ABSTRACT:

Several studies have identified social inequalities in low birth weight (LBW), preterm birth (PTB), and intrauterine growth restriction (IUGR), which, in recent years, have diminished or disappeared in certain locations.

Objectives:

Estimate the LBW, PTB, and IUGR rates in São Luís, Maranhão, Brazil, in 2010, and check for associations between socioeconomic factors and these indicators.

Methods:

This study is based on a birth cohort performed in São Luís. It included 5,051 singleton hospital births in 2010. The chi-square test was used for proportion comparisons, while simple and multiple Poisson regression models with robust error variance were used to estimate relative risks.

Results:

LBW, PTB and IUGR rates were 7.5, 12.2, and 10.3% respectively. LBW was higher in low-income families, while PTB and IUGR were not associated with socioeconomic factors.

Conclusion:

The absence or weak association of these indicators with social inequality point to improvements in health care and/or in social conditions in São Luís.

Keywords:
Socioeconomic factors; Infant, low birth weight; Premature birth; Fetal growth retardation

RESUMO:

Vários estudos mostram desigualdades sociais no baixo peso ao nascer (BPN), nascimento pré-termo (NPT) e restrição do crescimento intrauterino (RCIU), que nos últimos anos diminuíram ou desapareceram em determinados locais.

Objetivos:

Estimar as taxas de BPN, NPT e RCIU em São Luís, Maranhão, Brasil, em 2010, e verificar as associações entre fatores socioeconômicos e esses indicadores.

Métodos:

Este estudo baseia-se em uma coorte de nascimentos realizada em São Luís. Incluiu 5.051 nascimentos únicos hospitalares em 2010. O teste do qui-quadrado foi utilizado para comparação de proporções, enquanto modelos de regressão de Poisson simples e múltipla com variância robusta foram usados para estimar riscos relativos.

Resultados:

As taxas de BPN, NPT e RCIU foram de 7,5, 12,2 e 10,3%, respectivamente. O BPN foi maior em famílias de baixa renda, enquanto NPT e RCIU não estiveram associados com fatores socioeconômicos.

Conclusão:

A ausência ou associação fraca desses indicadores com desigualdades sociais aponta para melhorias na atenção à saúde e/ou em condições sociais em São Luís.

Palavras-chave:
Fatores Socioeconômicos; Recém-nascido de baixo peso; Nascimento prematuro; Retardo do crescimento fetal

INTRODUCTION

It is well documented in the literature that low birth weight (LBW), preterm birth (PTB) and intrauterine growth restriction (IUGR) are implicated in major infant and perinatal morbidities and mortality and are risk factors for diseases in adult life¹1. Blencowe H, Cousens S, Oestergaard MZ, Chou D, Moller AB, Narwal R, et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet 2012; 379(9832): 2162-72. DOI: 10.1016/S0140-6736(12)60820-4
https://doi.org/10.1016/S0140-6736(12)60... ^,22. World Health Organization (WHO). The world health report 2005: make every mother and child count. Geneva: WHO; 2005; 33(6): 409-11. DOI: 10.1080/14034940500217037
https://doi.org/10.1080/1403494050021703... ^,33. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet 2008; 371(9606): 75-84. DOI: 10.1016/S0140-6736(08)60074-4
https://doi.org/10.1016/S0140-6736(08)60... .

