Ciência & Saúde Coletiva

cienciaesaudecoletiva.com.br DOI: 10.1590/1413-81232024291.17462022EN

ISSN 1413-8123. v.29, n.1

Near miss neonatal in the capital of the Brazilian Midwest: 1

a case-control study

FREE THEMES
Priscilla Shirley Siniak dos Anjos Modes (https://orcid.org/0000-0003-2039-4505) 1
Maria Aparecida Munhoz Gaíva (https://orcid.org/0000-0002-8666-9738) 2
Amanda Cristina de Souza Andrade (https://orcid.org/0000-0002-3366-4423) 3
Elizabeth Fujimori (https://orcid.org/0000-0002-7991-0503) 4

Abstract We aimed to analyze factors associated

with neonatal near-miss in Cuiabá, State of Mato
Grosso, Brazil by performing a case-control study
of live births in a capital city of central-western
Brazil from January 2015 to December 2018 that
included 931 cases and 1,862 controls. Data were
obtained from the Live Births Information System
and the Mortality Information System and vari-
ables were organized according to the hierarchi-
cal model. Association was analyzed by logistic
regression with a 5% significance level. Data were
expressed as crude and adjusted odds ratio (OR)
1
Programa de Pós- and respective confidence intervals (95%CI). The
Graduação em Enfermagem, following factors were associated with neonatal
Faculdade de Enfermagem,
near miss: mothers with two (OR = 1.63; 95%CI:
Universidade Federal de
Mato Grosso. Av. Alexandre 1.01-2.63) or three or more previous pregnancies
Ferronato 1200. Bairro (OR=1.87; 95%CI: 1.09-3.21), without any live
Residencial Cidade Jardim.
children (OR = 2.57; 95%CI: 1.56-4.24 ) or one
78550-728 Sinop MT Brasil.
priscilladosanjos@ live child at birth (OR = 1.53; 95%CI: 1.04-2.26),
yahoo.com.br multiple pregnancy (OR = 4.57; 95%CI: 2.95-
2
Universidade Federal de
7.07), fewer than six prenatal consultations (OR
Mato Grosso. Cuiabá MT
Brasil. = 2.20; 95%CI: 1.77-2.72), whose deliveries took
3
Programa de Pós- place in public/university hospitals (OR = 2.25;
Graduação em Saúde
95%CI: 1.60-3.15) or philanthropic hospitals (OR
Coletiva, Instituto de Saúde
Coletiva, Universidade = 1.62; 95%CI: 1.16-2.26), with non-cephalic
Federal de Mato Grosso. presentation (OR = 2.71 95%CI: 1.87-3.94) and
Cuiabá MT Brasil.
uninduced labor (OR = 1.47; 95%CI: 1.18-1.84).
4
Departamento de
Enfermagem em Saúde Key words Near miss, Newborn, Morbidity, In-
Coletiva, Escola de formation systems
Enfermagem, Universidade
de São Paulo. São Paulo SP
Brasil.
Cien Saude Colet 2024; 29:e17462022
 2
Modes PSSA et al.

Introduction “Cases” were defined as hospital-delivered

newborns, from mothers residing in Cuiabá, who
Despite a noteworthy decrease in infant mortali- survived the first 27 days of life despite having
ty rates over the past 30 years, in 2019 more than presented one of the NNM criteria , adapted ac-
5 million children under five years of age died all cording to the definition by Silva et al (2017)12:
over the world. Almost half of these deaths oc- birth weight < 1,500g, 5-minute Apgar < 7, gesta-
curred in the first 28 days of life, i.e., in the neo- tional age < 32 weeks and congenital malforma-
natal period1. tion, excluding mechanical ventilation, which is
These deaths in the neonatal period are part not available in the SINASC database (Table 1).
of an greater problem, i.e., neonatal morbidity. Its “Controls” consisted of infants who were born
full extent and the factors that prevented death alive at hospitals in Cuiabá, whose mothers re-
still need to be fully understood. In this sense, sided in the city, who did not present any of the
it is important to examine, reflect and study the adapted pragmatic NNM criteria and who sur-
process involved in identifying the characteris- vived the first 27 days of life.
tics of newborns who escaped death so that fu- An odds ratio of 1.8 was used to make up a
ture deaths may be avoided. Newborns who sur- sample, which required two controls for each
vive despite complications are called “near miss” case (2:1), including power of 80%, alpha error
cases2. The concept of neonatal near miss (NNM) of 5% and relative frequency of 10% of a given
refers to newborn children with a life-threaten- exposure factor, considering the number of an-
ing condition at birth or an organ dysfunction alyzed variables, some with unknown frequency
during the neonatal period, who almost died but in the studied population13.
eventually survived3. SINASC and SIM data were obtained from
It is estimated that the worldwide NNM the Municipal Health Department of Cuiabá in
rate is 2.6 to 8 times higher than that of neona- form of an Excel file and were pre-processed to
tal deaths4. Therefore, analysis of these cases has correct and standardize variables before selection
been recommended to understand health system of cases and controls14. Next, we checked double
failures in comparison with neonatal mortality registration and absence of data. In the SIM data-
studies2. However, few studies have been devel- base, we found six duplicate records and 13 lack-
oped in Brazil that focus on factors potentially ing information on the birth certificate number
associated with NNM5. (DN), all of which were excluded. After that, us-
To date, it is known that the following factors ing the DN number as an identification variable,
are associated with the outcome of NNM or may
increase its risk: advanced maternal age5-6, black
maternal skin color7, twins and multiparity6,
lack of prenatal or inadequate prenatal care6-10,
breech presentation11, cesarean delivery10-12, type
of hospital doing the delivery12 and fetal malfor- Table 1. Distribution of criteria for defining cases of
mation10. neonatal near miss. Cuiabá, State of Mato Grosso,
Despite advances in research on this topic, Brazil, 2015-2018 (n = 931).
there are still few epidemiological studies7 and Variables n (%)
factors associated with NNM11. Therefore, this Birth weight
study aimed to analyze the factors associated < ,1500 g 379 (40,71)
with neonatal near miss in live births in Cuiabá, ≥ 1,500 g 552 (59,29)
State of Mato Grosso, Brazil. Five-minute Apgar score
<7 260 (27.93)
≥7 669 (71.86)
Method No information available 2 (0.21)
Gestational age
We performed a case-control study with live < 32 weeks 400 (42.96)
births in the city of Cuiabá, capital of the state of ≥ 32 weeks 524 (56.28)
Mato Grosso (MT), central-west region of Brazil No information available 7 (0.75)
from January 2015 to December 2018 and used Congenital malformation
data from the Born-Alive Infant Information Yes 187 (20.09)
System (SINASC) and the Mortality Information No 744 (79.91)
System (SIM). Source: Authors.
 3

