Services on Demand
Print version ISSN 0034-8910
Rev. Saúde Pública vol.38 n.1 São Paulo Feb. 2004
Ana Maria Krusser Zambonato; Ricardo Tavares Pinheiro; Bernardo Lessa Horta; Elaine Tomasi
Curso de Mestrado em Saúde e Comportamento da Universidade Católica de Pelotas. Pelotas, RS, Brasil
OBJECTIVE: To assess the risk factors for small-for-gestational-age
METHODS: All live births occurring in the city of Pelotas, Brazil, between October and December 1993 were identified and mothers interviewed soon after delivery. Birthweight was recorded by the maternity staff. Gestational age was obtained from the mothers recall of their date of last menstrual period. SGA was defined as a birthweight below the 10th percentile for gestational age and sex, according to the reference developed by Williams. Chi-square test and logistic regression were used in the crude and multivariate analysis, respectively.
RESULTS: In all, 1082 births were identified. The prevalence of SGA was 13.1%. Even after adjusting for possible confounding variables, the odds ratio for SGA among those infants whose family income was <1 minimum wage was 8.81 (95% CI, 1.12-69.46) times higher than among those for infants with a family income >10 minimum wages. Low quality antenatal care was associated with an odds ratio of 3.28 (95% CI, 1.09 - 9.91) for SGA. Short maternal stature and maternal smoking during pregnancy were also associated with SGA births.
CONCLUSIONS: Low socioeconomic status, maternal smoking during pregnancy, maternal height and low quality antenatal care are the main risk factors for SGA births.
Keywords: Small-for-gestational age. Birth weight.Gestational age. Fetal growth retardation. Risk factors. Prevalence. Socioeconomic factors.
Birth weight is a strong predictive factor for perinatal morbidity and mortality. Low birth weight infants, that is, those born with <2,500 g, present a higher mortality rate in the first weeks of life than those born with >2,500 g. Low birth weight may be due to prematurity, retarded intrauterine growth or a combination of both factors. Each of these factors present different etiologies and consequences.8,9
The majority of definitions consider as small-for-gestational-age (SGA) those infants whose birth weights are below a specific cut-off point for sex and gestational age. Other authors consider as SGA those infants born at term, but whose weight at birth is <2,500 g. Being SGA is utilized as an indicator of retarded intrauterine growth, but these terms are not synonyms for some infants considered SGA may simply be those located at the lower end of the normal curve of fetal growth distribution, for example, sons and daughters of short mothers. Infants exposed to agents that restrict intrauterine growth, such as cigarette smoking during gestation, may have a lower birth weight than they would have had if they hadnt been exposed to this agent, but not sufficient to be considered SGA. It is impossible to determine if an infant who is SGA did or didnt suffer from intrauterine growth retardation based on a simple evaluation of fetal size and gestational age.1 On the other hand, the larger the prevalence of SGA infants, the greater the proportion of intrauterine growth retardation. At present, the World Health Organization15 recommends that those infants with birth weights below the 10th percentile for gestational age and sex, according to the reference developed by Williams14 should be considered SGA.
Infants born SGA present an increased morbidity and mortality rate for the first years of life.2,4,6,10 Furthermore, as suggested in the studies edited by Barker et al,3 these infants present a higher risk for developing chronic illnesses as adults, such as diabetes, hypertension and coronary disease. Consequently, preventing the birth of infants who are small-for-gestational-age will have short and long run effects.
As to the etiology of SGA, the main risk factors are: black race; low caloric intake or low weight gain during gestation, low pre-gestational weight, maternal smoking, and short maternal stature.10
The objective of this study is to identify the main risk factors for small-for gestational-age births and collaborate towards the implementation of measures that result in a reduction in the prevalence of this pathology.
A cross-sectional study in which women that had given birth between October and December, 1996, in the five maternities of the city of Pelotas in Rio Grande do Sul State, Brazil were invited to participate. Initially, the studys objectives were explained to mothers who were then asked to give their informed consent. Those who participated in the study, a total of 1,082 mothers, were interviewed. The standardized questionnaire utilized for the interview included questions concerning socioeconomic as well as demographic conditions, reproductive history, maternal smoking during gestation, antenatal care and birth assistance. The maternities were visited daily in the mornings and afternoons by interviewers. The latter were medical students, previously trained in techniques for applying questionnaires.
Birth weight was measured by assistants working in the nursery wards. Gestational age was calculated by subtracting the date of the last menstruation from the date of birth. Those children whose birth weight, according to gestational age and sex, was below the 10th percentile, according to the reference developed by Williams,14 were considered SGA.