The World Health Organization (WHO) showed that newborns weighing between 1,500-2,000 grams have a 20 times higher mortality rate than newborns without LBW²2. World Health Organization (WHO). The world health report 2005: make every mother and child count. Geneva: WHO; 2005; 33(6): 409-11. DOI: 10.1080/14034940500217037
https://doi.org/10.1080/1403494050021703... . About 65% of deaths from children in the United States were due to LBW and PTB³3. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet 2008; 371(9606): 75-84. DOI: 10.1016/S0140-6736(08)60074-4
https://doi.org/10.1016/S0140-6736(08)60... , with PTB being responsible for three million deaths worldwide¹1. Blencowe H, Cousens S, Oestergaard MZ, Chou D, Moller AB, Narwal R, et al. National, regional, and worldwide estimates of preterm birth rates in the year 2010 with time trends since 1990 for selected countries: a systematic analysis and implications. Lancet 2012; 379(9832): 2162-72. DOI: 10.1016/S0140-6736(12)60820-4
https://doi.org/10.1016/S0140-6736(12)60... . Moreover, these indicators are related to morbidities in adult life: LBW is related to hypertension, diabetes, and metabolic syndrome⁴4. Chen W, Srinivasan SR, Yao L, Li S, Dasmahapatra P, Fernandez C, et al. Low birth weight is associated with higher blood pressure variability from childhood to young adulthood. Am J Epidemiol 2012; 176(7): S99-105. DOI: 10.1093/aje/kws298
https://doi.org/10.1093/aje/kws298... ^,55. Christensen DL, Kapur A, Bygbjerg IC. Physiological adaption to maternal malaria and other adverse exposure: low birth weight, functional capacity, and possible metabolic disease in adult life. Int J Gynaecol Obstet 2011; 115(1): S16-9. DOI: 10.1016/S0020-7292(11)60006-4
https://doi.org/10.1016/S0020-7292(11)60... ; PTB to lung, neurological and ophthalmological diseases⁶6. McCormick MC. The contribution of low birth weight to infant mortality and childhood morbidity. N Engl J Med 1985; 312(2): 82-90. DOI: 10.1056/NEJM198501103120204
https://doi.org/10.1056/NEJM198501103120... ^,77. Page JM, Schneeweiss S, Whyte HE, Harvey P. Ocular sequelae in premature infants. Pediatrics 1993; 92(6): 787-90.^,88. Moster D, Lie RT, Markestad T. Long-term medical and social consequences of preterm birth. New Engl J Med 2008; 359(3): 262-73. DOI: 10.1056/NEJMoa0706475
https://doi.org/10.1056/NEJMoa0706475... ; and IUGR to diabetes, hypertension and coronary artery disease⁹9. Barker DJ. Mothers, babies, and health in later life: Edinburgh: Churchill Livingstone; 1998. 217 p.^,1010. Leon DA. Fetal growth and adult disease. Eur J Clin Nutr 1998; 52(1): S72-8..

Birth weight is determined by two processes: duration of pregnancy and intrauterine growth, therefore LBW is due to either PTB, IUGR or an association of both. In developed countries, PTB is responsible for the largest number of neonates with LBW. On the other hand, in developing countries, IUGR is the most important factor¹¹11. Kramer MS. Determinants of low birth weight: methodological assessment and meta-analysis. Bull World Health Organ 1987; 65(5): 663-737.^,1212. Kramer MS. The epidemiology of adverse pregnancy outcomes: an overview. J Nutr 2003; 133(5 Suppl 2): 1592S-6S.^,1313. Chen Y, Li G, Ruan Y, Zou L, Wang X, Zhang W. An epidemiological survey on low birth weight infants in China and analysis of outcomes of full-term low birth weight infants. BMC Pregnancy Childbirth 2013; 13: 242. DOI: 10.1186/1471-2393-13-242
https://doi.org/10.1186/1471-2393-13-242... . In the year 2000 in China, 38.8% of LBW were due to PTB and 61.2% to IUGR, whereas in 2011 PTB had contributed to 69.6% of LBW and IUGR contributed to 30.4%. This change was attributed to improvements in the Chinese health system that have been occurring over the years¹³13. Chen Y, Li G, Ruan Y, Zou L, Wang X, Zhang W. An epidemiological survey on low birth weight infants in China and analysis of outcomes of full-term low birth weight infants. BMC Pregnancy Childbirth 2013; 13: 242. DOI: 10.1186/1471-2393-13-242
https://doi.org/10.1186/1471-2393-13-242... . In the city of Pelotas, Rio Grande do Sul, Brazil, similar changes were observed between 1993 and 2004, when PTB with LBW rates rose from 42.5 to 67.3%¹⁴14. Barros FC, Victora CG, Matijasevich A, Santos IS, Horta BL, Silveira MF, et al. Preterm births, low birth weight, and intrauterine growth restriction in three birth cohorts in Southern Brazil: 1982, 1993 and 2004. Cad Saúde Pública 2008; 24(3): S390-8. DOI: 10.1590/S0102-311X2008001500004
https://doi.org/10.1590/S0102-311X200800... .

Several studies have shown how socioeconomic inequality influences these indicators. In China, the LBW rate ranged from 2.5 to 9.4%, depending on the region studied, being higher in underdeveloped regions and lower in economically developed ones¹³13. Chen Y, Li G, Ruan Y, Zou L, Wang X, Zhang W. An epidemiological survey on low birth weight infants in China and analysis of outcomes of full-term low birth weight infants. BMC Pregnancy Childbirth 2013; 13: 242. DOI: 10.1186/1471-2393-13-242
https://doi.org/10.1186/1471-2393-13-242... . This same study showed that lower educational levels are associated with a greater risk of LBW.

