU between April 2016 and April 2017.
RESULTS: 220 neonates were admitted to the NICU within the study epoch. The mortality rate was 14.5%. Death was associated with a maternal diagnosis of hypertension (p = 0.03) and neonatal diagnoses of lower gestational age (p<0.0001), lower birth weight (p<0.0001), prematurity (p = 0.002), RDS p = 0.01), sepsis (p<0.0001) and kernicterus (p = 0.04). The most common diagnoses were sepsis, chorioamnionitis, respiratory distress syndrome, jaundice, prematurity and perinatal asphyxia.
CONCLUSIONS: This study demonstrates that preterm birth, sepsis, RDS and kernicterus are key contributors to neonatal mortality in a Haitian national pediatric referral center NICU and as such are promising interventional targets for reducing the neonatal mortality rate in Haiti.

Introduction

Neonatal and infant mortality rates in Haiti are among the highest in the Western hemisphere, with rates that are markedly disparate from those of its neighboring country on the island of Hispaniola, the Dominican Republic. The current neonatal mortality rate was 32 per 1000 live births, compared to the Dominican Republic’s rate of 16 per 1000 live births [1]. Of all deaths of children under the age of 5 in Haiti, 34% died within the first 28 days of life, the neonatal period [2]. Global Health Observatory data shows that between 2011–2017, only 41.6% of mothers delivered with a skilled birth attendant present [3]. The large percentage of women who deliver their infants without a skilled healthcare worker present and the significant shortage of neonatal intensive care units contribute to the high neonatal mortality rates [4]. Major causes of neonatal death include prematurity (34%), birth asphyxiation and trauma (26%), and sepsis/other infectious conditions (18%) [5]. In one study of a neonatal ICU in Haiti, sepsis accounted for 54.8% of admissions to the NICU between 2013 and 2015 [6].

Haiti is classified by the World Bank as low-income and is currently the poorest country in the Western Hemisphere, with more than half of Haitians living below the national poverty line [7]. Of the population of over ten million, one third are under the age of 15 [8]. While there are 23.5 physicians per 100,000 [9], the situation is even more dire for children. There are about 2.5 pediatricians per 100,000 children under the age of 15. In contrast, there are currently 101.2 certified pediatricians (including specialists) per 100,000 children in the United States [10]. The natural disasters that have occurred in the region have placed an additional strain on already limited healthcare services. There is a need for progressive, cost-effective health care practices that can alleviate both individual and governmental cost burden.

St. Damien Pediatric Hospital is a private non-profit hospital run by NPH/NPFS (Nuestros Pequeños Hermanos/Nos Petit Frères et Sœurs) in Tabarre, Haiti, supported through fundraising and charitable donations. The hospital was created in 1989, responding to need in the area for a hospital that could treat debilitating illnesses taking children’s lives. Its mission is to not only care for the sick, but to offset the injustices of poverty and unemployment that make healthcare inaccessible for many individuals in the country. The hospital provides quality and dignified healthcare to children of any social level in emergent condition, and encourages parent participation through ongoing dialogue, educational opportunities, and offering any material support available to families.

The neonatal intensive care unit (NICU) at St. Damien Pediatric Hospital is one of only five neonatology departments currently operating in Haiti. The NICU was established after the devastating 2010 earthquake, when damage to other hospitals in the area exacerbated the need for specialized neonatal care. The NICU currently has twenty beds and is staffed with pediatric physicians with neonatal training, as well as trained neonatal nurses, who provide specialized bedside care to the neonatal patients.

This study seeks to describe clinical care and outcomes in the St. Damien NICU, a national referral center NICU in Haiti, in order to target limited resources toward improving outcomes.