In order to quantify antenatal care, an indicator developed by Kessner and adapted by Takeda12 which combines the number of visits to the antenatal clinic and the period in which antenatal care was initiated was utilized. When the mother attended 5 or more antenatal visits and the first visit occurred up to the fourth month of gestation, antenatal care was considered adequate. Antenatal care was classified as inadequate when mothers attended antenatal clinics less than four times and initiated care after the seventh month of gestation. Antenatal care was classified as intermediate in all the remaining cases.
In addition to the Kessner index, quality of care was measured according to a score that was created based on procedures performed during the antenatal visit. Mothers were asked were asked whether the doctor had weighed her, measured her blood pressure, auscultated fetal heartbeats, measured uterine fundal height. Furthermore, mothers were asked whether in, any of the visits, the physician examined the breasts, encouraged breastfeeding, prescribed iron and tetanus toxoid, performed a gynecological examination and whether the latter included a Pap smear. Each item performed was given a score of 1. Consequently, scores varied from 0 to 8, indicating, respectively, antenatal care in which none of the above items were performed or antenatal care in which all of the above items were performed according to mothers.
As to maternal smoking, in order to ensure the quality of the information being collected, mothers were asked about their smoking habits during pregnancy. Those who stated they smoked during pregnancy were asked about their smoking habits during each trimester as well as the number of cigarettes smoked daily.
A data bank was created utilizing the Epi Info program. The Chi-Square test was performed on bivariate analysis with the objective of describing the occurrence of SGA, taking the independent variables and possible confounding factors into consideration.
In the logistic regression, the SPSS Program was utilized to analyze data hierarchically: socioeconomic and demographic variables were included in the first level; in the second level, anthropometric variables (height and corporal mass) were included; reproductive variables were in the third level; in the fourth level were maternal smoking variables; and in the fifth level were those variables related to antenatal care.
In the hierarchical model, each set of variables of a specific level were included. A retrospective selection was undertaken and all variables with p >0.25 were eliminated. In this type of model, variables situated in a higher hierarchical level than the variable being analyzed are considered potential confounding factors of the relation between this variable and the outcome of the study. The variables situated in lower levels are considered potential mediators of the effect of the exposure being studied. The variables selected in a particular level were retained in the subsequent models and were considered as risk factors for SGA, even if, with the inclusion of hierarchically inferior variables, they had lost their significance.
As a result of the 1,082 interviews conducted with mothers, 911 infants were included in the study and the remaining 171 were excluded for it was not possible to obtain information concerning their birth weight and gestational age. The prevalence of SGA within the study group was 13.1%.
Table 1 indicates that the prevalence of SGA was higher among those infants whose parents had low income or low educational level. Table 2 indicates that low maternal stature (<150 cm) was associated with a higher prevalence of infants with SGA [odds ratio; 2.88 (95% CI 1.35-6.15)]. On the other hand, parity, sex of the newborn and previous stillborn were not associated with SGA.
Table 3 indicates that those children whose mothers smoked during the entire pregnancy presented a greater chance of being SGA, whereas those children whose mothers stopped smoking during their pregnancy had the same chance of being SGA than those infants whose mothers did not smoke.
Table 4 indicates that poor quality antenatal care was associated with a greater prevalence of SGA.
In the multivariate analysis (see Figure), the variables entered the regression model according to the hierarchical theoretical model, previously discussed. In the first set, the marital status of the mother, maternal age and family income remained in the model, and the latter was the variable most strongly associated with SGA. Even after adjusting for socioeconomic and demographic variables, the chance of an infant being SGA was 2.3 (CI 95% 1.07-5.07) times greater among infants whose mothers height was <150 cm than among infants whose mothers height was >160 cm. The children whose mothers smoked during pregnancy also continued to present a greater chance of being SGA. Mothers with poor quality antenatal care continued to present a greater chance of having a SGA child as well.
This study encompassed almost all births which occurred in hospitals in the city of Pelotas, Rio Grande do Sul State, in southern Brazil, during a three month period in the year 1996, thus reducing the possibility of a selection bias. On the other hand, gestational age was estimated utilizing the date of the last menstruation as a parameter, which led to the exclusion of some mothers who did not remember this date. Mothers who do not remember this date present a greater prevalence of low birth weight and belong to a lower socioeconomic status.5 Consequently, the prevalence of SGA is underestimated in this study.
As to the effect of socioeconomic status on intrauterine growth, low socioeconomic status is associated to smoking, poor nutritional state and other factors which are considered determinants of intrauterine growth. In this study, even after adjusting for the variables situated in the lower hierarchical levels, the children from low income families continued to present a greater chance of being SGA. This is probably a consequence of not including some of the possible mediators of the effects of low socioeconomic status, such as infections during pregnancy and the drug abuse.
In our study, low maternal stature was associated with the birth of SGA infants, in consonance with the literature.9,11 Since this association was observed even after adjustment for confusion by socioeconomic status and pre-gestational weight, it is probably due to a genetic effect or is adaptive to the environmental conditions in the uterus.