Ciência & Saúde Coletiva, 29(1):1-12, 2024

a deterministic linkage was performed between pregnancy (single, double or more). Variables
the SINASC and SIM banks15. related to health care during pregnancy and de-
During the analyzed period, 40,741 children livery were characterized at the proximal level by:
were born, of which 306 (0.75%) died within the quarter in which prenatal care began (first, sec-
first month of life and were excluded from the ond, third), depending on gestational age; num-
study. Of the 40,435 survivors, 931 (2.30%) pre- ber of prenatal consultations (< 6, ≥ 6) ; health
sented at least one of the criteria adapted from establishment where the delivery took place, i.e.,
NNM at birth and made up the “cases”. Of the type of hospital (private hospital, private hospital
39,504 eligible subjects, 1,862 “controls” were associated with the Public Health System (SUS),
randomly selected, resulting in a final sample of philanthropic hospital, and public hospital that
2,793 live births (931 cases and 1,862 controls). merged with a university hospital); fetal presen-
The NNM was the dependent variable. For tation (cephalic, non-cephalic ([frank breech/
the analysis of variables taken from SINASC as- complete breech/transverse lie]); type of delivery
sociated with NNM cases, a hierarchical model7 (vaginal, cesarean); induced labor (yes/no). The
was adapted, which was based on the conceptual only characteristics of newborns analyzed was
theoretical model proposed to investigate factors their sex and year of birth.
associated with neonatal death10. In epidemio- Literature shows that the determinants of
logical studies using multivariate techniques, it is neonatal mortality and near miss morbidity are
suggested that the complex hierarchical interre- quite similar, including in twins.6 However, few
lationships between determinants be considered studies have investigated the relationship be-
to avoid underestimating the effects of distal (so- tween the type of pregnancy (single, double, tri-
cioeconomic) determinants, which may directly ple or more) and NM so far6,12. As the similar-
or indirectly affect all other variables, except sex ity between mortality and risk of complications
and age16. was understood, we decided to keep twins in the
Figure 1 shows the independent variables, analysis as an explanatory variable of the study.
which are organized into three hierarchical lev- Results of the descriptive analysis were pre-
els. Considering that the sex of the newborn is an sented in absolute and relative frequencies. The
important predictor of neonatal mortality10, this association between NNM cases and indepen-
variable was included in the final model, despite dent variables was analyzed using univariate and
the fact that it was not included in any of the lev- multiple logistic regression. Crude and adjusted
els7. odds ratio (OR) and the respective 95% confi-
The variables of this study were obtained dence intervals (CI) were used to measure as-
from SINASC and the categorization of mater- sociation. Variables with a p-value < 0.20 in the
nal sociodemographic variables at the distal level univariate analysis were included in the multiple
were: years of education completed (no school- model, following the proposed hierarchical lev-
ing, elementary school I and II, high school, full els.
and partial higher education, unknown) in years A hierarchical analysis was performed in
of schooling (≤ 8, 9-11, ≥ 12); professional oc- blocks according to the conceptual model (Fig-
cupation according to the Brazilian Classifica- ure 1). Newborn variables sex and year of birth
tion of Occupations (CBO, 2002), i.e., currently were inserted in the first model and variables of
working (yes/no), but the categories student, the distal level were included in the second mod-
housewife, unemployed, retired and pension- el, as well as sex and year of birth of the newborn,
ers were classified as “not working” and all the which were used as an adjustment. The signifi-
others occupations as “currently working”, and; cant variables (p ≤ 0.05) of the distal level were
race/color (white, brown, black, yellow and in- kept in the model and used to adjust the interme-
digenous). At the intermediate level, variables diate level block (model 3). The same procedure
were categorized into: mother’s age in years (< was repeated until the proximal variables were
20, 20-34, ≥ 35 years); marital status (lives with adjusted with the intermediate and distal ones
a partner [married, common-law marriage], (model 4). Those selected by the level of statis-
without a partner [single, widowed, separated/ tical significance at a certain level remained in
divorced]); number of previous pregnancies (0, the subsequent models, even if the inclusion of
1, 2, 3 or more), number of live births (0, 1, 2 or hierarchically inferior variables altered their lev-
more); number of fetal losses and abortions (0, 1, el of significance. A model was adjusted for each
2 or more); number of previous vaginal and ce- hierarchical level by excluding variables with the
sarean deliveries (0, 1, 2 or more); type of current highest pvalue and the model was re-evaluat-
 4
Modes PSSA et al.