Materials and methods

Data collection

A quantitative, descriptive, retrospective medical record review was performed on available medical records within the study timeframe. Haitian Institutional Review Board approval was obtained through Comité National de Bioéthique, which waived the requirement for informed consent. Data were collected from the medical records of newborns admitted to the NICU at St. Damien Pediatric Hospital between April 2016 and April 2017. Data were fully anonymized upon accessing and collection from the medical record. Inclusion criteria for the study included every infant admitted to the St. Damien NICU during the study period. The admissions book in the NICU was used to identify the names, birth date, birthweight, gestational age, mortality, and chart number of neonates. Typically, this information is recorded at time of admission or at death. Then, charts were physically identified and data transcribed in the hospital archives. Data collection included maternal and infant demographics, diagnoses, laboratory results, interventions and therapies. Due to flooding of basement archives, where charts of deceased patients were kept, some charts of neonates who had died were unable to be reviewed. These infants were excluded from the study, which in turn decreased the calculated mortality rate.

Data analysis

Descriptive statistics for the sample as a whole and by survival are shown. Comparisons between deaths and survivors were made by t-tests for normally distributed continuous variables and Wilcoxon tests when the data were non-normally distributed. Categorical variables were analyzed using the chi-square test, or in the case of small cell sizes, Fisher’s exact test. Statistical significance was defined as a two-sided p-value < 0.05.

Results

A total of 220 neonates admitted to the St. Damien Pediatric Hospital NICU were identified within the study epoch.

Of these, 32 (14.5%) died during the hospitalization. Those who died were more likely to have a maternal diagnosis of hypertension compared to those who survived to discharge, as well as lower gestational age, lower birth weight, and shorter length of hospitalization. The diagnoses associated with an increased outcome of death included prematurity, RDS, sepsis and kernicterus. Diagnoses associated with a decreased outcome of death were chorioamnionitis, transient tachypnea of the newborn (TTN) and jaundice (Table 1).

Table 1

Maternal and neonatal characteristics.

Characteristics (mean ± SD; n (%))	All admissions n = 220	Survivors n = 188 (85.0%)	Deaths n = 32 (14.5%)	Pa
Maternal
Age (n = 185)	29.5 ± 6.9	29.6 ± 6.9	28.7 ± 6.6	0.54
Gravida (n = 187)	2.3 ± 1.6	2.2 ± 1.5	2.8 ± 2.3	0.59
Para (n = 186)	1.4 ± 1.4	1.3 ± 1.2	1.9 ± 2.1	0.53
Cesarean section (n = 220)	115 (52.3)	100 (57.1)	14 (46.7)	0.50
HIV (n = 220)	3 (1.4)	1 (0.54)	2 (6.3)	0.058
Syphilis (n = 220)	3 (1.4)	2 (1.1)	1 (3.2)	0.37
Alcohol use (n = 220)	2 (0.9)	2 (1.1)	0 (0)	1.00
Tobacco use (n = 220)	0 (0)	0 (0)	0 (0)	n/a
Drug use (n = 220)	1 (0.5)	1 (0.54)	0 (0)	1.00
Hypertension (n = 220)	25 (12.9)	17 (10.4)	8 (27.6)	0.03
Infant
Gestational age (n = 218)	36.4 ± 4.1	37.1 ± 3.5	32.1 ± 4.8	<0.0001
Female (n = 220)	103 (46.8)	84 (44.9)	18 (56.3)	0.25
Birth weight (n = 219)	2304 ± 888	2427 ± 830	1597 ± 906	<0.0001
Singleton (n = 220)	206 (93.6)	176 (94.1)	29 (90.6)	0.44
Inborn (n = 220)	216 (98.8)	183 (97.9)	32 (100)	1.00
Apgar 1 min (n = 186)	6.1 ± 1.8	6.1 ± 1.8	5.7 ± 2.0	0.35
Apgar 5 min (n = 187)	7.2 ± 1.6	7.2 ± 1.6	6.8 ± 1.6	0.19
Apgar 10 min (n = 170)	7.8 ± 2.0	7.8 ± 2.0	7.3 ± 2.1	0.09
Length of hospitalization (n = 220)	14 ± 14.4	15.3 ± 15.0	6.5 ± 7.1	<0.0001
Neonatal diagnoses
Preterm	79 (35.9)	56 (30.0)	19 (59.3)	0.002
Chorioamnionitis	156 (70.9)	141 (75.4)	15 (46.9)	0.003
Sepsis	199 (91.5)	18 (9.6)	17 (53.1)	<0.0001
Respiratory distress syndrome (RDS)	91 (41.4)	71 (38.0)	20 (62.5)	0.01
Transient tachypnea of the newborn	21 (9.5)	20 (10.7)	0 (0)	0.05
Pneumonia	16 (7.4)	6 (3.2)	1 (3.1)	1.00
Meconium aspiration syndrome	19 (13.4)	12 (6.4)	2 (6.3)	1.00
Bronchopulmonary dysplasia	3 (1.4)	0 (0)	1 (3.1)	0.15
Perinatal asphyxia	62 (28.2)	55 (29.4)	10 (31.2)	0.84
Encephalopathy	11 (5.1)	0 (0)	1 (3.1)	0.15
Seizures	16 (7.4)	3 (1.6)	0 (0)	1.00
Meningitis	5 (2.3)	1 (0.5)	0 (0)	1.00
Hydrocephalus	2 (0.9)	1 (0.5)	1 (3.1)	0.27
Intraventricular hemorrhage (IVH)	4 (1.1)	2 (1.1)	2 (6.3)	0.10
Kernicterus	6 (1.6)	3 (1.6)	3 (9.4)	0.04
Jaundice	81 (36.8)	116 (62.0)	11 (34.4)	0.006
Congenital heart disease	10 (4.7)	14 (7.5)	2 (6.3)	1.00
Kidney disease	3 (1.6)	3 (1.6)	0 (0)	1.00
Necrotizing enterocolitis	10 (4.7)	5 (2.7)	1 (3.1)	1.00
Anemia	34 (16.9)	31 (16.6)	3 (9.4)	0.43
Patients who left AMA (n = 200)	20	20	n/a

