Alcohol intake during pregnancy among parturients in southern Brazil

Objectives: to assess alcohol intake prevalence and identify associated factors among pregnant women in the municipality of Rio Grande, RS, Brazil. Methods: this was a cross-sectional study which included all parturient women residing in the municipality who gave birth in 2013. Two outcomes were characterized: alcohol intake during pregnancy and excessive alcohol intake during pregnancy. In the analysis, proportions were tested using the Chi-square test, whilst Poisson regression was used in the multivariate analysis. Results: 9.4% (CI95%= 8.2-10.5) of the 2,685 parturient women studied reported having consumed alcohol during pregnancy, with beer being most common beverage. Following adjustment, the factors associated with alcohol intake in pregnancy were: age ≥30 years, brown skin colour, living without a partner, low schooling, tobacco smoking and illicit drug use, having had more children and late onset of prenatal care. Excessive alcohol intake was found in 12.7% (CI95%= 8.6-16.9) of those who reported drinking during pregnancy and the factors associated with this practice were age ≥30 years, lower schooling and illicit drug use. Conclusions: this study found high alcohol intake during the gestation period and identified women more susceptible to this practice. There is an evident need for health professionals to work on preventing and handling alcohol intake among pregnant women in the municipality


Introduction
Alcohol intake is a worldwide habit involving diverse cultures.It is considered to be a licit drug and is easy to access. 1 There are various definitions of alcohol intake patterns, but all of them take into consideration the amount consumed, intake frequency and its occurrence according to the person's sex. 1,2oderate alcohol intake (one drink a day for women and two drinks a day for men) does not represent a health problem for the general population.In the case of pregnant women, however, safe levels of exposure to alcohol have not been set, since even moderate intake can cause harm to the fetus.As such, any pattern of alcohol intake during pregnancy is considered to be a risk, and abstinence is recommended for pregnant women and women who intend to become pregnant. 3,4arm arising from fetal exposure to alcohol varies depending on the amount consumed, the pregnant mother's nutritional status and the metabolic capacity of mother and fetus, in addition to stage of pregnancy. 5In the first trimester of pregnancy there is a greater risk of physical abnormalities, whilst in the second trimester there is anincreased risk of miscarriage.In the third trimester, fetal growth may be restrict, particularlyhead circumference. 6One of the complications resulting from alcohol intake during pregnancy is fetal alcohol syndrome (FAS), which is characterized by craniofacial abnormalities, intra uterine growth retardation and central nervous system dysfunctions. 7espite potential risks to the health of fetuses and pregnant women, few studies assess alcohol intake prevalence during pregnancy.Furthermore, the use of different definitions and instruments to assess this outcome hinders result comparability.Studies assessing alcoholic drink intake during pregnancy report prevalence varying between 10.8% and 29.0%, [8][9][10][11] with most frequent intake among pregnant women aged over 30 years, 8,11 tobacco smokers, 10 illicit drug users, 10,12 women with depression13 and those having fewer prenatal care attendances. 11n view of the risks to which fetuses and pregnant women are exposed,owing to alcohol intake during pregnancy, as well as the scarcity of studies on this subject in Brazil, the purpose of this paper is to assess alcohol intake prevalence and associated factors during pregnancy, as well as excessive intake and types of drink consumed during pregnancy, among parturient women in the municipality of Rio Grande-RS in the year 2013.