In Quebec, Canada, between the years 2000 and 2008, mothers with low education and income had a higher risk of IUGR¹⁵15. Savard N, Levallois P, Rivest LP, Gingras S. Impact of individual and ecological characteristics on small for gestational age births: an observational study in Quebec. Chronic dis Inj Can 2014; 34(1): 46-54.. In Newcastle, England, between 1961 and 2000, PTB and LBW were higher in neonates of mothers from lower economic classes. In this same study, PTB rate increased in the lower classes but declined in the upper classes within the same period, but the factors involved in these changes have not yet been identified¹⁶16. Glinianaia SV, Ghosh R, Rankin J, Pearce MS, Parker L, Pless-Mulloli T. No improvement in socioeconomic inequalities in birthweight and preterm birth over four decades: a population-based cohort study. BMC Public Health 2013; 13: 345. DOI: 10.1186/1471-2458-13-345
https://doi.org/10.1186/1471-2458-13-345... . Despite studying different perinatal outcomes, both studies show the negative impact of low socioeconomic status in perinatal health.

In Pelotas, the risk of LBW was 2.8 times higher in families with lower income in 1982. In 2004, despite the drop in LBW with increasing income, there was an increase in the percentage of LBW among those in the higher wage stratum. The authors attributed this change to greater medical intervention in the high-income group¹⁴14. Barros FC, Victora CG, Matijasevich A, Santos IS, Horta BL, Silveira MF, et al. Preterm births, low birth weight, and intrauterine growth restriction in three birth cohorts in Southern Brazil: 1982, 1993 and 2004. Cad Saúde Pública 2008; 24(3): S390-8. DOI: 10.1590/S0102-311X2008001500004
https://doi.org/10.1590/S0102-311X200800... .

In the city of Ribeirão Preto, São Paulo, Brazil, in the years 1978/79 and 1994, LBW and IUGR rates were lower in families of higher income, education, and non-manual occupations. PTB rates had the same socioeconomic pattern in 1978/79, but this difference disappeared in 1994¹⁷17. Lamy Filho F, Assuncao Junior AN, Silva AA, Lamy ZC, Barbieri MA, Bettiol H. Social inequality and perinatal health: comparison of three Brazilian cohorts. Braz J Med Biol Res 2007; 40(9): 1177-86. DOI: 10.1590/S0100-879X2006005000154
https://doi.org/10.1590/S0100-879X200600... .

In São Luís, Maranhão, Brazil, in 1997/98 there were no differences in LBW and PTB rates regarding schooling, income, and occupation of the head of the family. On the other hand, IUGR was more prevalent in mothers with low income and schooling¹⁷17. Lamy Filho F, Assuncao Junior AN, Silva AA, Lamy ZC, Barbieri MA, Bettiol H. Social inequality and perinatal health: comparison of three Brazilian cohorts. Braz J Med Biol Res 2007; 40(9): 1177-86. DOI: 10.1590/S0100-879X2006005000154
https://doi.org/10.1590/S0100-879X200600... .

In Brazil, in the year 2005, a higher proportion of term neonates with LBW were found in mothers with low education¹⁸18. Coutinho PR, Cecatti JG, Surita FG, Souza JP, Morais SS. Factors associated with low birth weight in a historical series of deliveries in Campinas, Brazil. Rev Assoc Med Bras 2009; 55(6): 692-9. DOI: 10.1590/S0104-42302009000600013
https://doi.org/10.1590/S0104-4230200900... . In other study, with data from 2006/2007, LBW was again more prevalent in mothers with low education, however it was more prevalent in the South and Southeast Brazilian regions, considered the richest area in Brazil, and least prevalent in the North and Northeast regions, considered the poorest in Brazil¹⁹19. Viana KJ, Taddei JA, Cocetti M, Warkentin S. Birth weight in Brazilian children under two years of age. Cad Saúde Pública 2013; 29(2): 349-56. DOI: 10.1590/S0102-311X2013000600021
https://doi.org/10.1590/S0102-311X201300... . Other Brazilian study performed in 2009 found similar results, with LBW and PTB being more prevalent in the richest regions and less prevalent in the poorest regions of the country²⁰20. Lima MC, Oliveira GS, Lyra CO, Roncalli AG, Ferreira MA. The spatial inequality of low birth weight in Brazil. Cien Saúde Coletiva 2013; 18(8): 2443-52. DOI: 10.1590/S1413-81232013000800029
https://doi.org/10.1590/S1413-8123201300... .

The objectives of this study were to estimate LBW, PTB and IUGR rates in São Luís, in 2010; to verify if social inequalities were related to these indicators; and to verify if those socioeconomic inequalities remained, increased, or decreased between 1997/98 and 2010 in this town.

METHODS

This study used data from a birth cohort study called Brazilian Ribeirão Preto and São Luís Birth Cohort Studies (BRISA), carried out from January 1^st to December 31^st 2010. In this paper we used data only from the city of São Luís.