Distal Level Intermediate level Proximal Level

Maternal Maternal characteristics

sociodemographic and Healthcare
characteristics reproductive history

- Age
- Quarter in which
- Years of study - Marital status
prenatal care started
(schooling) - Previous pregnancies
- Number of prenatal
- Work (usual - Number of vaginal and
consultations
occupation) cesarean deliveries
- Type of delivery
- Race/color - Number of children
- Induced labor
born alive
- Fetal presentation
- Number of fetal losses
- Type of hospital where
and abortions
the delivery took place
- Type of pregnancy

Neonatal near niss

Sex

Figure 1. Conceptual hierarchical model* for analysis of factors associated with cases of neonatal near miss.

* Adapted from Kale et al., 2017.

Source: Authors.

ed after each exclusion until all variables of the Regarding the intermediate level variables
same level remained significant. One model was (Table 3), there was a higher proportion of moth-
adjusted for each hierarchical level and the vari- ers between 20 and 34 years old (70.4%) who
ables at the most distal levels remained as adjust- lived with a partner (60.2%). There were more
ment factors for those at the hierarchically lower single pregnancies (95.0%), by women without
levels. All analyzes were performed using the any fetal loss or abortion (80.5%), without previ-
STATA Software version 12. A significance level ous pregnancy (38.3%), who therefore had never
of 5% was adopted. had a normal delivery (65.3%) or cesarean sec-
Our research project was assessed and au- tion (70.6%) and no live births (43.6%). The uni-
thorized by the Ethics Committee by approval nº variate analysis of the intermediate block showed
3.734.141 and CAE 25558619.0.0000.5541. that mothers without a partner who had had two
or more fetal losses/abortions in a previous preg-
nancy and who had had multiple pregnancies
Results were statistically associated with NNM.
Health care analysis at the proximal block
Tables 2 to 4 present the distribution of cases and level showed that most mothers started prena-
controls by hierarchical level according to inde- tal care in the first quarter (80.7%) and had six
pendent variables. Table 1 shows that more than or more consultations (80.1%), but that condi-
half of the newborns were male in both groups tion was statistically lower among cases (70.1%)
(54.0% cases and 51.7% controls). There was a compared to controls (85.1%). There was a high-
slightly higher percentage of NNM cases in 2016 er proportion of births by cesarean delivery
(37.1%). Most mothers were brown (72.1%), had (55.8%), non-induced delivery (78.4%) and ce-
between 9 to 11 years of schooling (60.6%) and phalic delivery (94.0%). Outcome was associated
were currently working (51.0%). In the distal with mothers who had had less than six prenatal
block, schooling and maternal race/color were consultations, non-induced labor, non-cephalic
associated with NNM (Table 2). fetal presentation and delivery in a private hos-
 5

Ciência & Saúde Coletiva, 29(1):1-12, 2024

Table 2. Distribution of cases and controls according to year of birth and sex of newborns and variables of the
distal hierarchical level (maternal sociodemographic characteristics). Cuiabá, State of Mato Grosso, Brazil, 2015-
2018 (n = 2,793).
Total Cases Controls Gross OR
Variables p-value
n (%) (%) n (%) (95%CI)
Of the newborn
Year of birth
2015 721 (25.8) 219 (30.4) 502 (69.6) 0.83 (0.66-1.03) 0.092
2016 668 (23.9) 248 (37.1) 420 (62.9) 1.12 (0.90-1.39) 0.312
2017 677 (24.2) 213 (31.5) 464 (68.5) 0.87 (0.69-1.09) 0.223
2018 727 (26.0) 251 (34.5) 476 (65.5) 1.00
Sex
Female 1,326 (47.5) 427 (46.0) 899 (48.3) 1.00
Male 1,465 (52.5) 502 (54.0) 963 (51.7) 1.10 (0.94-1.28) 0.248
Distal level:
Years of study
≤8 411 (14.8) 169 (18.2) 242 (13.0) 1.36 (1.06-1.75) 0.016
9-11 1,688 (60.6) 525 (56.6) 1,163 (62.5) 0.88 (0.73-1.06) 0.190
≥ 12 688 (24.6) 233 (25.1) 455 (24.5) 1.00
Work
Yes 1,419 (51.0) 473 (51.1) 946 (50.9) 1.01 (0.86-1.18) 0.924
No 1,366 (49.0) 453 (48.9) 913 (49.1) 1.00
Race/Color
White 592 (21.3) 208 (22.5) 384 (20.7) 1.08 (0.97-12.02) 0.948
Brown 2007 (72.1) 653 (70.7) 1,354 (72.9) 0.96 (0.87-12.60) 0.977
Black 170 (6.1) 61 (6.6) 109 (5.9) 1.12 (0.99-14.34) 0.927
Yellow 10 (0.3) 1 (0.1) 9 (0.5) 0.22 (0.09-5.27) 0.352
Indigenous 3 (0.1) 1 (0.1) 2 (0.1) 1.00
* Not obtained for cases and controls. OR: odds ratio; 95%CI: Confidence Interval. The distal model in which the variables were
significant (p ≤ 0.05) were retained in the model and included in the adjustment in the second block of the intermediary level.