Methods
This paper refers to the 2013 Perinatal Study, a cross-sectional population-based study conducted every three years in the city of Rio Grande.This is a port city located in the far south of the state of Rio Grande do Sul.In 2013 the city had a population of approximately 205,000 inhabitants.
Given that the 2013 Perinatal Study included all pregnant women resident in the municipality, only the sample size calculations performed a posteriori in order to examine possible factors associated with alcohol intake in pregnancy are presented here.The following parameters were used: minimum statistical power of 80%, 95% confidence level, unexposed: exposed ratio of between 1:5 and 9:1, 5% outcome prevalence among the unexposed, and a prevalence ratio of 2.0.In addition, sample size was increased by 15% to control for confounding.Based on these parameters, this study should have comprised at least 2,680 pregnant women.It was possible to achieve this number since the Perinatal Study involved 2,685 parturients.
Data collection took place between January 1 st and December 31 st 2013 and included all parturient women resident in the municipality (both urban and rural areas) who gave birth to their children in either of the city's two maternity hospitals, namely the Dr. Miguel Riet Corrêa Jr. University Hospital, part of the Federal University of Rio Grande (UFRG), which provides services only to Unified Health System (Sistema Único de Saúde-SUS) users, and the Rio Grande Santa Casa de Misericórdia, which provides services to SUS, private health insurance and private patients.Parturient women were interviewed only once.Women who were excluded were not resident in Rio Grande, as well as those whose babies were born prior to the twentieth week of pregnancy or had birth weight below 500 grams, as this characterizes miscarriage.
A pre-coded questionnaire with mostly closedended questions was administered with mothers up to 48 hours after childbirth.The interviewers underwent 40-hour training which finished with a pilot study with parturient women admitted to hospital in the month prior to the start of data collection.
Two outcomes were characterized for this study.The first outcome, alcohol intake during pregnancy, was investigated by asking the question "Did you use to drink alcohol when you were pregnant?"Interviewees who replied positively were considered to be cases and were then asked about the kind of drink (wine, beer or distilled drinks) they had drunk in each trimester of pregnancy, as well as the type of drinking vessel (100 mL glass/wine glass/goblet, ordinary beaker, tin, small bottle,and 600-720 mL bottle) and the number of measures of drink per week.
The second outcome, excessive intake, was established as follows: first of all a standard measure of alcohol was defined according to the kind of vessel and type of drink; a standard measure was therefore equivalent to 330 mL of beer, 100 mL of wine and 30 mL of distilled drink; in this way the drinking vessels reported in answer to the questionnaire (ordinary 200 mL beaker, 100mL glass/wine glass/goblet, 350 mL tin, small 300 mLbottle and 600 mL bottle) were divided by the standard measure equivalent to each type of drink [(beer (330 mL), wine (100 mL) and distilled drinks (30 mL)]; the number of measures equivalent to each vessel was then multiplied by the number of days per week when alcohol was drunk and by the number of vessels drunk daily; the result, i.e. the total number of measures drunk in a week, was divided by seven to estimate the number of standard measures drunk per day in each trimester of pregnancy.Parturient womenwho reported drinking more than one standard dose per day in any trimester of pregnancy were considered to be excessive alcohol intake cases. 2 The independent variables were maternal age (up to 19 years, 20-29 yearsand 30 years or more), observed skin colour (white, brown, black, other), whether living with a partner, schooling in years (0-8 years, 9-11 years and 12 years or more), tobacco smoking (never; former smoker, smoker), illicit drug use during pregnancy (cocaine, marijuana, crack), self-reported depression during pregnancy, number of children given birth to (alive or dead), gestational age when starting prenatal care (in trimesters) and, having had a miscarriage or an abortionat any time in the past.
Every day at the two maternity hospitals, the interviewers checked the birth registry book to check the mothers' place of abode.When it was confirmed that a parturient woman lived in the urban or rural area of Rio Grande, the interviewers explained the study to her and invited her to take part.In case of agreeing, a Free and Informed Consent Form was signed by them.A copy of the form was kept by each mother.
Analysis was conducted using STATA 14. 14 Firstly, the proportions of all variable categories of interest were calculated.The prevalence of each outcome was then calculated according to the independent variables, followed by the calculation of the crude prevalence ratios and prevalence ratios adjusted using Poison regression with robust variance.The Wald linear trend statistical test was used for ordinal exposures and the Wald heterogeneity test for dichotomous/polytomous exposures.
Multivariate analyses with backward selection were performed according to a hierarchical analysis model comprised of three levels: the first level contained socio-demographic variables (age, skin colour, marital status and schooling); the second level contained behavioural and morbidity variables (tobacco smoking, illicit drug use and self-reported depression); the third level contained variables relating to reproductive history and antenatal care (number of children given birth to, miscarriage/abortion and trimester in which antenatal care was begun).Variables having a pvalue ≤0.20 were kept in the model.Variables with a p-value ≤0.05 were considered to have significant association.
Prevalence of intake of different kinds of alcoholic drink (wine, beer, distilled drinks) was also calculated according to the trimester of pregnancy in which the interviewees stated having drunk alcohol.The Chi-square linear trend test was performed to assess whether there had been significant change in the proportions of pregnant women drinking alcohol during the three trimesters of pregnancy for the different types of drink assessed.
Quality control was performed by means of 7% partial repetition of interviews with mothers by telephone or household visits.In the case of variables assessed that included, for instance, "age", "skin colour", "age when antenatal care begun", "number of antenatal sessions attended", the Kappa coefficient varied from 0.63 to 0.78, showing moderate to satisfactory agreement. The