The city of São Luís is the capital of the state of Maranhão, located in the northeast of Brazil, the country’s poorest region. The city had an estimated population of 1,082,935 inhabitants in 2016²¹21. Brasil. Instituto Brasileiro de Geografia e Estatística (IBGE). Censo Demográfico 2016. Brasil: Instituto Brasileiro de Geografia e Estatística; 2016. and a human development index (HDI) of 0.768, occupying the 249^th place in the national ranking in 2010²²22. Instituto de Pesquisa Econômica Aplicada (IPEA). Índice de Desenvolvimento Humano Municipal Brasileiro. PNUD, IPEA e FJP. Brasilia: IPEA; 2013..

SAMPLING

A population-based sample from hospital live births with gestational age (GA) > 20 weeks or birth weight > 500 grams was studied. The study was limited to women residing in the municipality of São Luís for at least six months. Hospital births accounted for 98% of births in 2010²³23. Silva AA, Coimbra LC, Silva RA, Alves MT, Lamy Filho F, Carvalho LZ, et al. Perinatal health and mother-child health care in the municipality of Sao Luis, Maranhao State, Brazil. Cad Saúde Pública 2001; 17(6): 1412-23. DOI: 10.1590/S0102-311X2001000600012
https://doi.org/10.1590/S0102-311X200100... ^,2424. Silva AA, Batista RF, Simoes VM, Thomaz EB, Ribeiro CC, Lamy Filho F, et al. Changes in perinatal health in two birth cohorts (1997/1998 and 2010) in São Luis, Maranhão State, Brazil. Cad Saúde Pública 2015; 31(7): 1437-50. DOI: 10.1590/0102-311X00100314
https://doi.org/10.1590/0102-311X0010031... .

For the selection of maternity hospitals, data recorded in the Brazilian National System of Information on Live Births (SINASC) in the year of 2008 were retrieved, and 18,255 live births of residents in the municipality were recorded. The live births were distributed across 16 health units, including public and private ones. After the exclusion of hospitals in which less than 100 births were performed in the year of 2008 (3.3% of all hospital births in the city of São Luís), 10 maternity wards were selected. Hence, the sample frame included 94.7% of all births that occurred in 2010²³23. Silva AA, Coimbra LC, Silva RA, Alves MT, Lamy Filho F, Carvalho LZ, et al. Perinatal health and mother-child health care in the municipality of Sao Luis, Maranhao State, Brazil. Cad Saúde Pública 2001; 17(6): 1412-23. DOI: 10.1590/S0102-311X2001000600012
https://doi.org/10.1590/S0102-311X200100... ^,2424. Silva AA, Batista RF, Simoes VM, Thomaz EB, Ribeiro CC, Lamy Filho F, et al. Changes in perinatal health in two birth cohorts (1997/1998 and 2010) in São Luis, Maranhão State, Brazil. Cad Saúde Pública 2015; 31(7): 1437-50. DOI: 10.1590/0102-311X00100314
https://doi.org/10.1590/0102-311X0010031... .

A systematic sampling technique was used, and the sample was stratified by maternity hospital with shares proportional to the number of births in each maternity ward. A sampling interval of three was defined, corresponding to approximately 6,000 deliveries, or 1/3 of all births that occurred in 2010, according to data from SINASC. A list of all births occurring in each hospital, according to the order of birth, was made. On the first day, for each hospital or maternity, a casual number between 1 and 3 was randomly chosen. Then, the sampling interval value was added to the casual number, and all births were randomly drawn for this study, successively²³23. Silva AA, Coimbra LC, Silva RA, Alves MT, Lamy Filho F, Carvalho LZ, et al. Perinatal health and mother-child health care in the municipality of Sao Luis, Maranhao State, Brazil. Cad Saúde Pública 2001; 17(6): 1412-23. DOI: 10.1590/S0102-311X2001000600012
https://doi.org/10.1590/S0102-311X200100... ^,2424. Silva AA, Batista RF, Simoes VM, Thomaz EB, Ribeiro CC, Lamy Filho F, et al. Changes in perinatal health in two birth cohorts (1997/1998 and 2010) in São Luis, Maranhão State, Brazil. Cad Saúde Pública 2015; 31(7): 1437-50. DOI: 10.1590/0102-311X00100314
https://doi.org/10.1590/0102-311X0010031... .

With the study sample, it was possible to estimate rates of LBW, PTB or IUGR of around 50% (maximum product of p and q, being p the estimated rate and q = 1 - p) with an accuracy of 2 and 99% confidence level. It was also possible to compare two proportions, considering a 5% probability of type I error, and an 80% study power, working with the product maximum of p × q (event proportion of 50%) and fixing in 4% the minimum difference to be detected as significant. For rates of less than 50%, it was possible to detect smaller differences (it was possible to detect a relative difference of 3% for rates of 10% and of 2% for rates of 5%)²⁵25. Rosner B. Fundamentals of biostatistics. Seventh Edition. Cengage Learning; 2011; 859..