As it has been observed in other studies9,11 maternal smoking during pregnancy increases the infants risk for SGA, in a direct dose-response relation, in which an increase in the number of cigarettes smoked during gestation increases the risk of SGA newborns. Maternal smoking may effect intrauterine growth in at least three different ways: fetal hypoxia due to increased levels of carboxyhemoglobulin; uterine vasoconstriction and interference in fetal metabolism.9
Theoretically, antenatal care should have a beneficial impact on intrauterine growth, for it makes it possible to identify and to treat complications during gestation and to eliminate or reduce modifiable risk factors, such as, for example, the habit of smoking. Evidence of the repercussion of antenatal care on the conditions of birth are weak, since the majority of studies failed to establish criteria capable of evaluating the quality of antenatal care. A reduced number of visits would be associated to poor quality antenatal care. For this reason, it would be interesting if the estimates were adjusted to the number of antenatal visits.12 The Kessner index evaluates adequacy, but not quality. The creation of a score related to the content of the visits is an attempt to objectify this process and is one more criteria for evaluating quality. Halpern et al7 demonstrated that, despite the high antenatal coverage in the municipality of Pelotas, Rio Grande do Sul State, there are inequities in the care pregnant women receive. Since the association between poor quality antenatal care and births of SGA infants was observed in this study, even after controlling for the number of antenatal visits, the effect of quality is independent of the number of antenatal visits during gestation, suggesting that, in addition to stimulating an increase in the number of visits, an increase in the quality of antenatal care should be sought.
1. Altman DG, Hytten FE. Intrauterine growth retardation: Let's be clear about it. Br J Obstet Gynecol 1989;96:1127-8. [ Links ]
2. Ashworth A. Effects of intrauterine growth retardation on mortality and morbidity in infants and young children. Eur J Clin Nutr 1998;52:S34-S42. [ Links ]
3. Barker DJP, editor. Fetal and infant origins of infant disease. London: BMJ Publishing Group; 1993. [ Links ]
4. Barros FC, Huttly SRA, Victora CG, Kirkwood BR, Vaughan JP. Comparison of the causes and consequences of prematurity and intrauterine growth retardation: a longitudinal study in Southern Brazil. Pediatrics 1992;90:238-44. [ Links ]
5. Buekens P, Delvoye P,Wollast E, Crobyn C. Epidemiology of pregnancies with unknown last menstrual period. J Epidemiol Community Health 1984;38:79-80. [ Links ]
6. Cnattingius S, Haglund B, Kramer MS. Differences in late fetal death rates in association with determinants of small for gestational age fetuses: population based cohort study. BMJ 1998;316:1483-7. [ Links ]
7. Halpern R, BarrosFC, Victora CG, Tomasi E. Atenção pré-natal em Pelotas, Rio Grande do Sul, Brasil, 1993. Cad de Saúde Pública 1998;14:487-92. [ Links ]
8. Horta BL, Victora CG, Menezes AM, Halpern R, Barros FC. Low birthweight, preterm births and intrauterine growth retardation in relation to maternal smoking. Paediatr Perinatal Epidemiol 1997;11:140-51. [ Links ]
9. Kramer MS. Determinants of low birth weight: methodological assessment and meta- analysis. Bull World Health Organ 1987;65:663-737. [ Links ]
10. Kramer MS, Oliver M, Mc Lean FH, Willis DM, Usher RH. Impact of intrauterine growth retardation and body proportionality on fetal and neonatal outcome. Pediatrics 1990;86:707-13. [ Links ]
11. Lang JM, Lieberman E, Cohen A.A comparison of risk factors for preterm labor and term small-for-gestational birth. Epidemiology 1996;7:369-76. [ Links ]
12. Takeda SMP. Avaliação de unidade de atenção primária: modificação dos indicadores de saúde e qualidade da atenção [dissertação de mestrado]. Porto Alegre: Universidade Federal de Pelotas; 1993. [ Links ]
13. Victora CG, Barros FC, Vaughan JP. Epidemiologia da desigualdade. São Paulo: Hucitec; 1989. [ Links ]
14. Williams RL, Creasy RK, Cunnigham GC, Hawes WE, Norris FD, Tashiro M. Fetal growth and perinatal viability in California. Obstet Gynecol 1982;59:624-32. [ Links ]
15. World Health Organization Expert Committee. Physical status: the use and interpretation of anthropometry. Geneva; 1995. (WHO-Technical Report Series, 854). [ Links ]
Ana Maria Krusser Zambonato
Rua Anchieta, 4374
96015-420 Pelotas, RS
Received on 5/9/2002. Reviewed on 18/6/2003. Approved on 21/6/2003.
Based on a Masters dissertation presented at the Curso de Mestrado em Saúde e Comportamento da Escola de Psicologia da Universidade Católica de Pelotas, 2001.