Source: Authors.

pital affiliated with the Brazilian Public Health Discussion

Care System (SUS), in a public/university hospi-
tal, or in a philanthropic hospital (Table 4). In the present study, variables that showed an as-
In the multiple analysis, whose results are de- sociation with the NNM outcome were mothers
scribed in Table 4, the following variables showed who had already been pregnant twice or more,
a statistically significant association (p < 0.05) who had either not given birth or one live birth,
with NNM: gave birth to two (OR = 1.63; 95%CI: who had had multiple gestation, less than six pre-
1.01-2.63) or more children (OR = 1.87; 95%CI: natal consultations, delivery in public/university
1.09-3.21) in previous pregnancies, where no hospitals and philanthropic hospitals, non-ce-
child (OR = 2.57; 95%CI: 1.56-4.24) or one child phalic presentation and non-induced labor.
was born alive (OR = 1.53; 95%CI: 1.04-2.26), The findings of this study, in which women
multiple pregnancies (OR = 4.57; 95%CI: 2.95- who attended less than six prenatal consultations
7.07), less than six prenatal consultations (OR = had a greater chance of NNM are corroborated by
2.20; 95%CI: 1.77-2.72), delivery in public/uni- a study performed in Gujarat, India, which shows
versity hospitals (OR = 2.25; 95%CI: 1.60-3.15) that having had less than four prenatal consulta-
and philanthropic hospitals (OR = 2.25; 95%CI: tions was associated with a greater risk of NNM9,
1.60-3.15) OR = 1.62; 95%CI: 1.16-2.26), non-ce- as well as by another study conducted in Ambo,
phalic presentation (OR = 2.71; 95%CI: 1.87- Ethiopia8. Studies performed in Ethiopia12 and
3.94) and non-induced labor (OR = 1.47; 95%CI: in Brazil7 show that inadequate and low-quality
1.18-1.84) (Table 5). prenatal care contribute to unfavorable outcomes
 6
Modes PSSA et al.

Table 3. Distribution of cases and controls according to variables of the intermediate hierarchical level (maternal
characteristics and reproductive history). Cuiabá, State of Mato Grosso, Brazil, 2015-2018 (n = 2,793).
Total Cases Controls Gross OR
Variables* p-value
n (%) n (%) n (%) (95%CI)
Intermediate level
Maternal age (years)
< 20 398 (14.2) 134 (14.4) 264 (14.2) 1.00
20-34 1967 (70.4) 628 (67.4) 1.339 (71.9) 0.92 (0.73-1.16) 0.498
≥ 35 428 (15.3) 169 (18.2) 259 (13.9) 1.28 (0.98-1.71) 0.083
Marital status
with a partner 1,677 (60.2) 511 (55.2) 1,166 (62.6) 1.00
without a partner 1,110 (39.8) 415 (44.8) 695 (37.4) 1.36 (1.16-1.60) < 0.001
Previous pregnancies
0 1,057 (38.3) 367 (39.9) 690 (37.5) 1.00
1 880 (31.9) 275 (29.9) 605 (32.9) 0.85 (0.71-1.03) 0.106
2 430 (15.6) 140 (15.2) 290 (15.8) 0.91 (0.71-1.15) 0.425
≥3 393 (14.2) 138 (15.0) 255 (13.8) 1.02 (0.80-1.30) 0.889
Vaginal delivery
0 1,803 (65.3) 603 (65.5) 1,200 (65.2) 1.00
1 518 (18.8) 169 (18.4) 349 (18.9) 0.96 (0.78-1.19) 0.727
2 or more 440 (15.9) 148 (16.1) 292 (15.9) 1.01 (0.81-1.26) 0.939
Cesarean delivery
0 1,950 (70.6) 668 (72.5) 1,282 (69.7) 1.36 (0.98-1.87) 0.062
1 609 (22.1) 197 (21.4) 412 (22.4) 1.25 (0.88-1.77) 0.219
2 or more 202 (7.3) 56 146 (7.9) 1.00
Number of children born alive
0 1,201 (43.6) 428 (46.6) 773 (42.0) 1.17 (0.95-1.43) 0.127
1 911 (33.0) 283 (30.8) 628 (34.2) 0.95 (0.77-1.18) 0.667
2 or more 645 (23.4) 207 (22.6) 438 (23.8) 1.00
Fetal loss / abortion
0 2,214 (80.5) 713 (77.8) 1,501 (81.8) 1.00
1 412 (15.0) 152 (16.6) 260 (14.2) 1.23 (0.99-1.53) 0.063
2 or more 124 (4.5) 51 (5.6) 73 (4.0) 1.47 (1.02-2.13) 0.040
Type of pregnancy
Single 2,649 (95.0) 823 (88.5) 1,826 (98.2) 1.00
Two or more 141 (5.0) 107 (11.5) 34 (1.8) 6.98 (4.70-10.36) < 0.001
* Information was not available for all cases and controls. OR: odds ratio; 95%CI: confidence interval. In the intermediate hierarchical
level model, the variables of the intermediate and distal model that had been significant (p ≤ 0.05) were retained and included in the
adjustment of the third block of the proximal level.