At the end of 2010, there was a total of 21,401 births in São Luís, one-third of which were picked by drawing lots (7,133). Of them, 5,475 were residents in the municipality for at least six months. With a loss of 4.6% from mothers that refused to be interviewed, and from early discharges, a total of 5,236 interviews was undertaken. After excluding multiple births, stillbirths and births with no weight record, the final sample of this study was 5,051 births.

INSTRUMENTS AND VARIABLES

For the collection of maternal, paternal and fetal data, a standardized questionnaire with questions related to identification, sociodemographic conditions, health, pregnancy, labor and birth was used.

For this study, the following variables were analyzed: newborn weight, date of last menstrual period, maternal education, head of the family occupation, family income and economic class²³23. Silva AA, Coimbra LC, Silva RA, Alves MT, Lamy Filho F, Carvalho LZ, et al. Perinatal health and mother-child health care in the municipality of Sao Luis, Maranhao State, Brazil. Cad Saúde Pública 2001; 17(6): 1412-23. DOI: 10.1590/S0102-311X2001000600012
https://doi.org/10.1590/S0102-311X200100... .

Prior to data collection, the research team was trained and a pilot study with all stages of the research was held for 24 hours for correction of possible errors.

Before the interview, mothers were informed about the objectives of the study and an informed consent was obtained.

Infants weighing < 2,500 grams were considered LBW. Birth weight was measured using infant digital scales adjusted to 10 grams²⁶26. Aragão VM, Silva AA, Aragão LF, Barbieri MA, Bettiol H, Coimbra LC, et al. Risk factors for preterm births in São Luís, Maranhão, Brazil. Cad Saúde Pública 2004; 20(1): 57-63. DOI: 10.1590/S0102-311X2004000100019
https://doi.org/10.1590/S0102-311X200400... . The newborns were weighed immediately after birth without clothes²³23. Silva AA, Coimbra LC, Silva RA, Alves MT, Lamy Filho F, Carvalho LZ, et al. Perinatal health and mother-child health care in the municipality of Sao Luis, Maranhao State, Brazil. Cad Saúde Pública 2001; 17(6): 1412-23. DOI: 10.1590/S0102-311X2001000600012
https://doi.org/10.1590/S0102-311X200100... .

Gestational age was calculated from the date of the last menstrual period reported by the mother. The 15^th day of the month was imputed in all cases for which only the day (not the month) of the last menstrual period was unknown. In cases of incompatible weight for gestational age, or gestational age located above the 99^th percentile of the English curve²⁷27. Altman DG, Coles EC. Nomograms for precise determination of birth weight for dates. Int J Gynaecol Obstet 1980; 87(2): 81-6. DOI: 10.1111/j.1471-0528.1980.tb04498.x
https://doi.org/10.1111/j.1471-0528.1980... , the date of the last menstrual period was recoded as missing. The same procedure was used for cases of implausible gestational age (less than 20 or more than 50 weeks). Finally, a process of imputation was performed for gestational age. All cases of originally missing data on gestational age or data recorded as missing were imputed in a linear regression model. Predictors of gestational age were birth weight, parity, family income, and sex of the newborn. A total of 446 cases were imputed, 29 as preterm and 458 as term based on the complete cases.

Newborns with a gestational age of less than 37 weeks were classified as preterm²⁶26. Aragão VM, Silva AA, Aragão LF, Barbieri MA, Bettiol H, Coimbra LC, et al. Risk factors for preterm births in São Luís, Maranhão, Brazil. Cad Saúde Pública 2004; 20(1): 57-63. DOI: 10.1590/S0102-311X2004000100019
https://doi.org/10.1590/S0102-311X200400... .

The classification of weight for gestational age was based on Williams curve²⁸28. Williams RL, Creasy RK, Cunningham GC, Hawes WE, Norris FD, Tashiro M. Fetal growth and perinatal viability in California. Obstet Gynecol 1982; 59(5): 624-32.. IUGR was considered when birth weight was below the 10^th percentile.

For the definition of socioeconomic indicators, maternal education was classified into four groups: 0-4, 5-8, 9-11 and greater than or equal to 12 years of education. The person with the highest income in the family was regarded as the head of the family, and his/her occupation was classified as non-manual, manual skilled/semi-skilled and manual unskilled/unemployed.