Abbreviations: AMA = against medical advice

aSurvivors vs. Deaths

Most neonates were inborn. Over half of births occurred via Cesarean section and the mean ten minute Apgar score was 7.8 (Table 1).

During the study period, a total of 20 (10%) parents and their neonates left the hospital against medical advice (Table 1).

The most common diagnosis was sepsis, at 91.5%. Other common diagnoses were chorioamnionitis, respiratory distress syndrome, jaundice, prematurity and perinatal asphyxia (Table 1). Although prematurity is defined as a gestational age of less than 37 weeks, the diagnosis of prematurity was also made using the Ballard score, given unreliable dating in the setting of low rates of prenatal care and access to dating ultrasound.

Endotracheal intubation and conventional ventilator support were rare, while respiratory support, which 60% of admissions received, most commonly consisted of nasal CPAP and nasal cannula (Table 2). Intubation was associated with death (total intubations n = 8; intubation among infants who died n = 6; p = 0.003).

Table 2

Interventions.

Respiratory	N = 220 (%)
Intubation	8 (4)
Ventilation	129 (60)
CPAP (n = 130)	68 (52)
Nasal cannula (n = 130)	61 (47)
Conventional ventilation (n = 130)	1 (1)
Nutrition	N = 181 (%)
Breast milk	152 (84)
Formula	1 (0.6)
Breast milk and formula	28 (16)
Hyperbilirubinemia	N = 220 (%)
Phototherapy	118 (54)
Sunbath	40 (18)
	Mean ± SD
Total bilirubin (n = 114)	8.33 ± 4.3
Direct bilirubin (n = 114)	1.65 ± 1.53
Medication	N = 220 (%)
Ampicillin	218 (99)
Gentamicin	214 (97.3)
Ranitidine	173 (78.6)
Cefotaxime	121 (55.0)
Phenobarbital	114 (51.8)
Caffeine Citrate	78 (35.5)
Sodium bicarbonate	62 (28.2)
Ceftazidime	44 (20.0)
Metronidazole	41 (18.6)
Phenytoin	24 (10.9)

Abbreviations: CPAP = continuous positive airway pressure

Most infants received breastmilk, with few receiving breastmilk plus formula, and only one receiving formula alone (Table 2).

55% of neonates received phototherapy, and 19% received a sun bath, in which infants are placed in the sun in a cot as a therapy for jaundice (Table 2). Mean total bilirubin was 8.33.

Almost all (99%) of admitted infants received ampicillin and 97.3% received gentamicin during the course of their hospitalization (Table 2).

