Do depressive symptoms among pregnant women assisted in Primary Health Care services increase the risk of prematurity and low birth weight?

Objective: to investigate associations between depressive symptoms during pregnancy, low birth weight, and prematurity among women with low-risk pregnancies assisted in public Primary Health Care services. Method: prospective cohort with 193 pregnant women, using the Edinburgh Postnatal Depression Scale, telephone interviews, and medical records available in the health services. Associations of interest were obtained using the Cox regression model. Results: the participants were aged 24.9 years old (median) and had 11 years of schooling (median); 82.4% lived with their partners, and gestational age at the birth was 39 weeks (median). Twenty-five percent of the participants scored ≥13 on the Edinburgh scale. Depressive symptoms did not appear associated with low birth weight (RR=2.06; CI95%=0.56-7.61) or prematurity (RR=0.86; CI95%=0.24-3.09) in the adjusted analysis. However, premature labor increased the risk of low birth weight (RR=4.81; CI95%=1.01-23.0) and prematurity (RR=7.70; CI95%=2.50-23.7). Additionally, each week added to gestational age decreased the risk of low birth weight (RR=0.76; CI95%=0.61-0.95). Conclusion: the presence of depressive symptoms among women with low-risk pregnancies was not associated with low birth weight or prematurity.


Introduction
Depression is considered a public health problem due to its severity, recurrence, and negative impact on health (1) . According to the World Health Organization (WHO), more than 254 million people of all ages are affected by depression worldwide. Hence, depression is among the primary causes of incapacity and significantly contributes to the global load of illness, especially among women (2) . Depression is a mental disorder characterized by depressive mood, sleep disorders, lack of interest in daily tasks, difficulty concentrating, guilt, lack of or excessive appetite, and occasionally thoughts and suicidal ideation (1,(3)(4) .
Depression is highly prevalent during pregnancy, affecting one in every five women worldwide (1) . The various physical, hormonal, emotional, and psychosocial transformations that occur during pregnancy directly influences these women's mental health, explaining its prevalence (4)(5) .
The consequences of depression during pregnancy in the short and long terms have been investigated, emphasizing the harm caused to the fetus's health, including low birth weight and prematurity. Despite the biological plausibility of this association, studies present conflicting results (6)(7)(8)(9) .
A systematic review reports an association between prenatal depression and premature birth in less than ¼ of the 50 studies included in the study, and approximately half (53%) of the studies addressing low birth weight found an association with prenatal depression (6) . A recent literature review included seven studies in which increased risk of prematurity was found among women with depression, while another five studies did not find such an association (9) .
A prospective cohort conducted in 2018 in Kenya, not included in the previous reviews, found an association between depression and premature labor: women with depression during pregnancy were approximately four times more likely to have premature babies, in comparison to women without depressive symptoms (7) . The cutoff point adopted in the study for the Edinburgh Postnatal Depression Scale (EPDS) was equal to or higher than 10 points. In the same year, a Vietnamese cohort study using the same scale and cutoff point found a risk three times greater for prematurity and a twice greater risk for low birth weight among women with depression (8) .
In Brazil, the relationship between depression during pregnancy, low birth weight, and prematurity is seldom addressed. A cohort study addressing women with low-risk pregnancies was conducted in the capital of São Paulo and did not find any association between depression and low birth weight or prematurity (10) . Another cohort study, conducted in a city located in the south of Brazil, addressed women exclusively assisted by the Unified Health System (SUS) during prenatal care and labor and verified that mothers experiencing depression during pregnancy were four times more likely to have an infant with low birth weight (11) .
The previous discussion shows no consensus in the international or Brazilian literature regarding the relationship between gestational depression, low birth weight, and prematurity, which justifies this study. This study's primary objective was to investigate the potential association between depressive symptoms during pregnancy, low birth weight, and prematurity among women with low-risk pregnancies assisted by Primary Health Care services. The secondary objective was to identify the factors associated with both outcomes.

Study design
This prospective cohort study was conducted in the mid-south region of the state of São Paulo, Brazil (12) .

Study setting
The study was conducted in Botucatu, SP, Brazil, a USF have a team composed of family physicians, nurses, dentists, nursing technicians, dental assistants, and community health agents located in the city's periphery, providing care to 37.0% of the city's population (15) .

Study participants
The pregnant women were recruited in all the PHC units in the city between May and December 2018. The following inclusion criteria were used: pregnant women at any gestational age, aged 18+ years old, and literate.
Multiple pregnancies were excluded due to the risk of low birth weight and prematurity.
This study addressed a convenience sample composed of 193 pregnant women, who were monitored until child delivery and had all data of interest available.