For the definition of economic class the Brazilian criterion of economic classification was used, ranking families in class A (high purchasing power), B, C, D or E (low purchasing power)²⁹29. Associação Brasileira de Empresas de Pesquisa (ABEP). Critério de Classificação Econômica Brasil 2010. São Paulo: IBOPE; 2012.. For the family income variable, the monthly family income group ranges were defined as up to and including 1 minimum wage, greater than 1 and less than or equal to 3 minimum wages, and more than 3 minimum wages. Since a high percentage of mothers did not report their family income, a missing category was added to this variable, instead of excluding cases with missing values from the analysis. The minimum wage as of August 17^th, 2015 was R$ 510.00, corresponding to US$ 146.97 monthly or US$ 1,910.61 annually.

STATISTICAL ANALYSIS

The data were entered in duplicates in Microsoft Office Access 2007 computer program, and were compared for error correction. Subsequently, they were transferred to Stata 12 (Stata Corporation, College Station, Texas, USA) and analyzed.

Absolute frequencies and percentages were calculated for descriptive analysis. For the comparison of proportions, we used the chi-square test with a significance level of 5% (p < 0.05). To check the associations between socioeconomic indicators with perinatal outcomes, we initially used the simple Poisson regression with robust adjustment of variance to calculate the relative risk (RR) with a 95% confidence interval (CI)³⁰30. Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004; 159(7): 702-6. DOI: 10.1093/aje/kwh090
https://doi.org/10.1093/aje/kwh090... . We then used separate models for LBW, PTB and IUCR, and, in each of those models, the socioeconomic variables were analyzed together, using multiple Poisson regression with robust adjustment of variance for control of confounding variables³⁰30. Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol 2004; 159(7): 702-6. DOI: 10.1093/aje/kwh090
https://doi.org/10.1093/aje/kwh090... .

ETHICAL ASPECTS

This study meets the criteria established by Resolution no. 196/96 of the National Health Council and its complementary regulations. Mothers who agreed to participate in the study signed an informed consent form. The participants had the option to leave the study at any time without any consequences to them or their families. The project and the informed consent form were approved by the Ethics Committee of the University Hospital of the Federal University of Maranhão.

RESULTS

LBW, PTB and IUGR rates were 7.5, 12.2 and 10.3%, respectively, in 2010. In the unadjusted analysis, LBW rates were not associated with maternal socioeconomic indicators. However, there was a 34% higher risk of PTB among infants of mothers with 5-8 years of education (Table 1).

There was also a 41% higher risk of IUGR among infants of mothers with 9-11 years of education, a 38% higher risk in newborns whose head of the family had a manual skilled/semiskilled occupation, a 27% higher risk among infants of mothers with a family income of > 1 and ≤ 3 minimum wages, and a 37% higher risk among those whose families earned ≤ 1 monthly minimum wage (Table 1).

After adjustment for confounding variables, a 50% higher risk of LBW among infants of mothers ≤ 1 minimum wage (p = 0.040; 95%CI 1.02 - 2.20) was demonstrated. There were no associations between the socioeconomic variables and PTB or IUGR after adjustment for the confounding variables (Table 2).

DISCUSSION

PTB and IUGR rates were not associated to socioeconomic variables. LBW rate was higher in infants from low-income families, but that association was close to the significance level.

Compared to a similar cohort performed in 1997/98 in São Luís, there was a drop in IUGR rate from 13.3 to 10.3% (p < 0.001), but LBW (7.5%) and PTB (12.2%) rates remained stable¹⁷17. Lamy Filho F, Assuncao Junior AN, Silva AA, Lamy ZC, Barbieri MA, Bettiol H. Social inequality and perinatal health: comparison of three Brazilian cohorts. Braz J Med Biol Res 2007; 40(9): 1177-86. DOI: 10.1590/S0100-879X2006005000154
https://doi.org/10.1590/S0100-879X200600... .