Discussion

Neonatal mortality and morbidity are increased in low resource settings, including Haiti [11]. In order to target limited resources toward improving outcomes, we performed a chart review of a national referral NICU in Haiti to ascertain areas for targeted interventions. Similar chart reviews in other low and mid income countries (LMICs), such as Uganda [12] have been performed and have shown that evaluation of current practices may lead to decreased mortality in low resource settings such as in Rwanda [13]. The differences in characteristics observed between the neonates who died in the NICU and those that survived to discharge suggest specific patterns of neonatal morbidity and mortality.

Maternal hypertension may lead to a decision to deliver an infant, subsequently leading to prematurity and thus lower gestational age and lower birth weight. Prematurity, in turn, is associated with an increased risk of RDS [14]. Meanwhile, sepsis is associated with prematurity in general, but is also well documented as a common cause of neonatal mortality in low resource settings [15]. Similarly, kernicterus is a complication of prematurity that is virtually nonexistent in otherwise healthy neonates in high resource settings but is prevalent in LMICs [16]. A shorter length of hospitalization is likely a reflection of neonates who die early during the course of hospitalization. The diagnoses associated with a decreased outcome of death, on the other hand, chorioamnionitis, TTN and jaundice, may be a reflection of these characteristics indicating a less severe manifestation of the same pathophysiology that the diagnoses associated with higher death rates, sepsis, RDS and kernicterus, represent. Additional factors that may affect length of hospitalization, in addition to diagnosis, may include socioeconomic factors, such as parental ability to take the infant home and care for them outside the hospital. The high number of deaths associated with intubation and the low number of intubated infants is reflective of the lack of access to an adequate number of ventilators and related supplies.

St. Damien Pediatric Hospital is a national pediatric referral center that also houses a labor and delivery suite. Because of previous nosocomial infectious outbreaks, the NICU is restricted to neonates who were born at St. Damien. A separate NICU for children born outside of St. Damien is currently being developed. Thus, most patients are inborn. The high Cesarean section rate is similar to rates in other limited resource settings, such as the Dominican Republic [17]. Given the fact that less than half of births occur in the presence of a trained birth attendant [3], women are more likely to present to the hospital without adequate prenatal care or during complicated labor after attempting home birth, thus leading to referral centers such as St. Damien Pediatric Hospital having a higher risk population. Although high perinatal asphyxia rates may be secondary to delivery practices, they are more likely also a reflection of this high risk patient population. Increasing the number of births that are attended by trained birth attendants in the community may be a viable strategy to decrease the rates of both Cesarean sections and perinatal asphyxia.

It is interesting to note the 10% rate of parents leaving the NICU with their infant against medical advice. A Saudi Arabian study demonstrated a rate of 1.6% [18], while a study in India showed a rate of 25.4% [19]. In the latter study, economic considerations were the most commonly indicated reason by parents, followed by lack of improvement and poor prognosis [19]. The most plausible explanation for this phenomenon within the Haitian economic and cultural framework is a lack of resources and the complex socioeconomic situation of families in addition to the medical challenges their infants face.

Sepsis was the most common diagnosis noted, and antibiotic usage was high. While multiple factors leading to increased rates of sepsis and mortality from sepsis have been documented in LMICs [15], evaluating these outcomes may provide direction in prioritizing effective countermeasures. The high rates of antibiotic use compared to rates in high resource settings such as 17% in a study of US NICUs [20] are likely a reflection of the sepsis rates.

Jaundice is another common diagnosis in the St. Damien NICU, with a rate of kernicterus that is higher than in high resource settings, such as a rate of 1.3 per 100,000 in a Swedish population based study [21]. Standard phototherapy was a common intervention, as well as sun baths when phototherapy was not available. Given the dangers associated with sun bath usage, the procurement of adequate phototherapy lights would likely lead to more appropriate therapeutic interventions.

An interesting observation is the very high rate of breastfeeding and exclusive breastfeeding compared to high resource settings. In the US, 83.8% of newborns were ever breastfed, 47.5% were exclusively breastfed at 3 months of age [22], and breastfeeding initiation in a NICU was 85% [23]. This may be secondary to the lack of access to newborn formula and thus a higher dependence on mother’s milk in the Haitian population.