Instruments used in data collection
The EPDS, a scale initially developed to investigate postpartum depression (5) , was used to collect data concerning depressive symptoms during pregnancy. It was translated into and validated in many countries, including Brazil (16) , and was indicated and validated to be used during pregnancy. A systematic review, addressing instruments designed to assess depression among pregnant women and identify the most appropriate instruments to apply in prenatal care services, reports that the EPDS was the most accurate scale (area under the curve ROC=0.965, sensitivity=0.80, and specificity=0.81) (17) . Its advantages include the fact that it is a self-reported, simple, and easy-to-apply instrument, simple to be interpreted, and any professional can apply (4,16) .
The EPDS is a questionnaire composed of ten topics with four options of answers, ranging from zero to three, according to the absence or presence and intensity of symptoms (18)(19)(20) . The questions address the presence of depressive or dysphoric mood, guilt, sleep disorders, loss of pleasure, decreased performance, and death and suicide ideation (4)(5) . The final score ranges from 0 to 30 points; higher scores represent a more severe condition (4) .
Similar to other studies, a score equal to or higher than 13 points was the cutoff point used to indicate the risk of depression (21)(22) .
Three other forms specifically designed for this study were used. The first collected the participants' identification information; the second addressed sociodemographic data and information concerning pregnancy planning and acceptance, availability of social support, and the participants' lifestyle; and the third form addressed data concerning the pregnancy, labor, and childbirth, which were available in the medical records available in the public maternal hospitals located in Botucatu.

Study variables
The exposure variable was a score indicating depressive symptoms.
Two outcomes were considered: low birth weight and prematurity, and the covariates included maternal sociodemographic factors and complications in the current pregnancy. A synthesis of the variables is presented in Figure 1. Strategies were adopted to avoid a follow-up loss such as asking the participants' different telephone numbers and those of their partners or relatives; the telephone interview was scheduled on the day the EPDS was applied according to the participants' convenient day and time, and contact was attempted at least six times before considering it a loss.

Data analysis
Association between depressive symptoms during

Ethical aspects
The study was approved by the Research Ethics

Board at the Medical School of Botucatu, under protocol
No. 2,641,633. The participants signed free and informed consent forms, and whenever depressive symptoms were identified (i.e., equal to or higher than 13), the participants were referred to medical assessment in the health units where they received prenatal care.

Discussion
Contrary to the expected, no association was found in this study between depressive symptoms during pregnancy and low birth weight or prematurity. On the other hand, a large number of pregnant women presented a positive score for depression. This result was similar to that reported by a Brazilian study conducted in another city, which adopted the EPDS and considered a validity is that the gestational trimester in which mothers were at the time the EPDS was applied was not associated with the final score or the two outcomes addressed here.
Therefore, the fact that the pregnant women were not all assessed in the same stage of pregnancy did not affect the results.
The depression rates reported by international studies among pregnant women vary among countries and within the same country. Using the same scale adopted here and the same cutoff point, studies conducted in China report 28.5% (23) , 33.8% in Tanzania (24) , and 7.0% in Australia (25) . Different rates were found in different places within Ethiopia 11.8% (26) , 21.5% (1) , and 24.9% (27) so that only the highest rate found in Ethiopian studies was similar to the one reported here. Depressive symptoms identified with the EPDS among pregnant women in Nigeria were similar to those found in this study, though the cutoff point used was 12, limiting comparisons between results (3) .
Note that the main result found in this study was no association between depressive symptoms during pregnancy and low birth weight or prematurity. Therefore, the hypothesis that depressive symptoms during pregnancy would increase the risk of these two adverse outcomes was not confirmed. This result is in line with some previous studies (6)(7)(8) , except for one (10) , so that the question remains, especially in Brazil. and are more susceptible to consume poor quality food (32) , alcohol, and smoking (3,(33)(34) . A literature review addressing the repercussion of maternal depression, including studies conducted in low-and middle-income countries, reports a greater risk of psychological disorders among infants, children, and adolescents; deficient growth and postnatal child development; early weaning; and childhood illnesses, such as diarrhea and other infectious diseases (35) .
This study also identified that premature labor increased by approximately five and eight times the risk of low birth weight and prematurity, respectively. On the other hand, each week added to gestational age decreased the risk of low birth weight by 24%. These relations are well known and corroborate the validity of the results reported here.
Premature labor is a significant complication that determines neonatal morbidity and mortality due to the progression to premature childbirth, with consequences in the long term including neurological, cognitive, respiratory, cardiovascular, and psychosocial disorders (36) . Studies using varied designs, addressing different populations, and conducted in different countries consistently report an association between premature labor and low birth weight (37)(38)(39)(40)(41)(42) .
The large portion of women with low-risk pregnancies in this study presenting depressive symptoms indicates that actions should be implemented in health services to identify, manage, and monitor prenatal depression.
Considering the vital work of PHC nurses in prenatal care, these professionals are expected to lead actions intended to incorporate routine screening for depression among pregnant women.

Conclusion
Many women with depressive symptoms were identified in this study. Slightly more than ¼ of the participants obtained scores equal to or higher than 13 in the EPDS. However, no association was found between depressive symptoms during pregnancy and low birth weight or prematurity. On the other hand, premature labor was confirmed as an independent risk factor influencing low birth weight and prematurity, while increased gestational age was confirmed as a protection factor against low birth weight.