Those rates were comparable to the ones in developed countries like the United States, which had a LBW rate of 8% and PTB of 11.4% in 2013³¹31. Martin JA, Hamilton BE, Osterman MJ, Curtin SC, Matthews TJ. Births: final data for 2013. Natl Vital Stat Rep 2015; 64(1): 1-65.. However, they are still high compared to other countries like Spain, with a PTB rate of 7% in 2010³²32. Fuster V, Zuluaga P, Colantonio SE, Roman-Busto J. Factors determining the variation in birth weight in Spain (1980-2010). Ann Hum Biol 2013; 40(3): 266-75. DOI: 10.3109/03014460.2013.765034
https://doi.org/10.3109/03014460.2013.76... , Iceland, with a 3% LBW, 4.6% PTB and 1.5% IUGR rate from 2006 to 2009³³33. Eiriksdottir VH, Asgeirsdottir TL, Bjarnadottir RI, Kaestner R, Cnattingius S, Valdimarsdottir UA. Low birth weight, small for gestational age and preterm births before and after the economic collapse in Iceland: a population based cohort study. PloS One 2013; 8(12): e80499. DOI: 10.1371/journal.pone.0080499
https://doi.org/10.1371/journal.pone.008... , Finland, with a 3.1% IUGR rate from 1967 to 2010³⁴34. Raisanen S, Gissler M, Sankilampi U, Saari J, Kramer MR, Heinonen S. Contribution of socioeconomic status to the risk of small for gestational age infants--a population-based study of 1,390,165 singleton live births in Finland. Int J Equity Health 2013; 12: 28. DOI: 10.1186/1475-9276-12-28
https://doi.org/10.1186/1475-9276-12-28... , or Sweden, with a 4.8% PTB and 2% IUGR rate from 1999 to 2010³⁵35. Baba S, Wikstrom AK, Stephansson O, Cnattingius S. Influence of smoking and snuff cessation on risk of preterm birth. Eur J Epidemiol 2012; 27(4): 297-304. DOI: 10.1007/s10654-012-9676-8
https://doi.org/10.1007/s10654-012-9676-... .

Compared to a cohort in São Luís in 1997/98, LBW and PTB rates stayed the same in the socioeconomic groups analyzed. IUGR decreased among less educated mothers (from 14.2 to 10.4%, p = 0.007 in mothers with 5-8 years of education, from 15.3 to 8.8%, p = 0.018 in mothers with 0-4 years of education) and in mothers whose head of the family had manual unskilled jobs or were unemployed (14.8 to 9.7%, p < 0.001)¹⁷17. Lamy Filho F, Assuncao Junior AN, Silva AA, Lamy ZC, Barbieri MA, Bettiol H. Social inequality and perinatal health: comparison of three Brazilian cohorts. Braz J Med Biol Res 2007; 40(9): 1177-86. DOI: 10.1590/S0100-879X2006005000154
https://doi.org/10.1590/S0100-879X200600... . The decrease in IUGR rate over the past 10 years, accompanied by a reduction in the number of growth restricted neonates in less advantaged social groups, suggested an improvement in health care and/or health conditions, especially in the lower classes, the most prevalent in this study. The improvement in health care and/or health conditions may be in part explained by the implementation of the Bolsa Família Program (conditional cash transfer program) in 2003 and the Family Health Program (geographically organized public family health multidisciplinary teams providing primary care for defined populations) in 1994³⁶36. Macinko J, Harris MJ. Brazil's family health strategy--delivering community-based primary care in a universal health system. N Engl J Med 2015; 372(23): 2177-81. DOI: 10.1056/NEJMp1501140
https://doi.org/10.1056/NEJMp1501140... ^,3737. Nery JS, Rodrigues LC, Rasella D, Aquino R, Barreira D, Torrens AW, et al. Effect of Brazil's conditional cash transfer programme on tuberculosis incidence. Int J Tuberc Lung Dis 2017; 21(7): 790-6. DOI: 10.5588/ijtld.16.0599
https://doi.org/10.5588/ijtld.16.0599... . Both programs are geared towards the improvement of quality of life of underserved populations. Aquino et al. showed that infant mortality rates decreased as the Family Health Program coverage increased³⁸38. Rasella D, Aquino R, Barreto ML. Impact of the family health program on the quality of vital information and reduction of child unattended deaths in Brazil: an ecological longitudinal study. BMC Public Health 2010; 10: 380. DOI: 10.1186/1471-2458-10-380
https://doi.org/10.1186/1471-2458-10-380... ; and Rasella et al. showed that, among children under 5 years old, mortality rates decreased as the Bolsa Família Program coverage increased³⁷37. Nery JS, Rodrigues LC, Rasella D, Aquino R, Barreira D, Torrens AW, et al. Effect of Brazil's conditional cash transfer programme on tuberculosis incidence. Int J Tuberc Lung Dis 2017; 21(7): 790-6. DOI: 10.5588/ijtld.16.0599
https://doi.org/10.5588/ijtld.16.0599... . It is possible that weak or non-existing socioeconomic inequalities in perinatal indicators, depicted here, could be due, at least partially, to the effects of those programs.

Lamy et al. showed that in São Luís, in 1997/98, there were no associations of LBW with socioeconomic variables. PTB was lower in middle-income families (> 1 and ≤ 3 minimum wages), and IUGR was associated with low education (0-4 years) and low income (≤ 1 minimum wage)¹⁷17. Lamy Filho F, Assuncao Junior AN, Silva AA, Lamy ZC, Barbieri MA, Bettiol H. Social inequality and perinatal health: comparison of three Brazilian cohorts. Braz J Med Biol Res 2007; 40(9): 1177-86. DOI: 10.1590/S0100-879X2006005000154
https://doi.org/10.1590/S0100-879X200600... . In 2010 minimal socioeconomic inequality in adverse perinatal outcomes were detected because, only in the case of low birth weight, families with ≤ 1 minimum wage had a 50% higher risk of LBW. However, since the p-value was very close to the significance level, that association may be due to random error.