Source: Authors.

for neonatal health. Such results confirm the ur- cedures, to follow the pregnancy periodically, to
gent need to improve access to qualified care for detect issues at an early stage and to treat health
pregnant women and to address the issue of the risk factors in time17 to prevent neonatal deaths18.
number of consultations. Moreover, prevention Regarding the number of pregnancies, an
and early detection of both maternal and fetal association was identified between NNM and
pathologies is fundamental to reduce life-threat- mothers who had already had two or more chil-
ening conditions in newborn children. dren in their obstetric history, corroborating the
It is widely known that starting prenatal care findings of a study carried out in Ethiopia11. On
early and performing follow-up appropriately the other hand, these findings differ from those
ensures more beneficial health outcomes for both found by a study performed in southeastern
the mother and the baby, since consultations Brazil, in which primiparous mothers showed
provide the opportunity to perform basic pro- a higher risk of NNM7. That divergence may be
 7

Ciência & Saúde Coletiva, 29(1):1-12, 2024

Table 4. Distribution of cases and controls according to variables of the proximal hierarchical level (health care).
Cuiabá, MT, Brazil, 2015-2018 (n = 2,793).
Total Cases Controls Crude OR
Variables* p-value
n (%) n (%) n (%) (95%CI)
Proximal level
Quarter in which prenatal care started
First 2,200 (80.7) 721 (80.0) 1,479 (81.0) 1.03 (0.63-1.70) 0.888
Second 452 (16.6) 156 (17.3) 296 (16.2) 1.12 (0.66-1.89) 0.671
Third 75 (2.7) 24 (2.7) 51 (2.8) 1.00
Number of prenatal consultations
<6 553 (19.9) 276 (29.9) 277 (14.9) 2.43 (2.01-2.94) < 0.001
≥6 2,225 (80.1) 647 (70.1) 1,578 (85.1) 1.00
Type of delivery
Vaginal 1,233 (44.2) 388 (41.7) 845 (45.38) 1.00
Cesarean 1,560 (55.8) 543 (58.3) 1,017 (54.62) 1.16 (0.99-1.36) 0.063
Induced labor
Yes 598 (21.6) 153 (16.6) 445 (24.1) 1.00
No 2,167 (78.4) 769 (83.4) 1,398 (75.9) 1.60 (1.30-1.96) < 0.001
Fetal presentation
Cephalic 2,601 (94.0) 804 (88.3) 1,797 (96.9) 1.00
Non-cephalic 165 (6.0) 107 (11.7) 58 (3.1) 4.12 (2.96 to 5.74) < 0.001
Type of hospital
Private 762 (27.4) 239 (26.2) 523 (28.1) 1.00
Private/affiliated with SUS 1,353 (48.8) 369 (40.4) 984 (52.8) 0.82 (0.67-0.99) 0.046
Public/university 319 (11.5) 156 (18.1) 154 (8.3) 2.34 (1.79-3.06) < 0.001
Philanthropic 341 (12.3) 140 (15.3) 201 (10.8) 1.52 (1.17-1.98) 0.002
* Not available for cases and controls. OR: odds ratio; 95%CI: confidence interval.

Source: Authors.

due to regional and cultural differences, socio- research that used a hierarchical neonatal near
economic inequalities, maternal preparation and miss model20 and to a third study that showed
adherence to prenatal care, different care systems, an association with the outcome among adoles-
quality of prenatal care, professional qualification cent mothers21. Thus, attention needs to be paid
and accessibility. to multiple pregnancies and timely and essential
Regarding the women’s obstetric history, care must be provided to this type of pregnancy,
mothers who had had no or one live birth in a as studies show that it is not only associated with
previous pregnancy were associated with the out- NNM, but also with neonatal death and a higher
come. It should be noted that the relationship of risk of prematurity and low birthweight22.
this variable with NNM has been little studied In the present study, the largest number of
so far. However, it is known that negative ma- births occurred in private hospitals associated
ternal and perinatal outcomes may be triggered with SUS, both among cases and controls. How-
by obstetric complications19. Thus, it is essential ever, births in public/university and philanthrop-
to pay more attention to women’s health before ic hospitals were the ones significantly associated
pregnancy and help them get ready by means of with the outcome. This suggests that the public
health prevention and promotion actions and by assisted by these hospitals shows specific char-
properly diagnose and treat issues that may arise. acteristics, such as social inequalities, which is
Double or more pregnancies were associated revealed by patient profile and care service type
with NNM by the present study, which matches offered to the population. That care service is
the findings of the “Nascer no Brasil” research specialized and a reference in assisting mothers
that also identified that kind of association in and babies at risk. It relies on assistance protocols
multiparous and nulliparous women6, in addi- and trained professionals to provide care based
tion to another study based on data of the same on scientific evidence.
 8
Modes PSSA et al.