A limitation of the study is that it is a retrospective chart review, and thus the data quality is lower than it would be had the data been collected prospectively, given the inherent limitation associated with data extracted from existing medical records. Thus, the internal validity of the study is not 100%. Nonetheless, all medical records that were available were reviewed, so the data represent as accurate a description of NICU patients during the study timeframe as is possible. The results of this study cannot be extrapolated to all of Haiti given its limitations, and thus external validity is also not 100%. However, because St. Damien Pediatric Hospital plays a unique role in Haitian healthcare for neonates as a national pediatric referral center, this study, which has elucidated outcomes for neonates beyond what was previously known about the state of neonatal care in Haiti, is a reflection of the highest level of neonatal care available in Haiti and as such is a useful tool in the development of effective neonatal interventions in Haiti.

As a next step, randomized studies of targeted neonatal interventions are necessary to achieve improved outcomes. In a trial in Guatemala, identifying women at risk for preterm birth, administering corticosteroids and encouraging delivering at a health care facility, reduced neonatal mortality [24].

Additional areas not reflected in this study include larger structures that can affect neonatal health care, such as sociopolitical climate, resource supply, and financial strain that are connected to the hospital’s functioning and thus, neonatal health outcomes.

Conclusions

In summary, this study demonstrates that preterm birth, sepsis, RDS and kernicterus are key contributors to neonatal mortality in a Haitian national pediatric referral center NICU and as such are promising interventional targets for reducing the neonatal mortality rate in Haiti.

19 Aug 2020

PONE-D-20-12677

Mortality, morbidity and clinical care in a referral neonatal intensive care unit in Haiti

PLOS ONE

Dear Dr. Beatrice Lechner,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by 16th October 2020. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Victor Adekanmbi, M.D., PhD

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In the ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records used in your retrospective study. Specifically, please ensure that you have discussed whether all data were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data from their medical records used in research, please include this information.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Manuscript Number: PONE-D-20-12677

A- Summary

Josie et al. aim at describing the clinical care and services provided at St Damien Neonatal Intensive Care Unit (NICU). To do so, they conducted a review of existing patient medical records over one year from April 2016 to Avril 2017. They used descriptive statistics including student and chi-square or Fisher exact tests to compare means and proportions, respectively. After describing the most common diagnosis by outcome status and the captured clinical interventions, they concluded that the findings provided an opportunity to develop targeting strategies to improve neonatal outcomes in Haiti.

Although there are significant gaps in the literature in describing the clinical features of NICU in Haiti, I have noticed critical methodological issues that should be addressed before this study can be considered as eligible for publication.

B- Major issues

• The method section only contains a few words on the statistical analyses performed. While they provided enough information on the study site in the background section, more information should be provided on the following elements:

a. The study design;

b. inclusion and exclusion criteria;

c. the outcome definition and calculation; need to be defined;

d. and data quality (Given the inherent limitation associated with data extracted from existing medical records);

• The way the results were being interpreted suggest an analytic study that would aim to establish the risk factors to neonatal death at the study site, which would require other types of study designs and statistical analysis. For instance, the following sentence clearly suggests that a relative risk has been estimated to establish the association between the risk factors and the outcomes: "The diagnoses associated with an increased risk of death included prematurity, RDS, sepsis, and kernicterus. Diagnoses associated with a decreased risk of death were chorioamnionitis, transient tachypnea of the newborn (TTN), and jaundice (Table 1)". Another example is this sentence: "Intubation was associated 158 with death (total intubations n=8; intubation among infants who died n=6; 159 p=0.003)". The most important point here is the need to keep the result interpretation consistent with the study objectives and methods.

• The study limitations that are subject to affect both internal and external validity should be clearly stated and particularly those inherent to the study design itself;

• The conclusion seems to suggest that these findings could be extrapolated to the whole country should be reconsidered once the study limitations are being clearly stated;

C- Minor issues

• The authors need to make sure that the most recent pieces of evidence are being considered. For instance, the current neonatal mortality rate in Haiti is 32 per 100 live birth (https://www.dhsprogram.com/pubs/pdf/FR326/FR326.pdf) instead of 25 per 100 live birth, as mentioned.