Results
According to the Live Birth Information System, in 2013 there were 2,713 parturient women eligible for inclusion in the 2013 Perinatal Study. 15Of this total, 2,685 women were interviewed, including 32 stillborn baby cases.
Table 1 shows that around 17% of parturients were adolescents and that over 50% were aged 20-29.The majority had white skin colour, lived with a partner and had up to 11 years of schooling.At least one in six (16%) smoked tobacco.Almost 2% said they had used some sort of illicit drug during pregnancy and 10% reported depression during pregnancy.Almost half (47.3%) were primiparous, eight in ten (78.6%) began prenatal care in the first trimester of pregnancy and 15.3% reported miscarriage or abortion.Overall prevalence of alcoholic drink intake during pregnancy was 9.4% (CI95%= 8.2-10.5).
Adjusted analysis showed that among all the variables included in the model, only the miscarriage/abortion variable lost statistical  of drink with a reduced intake trend, varying between 84.1% and 70.1% in the first and third trimesters, respectively.Around 15% of parturients reporting alcohol intake during pregnancy drank wine and less than 10% drank distilled drinks.There was no significant reduction in the proportion of pregnant women who drank wine and/or distilled drinks between the trimesters of pregnancy (Figure 1).3).
Beer was the most commonly consumed beverage during pregnancy.This was the only kind

Alcohol intake during pregnancy
Studies conducted in the United States and Canadashow prevalence rates of alcohol intake during pregnancy similar to those found in Rio Grande,i.e. 10.4% and 10.8%, respectively. 17,18lthough socio-economic and methodological differences prevent more in-depth comparison, it is evident that despite abstinence from alcohol during pregnancy being recommended, at least one in tenpregnant women uses this substance at some time during pregnancy.It is therefore a frequent problem occurring in countries/regions with great differences in relation to levels of economic development and population and cultural differences.Women aged 30 or over are more likely to drink alcohol during pregnancy and also drink excessively.This finding may arise from the fact that older women are more subject to behaviours that become established over the course of life, such as, for instance, tobacco smoking.Moreover, these women have higherparity, which may give them a false sense of security about the outcome of their pregnancy, whereby they take on risk behaviours. 10,11[21][22]