Table 5. Hierarchical multiple logistic regression model of factors associated with cases of neonatal near miss in
Cuiabá, State of Mato Grosso, Brazil, 2015-2018 (n = 2,793).
Neonatal near miss
Variables
Adjust.
OR* 95%CI ** p-value***
Model 1 – Characteristics of the newborn
Sex
Male 1.13 0.95-1.35 0.153
Female 1.00 -
Year of birth of the newborn
2015 0.79 0.62-1.00 0.056
2016 1.11 0.87-1.41 0.405
2017 0.82 0.64-1.05 0.119
2018 1.00 -
Model 2 – Distal*
Education (years of study) (6 missings )
0 to 8 1.22 0.87-1.72 0.232
9 to 11 0.87 0.67-1.11 0.273
12 or more 1.00 -
Model 3 – Intermediate **
Marital status (6 missings)
Without a partner 0.99 0.82-1.20 0.958
With a partner 1.00 -
Number of pregnancies (33 missing )
0 1.00 -
1 1.30 0.89-1.89 0.168
2 1.63 1.01-2.63 0.044
3 or more 1.87 1.09-3.21 0.021
Number of children born alive (36 missings)
0 2.57 1.56-4.24 < 0.001
1 1.53 1.04-2.26 0.031
2 or more 1.00 -
Type of pregnancy (3 missings)
Single 1.00 -
Two or more 4.57 2.95 to 7.07 < 0.001
Model 4 – Proximal ***
Number of prenatal consultations (15 missings)
<6 2.20 1.77-2.72 < 0.001
≥6 1.00 -
Type of Hospital (18 missings)
Private 1.00 -
Private affiliated with SUS 0.98 0.75-1.26 0.871
Public / University 2.25 1.60-3.15 < 0.001
Philanthropic 1.62 1.16-2.26 0.005
Fetal presentation (27 missings)
Cephalic 1.00 -
Non-cephalic 2.71 1.87-3.94 < 0.001
Induced labor (27 missings)
No 1.47 1.18-1.84 < 0.001
Yes 1.00 -
OR: odds ratio; 95%CI: confidence interval; * model adjusted by sex and year of birth of newborns; ** model adjusted by the variables
of the distal block, sex and year of birth of newborns; *** model adjusted by the variables of the distal and intermediate block, sex
and year of birth of newborns.

Source: Authors.
 9

Ciência & Saúde Coletiva, 29(1):1-12, 2024

University hospitals are characterized by appropriately to induce labor, it may put the safe-
offering better obstetric and neonatal care, by ty of maternal and fetal health31 at risk and cause
qualified teams that follow protocols support- serious issues, such as uterine hyperstimulation
ed by scientific evidence and by using advanced and rupture, fetal distress, very painful uterine
medical technology7, which may explain the as- contractions, hyponatremia, fetal hypoxia and
sociation between the type of hospital and the acidemia, which contribute to an increase in the
identified NNM outcome in this study and which cesarean delivery rate32.
is therefore a protective factor against neonatal Since the concept of NNM is rather recent
mortality due to their care features. and still being discussed, the present research
In this sense, findings on hospital type may contributes to the field of health by assuming an
be useful for the surveillance of neonatal care in adapted definition of the pragmatic criteria that
institutions, even if evaluating different types of threaten life at birth (gestational age less than 32
hospitals is a complex matter. Surveillance can weeks, birthweight below 1,500g, 5-minute Ap-
be a monitoring tool for neonatal care in differ- gar score < 7, in addition to including congenital
ent institutions that support newborns at risk, as malformation as an NNM criterion and exclud-
long as only establishments of similar complexity ing mechanical ventilation, as it is not included
are compared among each other, considering in- in the Information Systems (IS) used.
stitutional profile and assisted population, in ad- Moreover, the study provided an expressive
dition to case severity and the different technol- assessment of the local context by analyzing the
ogies used to identify alert situations that require entire neonatal period and by using information
taking action23. provided by SIS, whose data is largely available.
Although most deliveries showed a cephal- In turn, the fact that some variables are incom-
ic presentation, both among cases and controls, plete may be considered a limitation of the pres-
which was similar to the findings of the two ent study. However, using a hierarchical model-
studies performed in Ethiopia8,11, non-cephalic ing strategy avoided weakening associations with
presentation was associated with NNM in both factors at distal levels by incorporating more
studies. In the present study, this type of presen- proximal ones in the model. Thus, the theoretical
tation had 2.71 chances of NNM compared to model based on literature improved our analysis
cephalic presentation. and helped integrate and interpret variables and
Regardless of delivery type, breech presenta- their respective statistical associations8.
tion results in a greater risk of gestational com- Based on these findings, we may claim that
plications24, stillbirth and neonatal death compared care quality has to be improved, including pre-
to the cephalic position25 and is associated with natal care, delivery and birth, correction of defi-
obstetric risk factors that increase linearly as the ciencies, planning and organization of improve-
gestational age is lower26, in addition to a higher ments, definition of priorities regarding actions
frequency of small-for-gestational-age births25-27, that have contributed most to “near deaths”, iden-
episiotomy27, and labor induction. Nevertheless, tification and tackling of issues at the intermedi-
studies show that the pelvic position does not ate and proximal hierarchical levels by highlight-
present a statistically significant difference in ing the significance of qualified prenatal care,
perinatal and maternal morbidity27-28 and as long considering obstetric history, paying thorough
as qualified assistance is provided, this can be a attention to multiple pregnancies and deliveries,
safe option for rigorously selected cases27. investing in training of professionals who assist
In the present study, most deliveries were newborns at risk at all levels of care, especially
noninduced, both among cases and controls, and those who work in public/university hospitals
were associated with NNM, which shows that not and are a reference in their field to avoid cases of
inducing labor favors the outcome. Pros and cons neonatal near miss. Further, these findings may
of inducing labor has been discussed worldwide contribute to the development of management
and involves several other issues, e.g., use of syn- strategies to reduce neonatal mortality and long-
thetic oxytocin, which is not recommended by term sequelae.
the World Health Organization (WHO) to accel-
erate childbirth29.
However, there are appropriate indications Conclusion
for its use for labor induction, e.g., in the first 24
hours of premature membrane rupture30. On the The results of this study point out that being a
other hand, when synthetic oxytocin is used in- mother who had had two or more pregnancies,
 10
Modes PSSA et al.