• Need an in-text citation for this sentence: "The large percentage of women who deliver their infants without a skilled healthcare worker present and the significant shortage of neonatal intensive care units contributes to the high neonatal mortality rates"; If available, it would be great if more socio-demographic information (maybe characteristics of the mothers) can be provided in table one beyond age;

• The percentage for the other diagnosis needs to be provided, including the n value "The most common diagnosis was sepsis, at 91.5%. Other common diagnoses 149 were chorioamnionitis, respiratory distress syndrome, jaundice, prematurity, and 150 perinatal asphyxia (Table 1)";

• It would be great if the author could present table 2 according to the neonates' outcomes as well;

• The authors should describe how data quality may have impacted the outcome (neonatal death) measurement (Potential information bias);

• The authors should explore other factors that can affect the Length of Stay in NICU;

Given both the major and minor issues listed above, I do not recommend publishing this article as submitted. Significant modification in the method and the conclusion sections will be needed to address the issues listed. It seems like the authors have enough variable to estimate a GLM to more highlight the roles of the contributors discussed in the neonatal death rate at the study site while clearly state the study limitations.

Reviewer #2: The paper is well articulated- the study purpose, methodology, data collection and analysis are well documented. The results of the study are also well documented and were performed to a high technical standard with sufficient detail. The results of this study have not been published elsewhere. The focus of this study represents findings that would be similar in most low resource setting countries. The authors have ably described the causes of neonatal mortality in a facility in Haiti and include causes such as maternal hypertension leading to low gestational age, low birth weight, prematurity, jaundice among others. Data was collected from medical records and was analyzed using T-test and wilcoxon tests and chi square. Overall 32 out of 220 neonates died representing 14.5%. The authors were able to provide conclusions in a comprehensive manner and the article is presented in a scholarly manner with high academic standard. The research therefore meets all required standards for publication. There is evidence of ethical approval and research integrity. However, despite mentioning the issue of low Skilled Birth Attendants and low number of pediatricians, the authors did not discuss the role of neonatal nurses in such a health care setting as they would normally spend more time in the clinical setting.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Professor Address Malata

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (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. Registration is free. 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 PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

11 Sep 2020

We thank the reviewers for their encouraging comments and their thorough and helpful critique. We have carefully considered their comments and have made changes based on them that we believe improve the overall manuscript. Most noteworthy is the addition of further details in Methods as well as the addition of study limitations to the Discussion. Please find below a point-by-point response to the Reviewer’s critique.

Reviewer 1

Critique 1: “The method section only contains a few words on the statistical analyses performed. While they provided enough information on the study site in the background section, more information should be provided on the following elements: a. The study design; b. inclusion and exclusion criteria; c. the outcome definition and calculation; need to be defined; d. and data quality (Given the inherent limitation associated with data extracted from existing medical records).”

We have added additional information on the statistical analysis to the methods section (lines 130-135), as well as further clarifications on points a and b (lines 110, 116-117, 125-126). We added a paragraph on point d to the Discussion (lines 226-279). Given that this study is a descriptive study, an outcome definition or calculation is not appropriate.

Critique 2: “The way the results were being interpreted suggest an analytic study that would aim to establish the risk factors to neonatal death at the study site, which would require other types of study designs and statistical analysis. For instance, the following sentence clearly suggests that a relative risk has been estimated to establish the association between the risk factors and the outcomes: "The diagnoses associated with an increased risk of death included prematurity, RDS, sepsis, and kernicterus. Diagnoses associated with a decreased risk of death were chorioamnionitis, transient tachypnea of the newborn (TTN), and jaundice (Table 1)". Another example is this sentence: "Intubation was associated 158 with death (total intubations n=8; intubation among infants who died n=6; 159 p=0.003)". The most important point here is the need to keep the result interpretation consistent with the study objectives and methods.”