Discussion
This study found high alcohol intake prevalence during pregnancy.It also showed thatbeing older, having brown skin colour, living without a partner, having less schooling, using illicit drugs, having self-reported depression during pregnancy, higher parity and beginning prenatal care late are all associated with the outcome.Beer was the most frequently consumed beverage, with a tendency of intake reducing between the first and third trimesters of pregnancy.Among mothers who drank alcohol during pregnancy, those who were older, had less schooling and used illicit drugs were found to have a greater propensity to consuming larger quantities of alcohol.
Alcohol intake prevalence during pregnancy in Rio Grande is similar to that reported in one study conducted in a public maternity hospital in Rio de Janeiro city. 12Notwithstanding, it is lower than the prevalence reported by another study also conducted in public maternity hospitals in Rio de Janeiro city. 16hese two studies only included Unified Health System users.As such, the parturients assessed may not represent the entire population of pregnant women in Rio de Janeiro city.In addition, different instruments and different definitions of alcohol intake may also explain the variability in the findings.to prenatal care.
In addition to maternal age, schooling and illicit drug use were also variables strongly associated with excessive alcohol use among parturients reporting alcohol intake during pregnancy.These results show that women with these characteristics are even more susceptible to consuming alcoholic drinks and, therefore, need greater care from health professionals and health services.
This study has some limitations.One of them relates to self-reported alcohol intake.It is possible that prevalence reported in this study may be underestimated owing to the fact that some interviewees may have answered the questionnaire negatively for fear of being criticized.Moreover, a standardized/validated instrument was not used to assess alcohol consumption.In addition, there may be memory recall error in relation to the number of drinks consumed in the different stages of pregnancy assessed.With regard to the analysis of factors associated with excessive intake, the "tobacco smoking" and "parity" variables probably lost statistical significance in their association with the outcome owing, fortunately, to the low number of pregnant women reporting this practice.
This study provides a description of parturient women in Rio Grande who consumed alcoholic drink during pregnancy, showing which of them are more vulnerable to a risk factor that is extremely harmful to maternal and fetal health.It is important to stress the need for early identification by prenatal care services of women who consume alcohol during pregnancy.Health professionals also need to be trained in order to be able to deal with this problem which often goes unnoticed during routine appointments.In addition, campaigns are needed to guide women of child-bearing age and who intend to become pregnant about the importance of abstaining from alcohol intake during pregnancy.
Several studies show that living without a partner is a risk factor for alcohol intake during pregnancy. 12,16,20,21This is due to the fact that pregnancy among single women is frequently related to factors associated with drinking alcohol, including low socio-economic status, stress because of being the main/only family breadwinners and the occurrence of unplanned or unwanted pregnancy. 12,13,23iverse studies have shown that alcohol use is more prevalent among women with low levels of schooling. 10,20,23,24This may be the result of less access to information and greater propensity to behaviours associated with alcohol consumption, including tobacco smoking and illicit drug use.It has also been found that the prevalence of alcohol intake among tobacco smokers is twice that found among non-smokers 25,26 and up to five times higher among illicit drug users, thus providing evidence that these are behaviours associated with alcoholic drink intake. 19,21ccording to the literature, self-reported depression increases the likelihood of alcohol use as mental morbidities are commonly associated with the use of substances such as alcohol, cigarettes and other illicit drugs.The possibility of reverse causality exists in this study, given that self-reported depression is susceptible to the outcome studied here.Notwithstanding, screening psychiatric symptoms during pregnancy can help to identify alcohol use, as well as the use of tobacco and illicit drugs. 13he pregnant women who began prenatal care late are, probably, women from settings with poor living conditions and difficult access to health services, and this may also favour alcohol intake.In addition, women who use alcohol have characteristics and behaviours that may lead them to use health services less, thus resulting in a late start

Figure 1
Figure 1Prevalence of consumption of different alcoholic drinks according to trimester of pregnancy among pregnant women in Rio Grande-RS, who consumed alcohol during pregnancy; 2013 (N=251).

Table 1
Characteristics of pregnant women and prevalence of alcohol intake during pregnancy among women resident in the municipality of Rio Grande-RS, 2013 (N=2685).
+ N= 2613, 71 pregnant women had no prenatal care, 1 pregnant woman with no information on trimester in which prenatal care began.

Table 2
Crude and adjusted analysis of factors associated with alcohol intake during pregnancyin women resident in the municipality of de Rio Grande -RS, 2013 (N=2685).
+ N= 2613, 71 pregnant women had no prenatal care, 1 pregnant woman with no information on trimester in which prenatal care began.++ Prevalence ratio; * Wald linear trend test; ** Wald heterogeneity test.

Table 3
Prevalence, crude and adjusted analysis of factors associated with excessive alcohol intake (>1 standard measure per day) in any trimester of pregnancy among pregnant women in Rio Grande, RS, 2013.(N=251).