no or one live child at birth, multiple pregnan- delivery would favor care quality of this popula-
cies, less than six prenatal consultations, deliver- tion and help avoid this outcome.
ies in public/university and philanthropic hospi- The results of this investigation are intended
tals, non-cephalic presentation and non-induced to contribute to an initial reflection process on
labor was associated with NNM. Investing in factors associated with life-threatening condi-
improving care during the gestational period and tions in neonates, especially NNM, a topic that
still lacks epidemiological studies.

Collaborations

All authors contributed fundamentally to this

study.
 11

Ciência & Saúde Coletiva, 29(1):1-12, 2024

References

1. United Nations Children’s Fund (UNICEF), World 16. Victora CG, Huttly SR, Fuchs SC, Olinto MTA. The
Health Organization (WHO). Levels & trends in child role of conceptual frameworks in epidemiological
mortality: report 2019. Estimates developed by the Uni- analysis: a hierarchical approach. Int J Epidemiol 1997;
ted Nations Inter-agency Group for Child Mortality Es- 26:224-227.
timation (UN IGME). New York: UNICEF; 2019. 17. Brito LME, Mesquita KKCB, Melo JS, Santos TP. A
2. Surve S, Chauhan S, Kulkarni R. Neonatal near miss importância do pré-natal na saúde básica: uma revi-
review: tracking its conceptual evolution and way fo- são bibliográfica. Res Soc Develop 2021; 10:15.
rward. Curr Pediatr Res 2017; 21(2):264-271. 18. Saloio CA, Morais Neto OL, Gonçalves DA, Bessa
3. Pileggi C, Souza JP, Cecatti JG, Faúndes A. Neonatal HEM, Coelho Júnior JP, Afonso MSM, Carvalho SR.
near miss approach in the 2005 WHO Global Survey Magnitude e determinantes da mortalidade neonatal
Brazil. J Pediatr (Rio J) 2010; 86(1):21-26. e pós-neonatal em Goiânia, Goiás: um estudo de co-
4. Santos JP, Pileggi-Castro C, Camelo Jr JS, Silva AA, orte retrospectivo, 2012. Epidemiol Serv Saude 2020;
Duran P, Serruya SJ, Cecatti JG. Neonatal near miss: 29(5):e2020132.
a systematic review. BMC Pregnancy Childbirth 2015; 19. Moura BLA, Alencar GP, Silva ZPD, Almeida MF.
15:320. Internações por complicações obstétricas na gestação
5. Kale PL, Jorge MHPM, Laurenti R, Fonseca SC, Silva e desfechos maternos e perinatais, em uma coorte
KS. Critérios pragmáticos da definição de near miss de gestantes no Sistema Único de Saúde no Muni-
neonatal: um estudo comparativo. Rev Saude Publica cípio de São Paulo, Brasil. Cad Saude Publica 2018;
2017; 51:111. 34(1):e00188016.
6. Martinelli KG, Gama SGN, Almeida AHV, Pacheco 20. Pereira TG, Rocha DM, Fonseca VM, Moreira MEL,
VE, Neto ETS. Advanced maternal age and factors Gama SGN. Factors associated with neonatal near
associated with neonatal near miss in nulliparous miss in Brazil. Rev Saude Publica 2020; 54:123.
and multiparous women. Cad Saude Publica 2019; 21. Assis TSC, Martinelli KG, Gama SGN, Santos Neto
35(12):e00222218. ET. Associated factors of neonatal near miss among
7. Kale PL, Mello-Jorge MHP, Silva KS, Fonseca SC. newborns of adolescent mothers in Brazil. Rev Esc En-
Neonatal near miss and mortality: factors associated ferm USP 2022; 56:e20210359 .
with life-threatening conditions in newborns at six 22. Demitto MO, Gravena AAF, Dell’Agnolo CM, Antu-
public maternity hospitals in Southeast Brazil. Cad nes MB, Pelloso SM. Gestação de alto risco e fatores
Saude Publica 2017; 33(4):e00179115. associados ao óbito neonatal. Rev Esc Enferm USP
8. Yohannes E, Assefa N, Dessie Y. Determinants of ne- 2017; 5:e03208.
onatal near miss among neonates admitted to Ambo 23. França, KEX, Vilela BR, Frias PG, Chaves MA, Sari-
University Referral Hospital and Ambo General Hos- nho. Near miss neonatal em hospitais de referência