In order to clarify the objectives of the study and thus the interpretation of the results more clearly, we have edited the abstract and background (lines 32-34, 102-104) to state that this study seeks to describe clinical care in the St. Damien NICU in order to target limited resources toward improving outcomes. We did not calculate relative risks. In order to avoid confusion, we changed the word “risk” to “outcome” to more accurately reflect our analysis (lines 146-147, 209). Finally, this retrospective cohort study is adequate to describe the relationship of characteristics with death, without positing that these are causes.

Critique 3: “The study limitations that are subject to affect both internal and external validity should be clearly stated and particularly those inherent to the study design itself.”

We have added a discussion of internal and external validity to the Discussion (lines 266-279).

Critique 4: “The conclusion seems to suggest that these findings could be extrapolated to the whole country should be reconsidered once the study limitations are being clearly stated.”

We have clarified in the Discussion (lines 266-279) that there are limitations to the study that preclude extrapolating to all of Haiti based on this study.

Critique 5: “The authors need to make sure that the most recent pieces of evidence are being considered. For instance, the current neonatal mortality rate in Haiti is 32 per 100 live birth (https://www.dhsprogram.com/pubs/pdf/FR326/FR326.pdf) instead of 25 per 100 live birth, as mentioned.”

Thank you for bringing the more current citation to our attention. We have updated the manuscript (line 60).

Critique 6: “Need an in-text citation for this sentence: "The large percentage of women who deliver their infants without a skilled healthcare worker present and the significant shortage of neonatal intensive care units contributes to the high neonatal mortality rates"; If available, it would be great if more socio-demographic information (maybe characteristics of the mothers) can be provided in table one beyond age.”

We have added a citation for the sentence above (line 63). Unfortunately, additional demographic information regarding the mothers beyond what was presented in Table 1 was not available in the charts.

Critique 7: “The percentage for the other diagnosis needs to be provided, including the n value "The most common diagnosis was sepsis, at 91.5%. Other common diagnoses 149 were chorioamnionitis, respiratory distress syndrome, jaundice, prematurity, and perinatal asphyxia (Table 1).”

The percentages, as well as the n values, for each diagnosis, are listed in Table 1.

Critique 8: “It would be great if the author could present table 2 according to the neonates' outcomes as well.”

We weighed this suggestion in depth. While Table 1 was structured according to the outcomes of death vs. survival so as to demonstrate which pre-existing patient characteristics were associated with death, the role of Table 2 was to show which interventions were available in the NICU and utilized. Thus, demonstrating outcomes according to interventions utilized would not result in meaningful metrics.

Critique 9: “The authors should describe how data quality may have impacted the outcome (neonatal death) measurement (Potential information bias).”

We have included this information in the Methods section (line 125).

Critique 10: “The authors should explore other factors that can affect the Length of Stay in NICU.”

We have included an exploration of this topic under Discussion (lines 212-215).

Critique 11: “It seems like the authors have enough variable to estimate a GLM to more highlight the roles of the contributors discussed in the neonatal death rate at the study site while clearly state the study limitations.”

While there may be enough deaths in the data to do a GLM (generalized linear model/logistic regression), those models usually involve designating a particular variable as a focal cause of death and include other variables as possible confounders. This is a descriptive study looking at associations of a number of different outcomes relating to death, not one particular thing. Thus, a GLM would not add useful information.

Reviewer 2

Critique 1: “However, despite mentioning the issue of low Skilled Birth Attendants and low number of pediatricians, the authors did not discuss the role of neonatal nurses in such a health care setting as they would normally spend more time in the clinical setting.”

Thank you for pointing out this oversight on our part. In addition to birth attendants and pediatricians, neonatal nurses play a central role in providing care to neonates. We have added this information to the Introduction (lines 98-100).

Submitted filename: Response to Reviewers.doc

Click here for additional data file.

28 Sep 2020

Mortality, morbidity and clinical care in a referral neonatal intensive care unit in Haiti

PONE-D-20-12677R1

Dear Dr. Beatrice Lechner,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Victor Adekanmbi, M.D., PhD

Guest Editor

PLOS ONE

5 Oct 2020

PONE-D-20-12677R1

Mortality, morbidity and clinical care in a referral neonatal intensive care unit in Haiti

Dear Dr. Lechner:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr Victor Adekanmbi

Guest Editor

PLOS ONE