pital, Ethiopia, 2019. J Preg Child Health 2020; 7:1. para gestação e parto de alto risco: estudo transversal.
9. Shroff BD, Ninama NH. A call for eminence obstetrics Cad Saude Publica 2021; 37(6):e00196220.
care by way of “Neonatal Near Miss” events (NNM): 24. Fonseca A, Nogueira-Silva C, Silva IS, Casal Es, Pinto
a hospital-based case-control study. J Obstet Gynaecol L, Clode N. Vaginal breech delivery: guidelines/nor-
India 2019; 69(1):50-55. mas de orientação clínica. Acta Obstet Ginecol Port
10. Lima THB, Katz L, Kassar SB, Amorim MM. Neonatal 2021; 15(1):75-83.
near miss determinants at a maternity hospital for hi- 25. Bjellmo S, Andersen GL, Martinussen MP, Romunds-
gh-risk pregnancy in Northeastern Brazil: a prospec- tad PR, Hjelle S, Moster D, Vik T. Is vaginal breech
tive study. BMC Pregnancy Childbirth. 2018; 18:401. delivery associated with higher risk for perinatal dea-
11. Mersha A, Bante A, Shibiru S. Factors associated with th and cerebral palsy compared with vaginal cephalic
neonatal near-miss in selected hospitals of Gamo and birth? Registry-based cohort study in Norway. BMJ
Gofa zones, southern Ethiopia: nested case-control Open 2017; 7(4):e014979.
study. BMC Pregnancy Childbirth 2019; 19:516. 26. Toijonen AE, Heinonen ST, Gissler MVM, Macharey
12. Silva GA, Rosa KA, Saguier ESF, Henning E, Mucha F, G. A comparison of risk factors for breech presenta-
Franco SC. Estudo de base populacional sobre a pre- tion in preterm and term labor: a nationwide, popu-
valência de near miss neonatal em município do Sul lation-based case-control study. Arch Gynecol Obstet
do Brasil: prevalência e fatores associados. Rev Bras 2020; 301(2):393-403.
Saude Matern Infant 2017; 17(1):169-177. 27. Valente MP, Afonso MC, Clode N. O parto pélvico
13. Almeida SDM, Barros MBA. Atenção à saúde e mor- vaginal ainda é uma opção segura? Rev Bras Gine-
talidade neonatal: estudo caso-controle realizado em col Obstet 2020; 42(11):712-716.
Campinas, SP. Rev Bras Epidemiol 2004; 7(1):22-35. 28. Gaillard T, Girault A, Alexander S, Goffinet F, Le Ray
14. Oliveira GP, Bierrenbach ALS, Camargo Júnior KR, C. Is induction of labor a reasonable option for bre-
Coeli CM, Pinheiro RS. Accuracy of probabilistic and ech presentation? Acta Obstet Gynecol Scand 2019;
deterministic record linkage: the case of tuberculosis. 98(7):885-893.
Rev Saude Publica 2016; 50:49. 29. World Health Organization (WHO). Evaluating the
15. Brustulin R, Marson PG. Inclusão de etapa de pós quality of care for severe pregnancy complications: the
-processamento determinístico para o aumento de who near-miss approach for maternal health. Geneva:
performance do relacionamento (linkage) probabilís- WHO; 2011.
tico. Cad Saude Publica 2018; 34(6):e00088117.
 12
Modes PSSA et al.

30. Ministério da Saúde (MS). Secretaria de Ciência,

Tecnologia e Insumos Estratégicos. Departamento
de Gestão e Incorporação de Tecnologias em Saúde.
Diretrizes nacionais de assistência ao parto normal:
versão resumida. Brasília: MS; 2017.
31. Schincaglia CY, Santos GC, Ribeiro JA, Figueiredo RY,
Menezes S, Maia JS, Maia LFS. As consequências do
uso de ocitócitos durante o parto. Rev Recien 2017;
7(19):75-82.
32. Santos KLA, Farias CRBL, Cavalcante JS, Santos EA,
Silva JM, Duarte APRSD. Ocitocina sintética no tra-
balho de parto induzido e suas repercussões materno-
fetais. Diversitas J 2020; 5(3):1787-1804.

Article submitted 29/10/2022

Approved 17/04/2023
Final version submitted 19/04/2023

Chief editors: Romeu Gomes, Antônio Augusto Moura da

Silva

CC BY This is an Open Access article distributed under the terms of the Creative Commons Attribution License