LBW is caused mainly by IUGR in developing countries, and by PTB in developed countries. In China, for example, there was a reversal over ten years; LBW was mainly caused by IUGR in 2000, but became mainly caused by PTB in 2011¹¹11. Kramer MS. Determinants of low birth weight: methodological assessment and meta-analysis. Bull World Health Organ 1987; 65(5): 663-737.^,1212. Kramer MS. The epidemiology of adverse pregnancy outcomes: an overview. J Nutr 2003; 133(5 Suppl 2): 1592S-6S.^,1313. Chen Y, Li G, Ruan Y, Zou L, Wang X, Zhang W. An epidemiological survey on low birth weight infants in China and analysis of outcomes of full-term low birth weight infants. BMC Pregnancy Childbirth 2013; 13: 242. DOI: 10.1186/1471-2393-13-242
https://doi.org/10.1186/1471-2393-13-242... . In Pelotas, similar results from 1993 to 2004 were observed¹⁴14. Barros FC, Victora CG, Matijasevich A, Santos IS, Horta BL, Silveira MF, et al. Preterm births, low birth weight, and intrauterine growth restriction in three birth cohorts in Southern Brazil: 1982, 1993 and 2004. Cad Saúde Pública 2008; 24(3): S390-8. DOI: 10.1590/S0102-311X2008001500004
https://doi.org/10.1590/S0102-311X200800... . However, in São Luís, the percentage of PTB among LBW did not change significantly: from 51.1% in 1997/98 to 54.3% in 2010 (p = 0.646)¹⁷17. Lamy Filho F, Assuncao Junior AN, Silva AA, Lamy ZC, Barbieri MA, Bettiol H. Social inequality and perinatal health: comparison of three Brazilian cohorts. Braz J Med Biol Res 2007; 40(9): 1177-86. DOI: 10.1590/S0100-879X2006005000154
https://doi.org/10.1590/S0100-879X200600... ^,3939. Silva AA, Bettiol H, Barbieri MA, Ribeiro VS, Aragão VM, Brito LG, et al. Infant mortality and low birth weight in cities of Northeastern and Southeastern Brazil. Rev Saúde Pública 2003; 37(6): 693-8. DOI: 10.1590/S0034-89102003000600002
https://doi.org/10.1590/S0034-8910200300... .

The results of our study oppose several studies in Brazil, Europe, Asia and North America, which found a higher risk of adverse perinatal outcomes among families of low socioeconomic class, education, and whose parents had manual jobs¹³13. Chen Y, Li G, Ruan Y, Zou L, Wang X, Zhang W. An epidemiological survey on low birth weight infants in China and analysis of outcomes of full-term low birth weight infants. BMC Pregnancy Childbirth 2013; 13: 242. DOI: 10.1186/1471-2393-13-242
https://doi.org/10.1186/1471-2393-13-242... ^,1515. Savard N, Levallois P, Rivest LP, Gingras S. Impact of individual and ecological characteristics on small for gestational age births: an observational study in Quebec. Chronic dis Inj Can 2014; 34(1): 46-54.^,1616. Glinianaia SV, Ghosh R, Rankin J, Pearce MS, Parker L, Pless-Mulloli T. No improvement in socioeconomic inequalities in birthweight and preterm birth over four decades: a population-based cohort study. BMC Public Health 2013; 13: 345. DOI: 10.1186/1471-2458-13-345
https://doi.org/10.1186/1471-2458-13-345... ^,1717. Lamy Filho F, Assuncao Junior AN, Silva AA, Lamy ZC, Barbieri MA, Bettiol H. Social inequality and perinatal health: comparison of three Brazilian cohorts. Braz J Med Biol Res 2007; 40(9): 1177-86. DOI: 10.1590/S0100-879X2006005000154
https://doi.org/10.1590/S0100-879X200600... ^,1818. Coutinho PR, Cecatti JG, Surita FG, Souza JP, Morais SS. Factors associated with low birth weight in a historical series of deliveries in Campinas, Brazil. Rev Assoc Med Bras 2009; 55(6): 692-9. DOI: 10.1590/S0104-42302009000600013
https://doi.org/10.1590/S0104-4230200900... ^,1919. Viana KJ, Taddei JA, Cocetti M, Warkentin S. Birth weight in Brazilian children under two years of age. Cad Saúde Pública 2013; 29(2): 349-56. DOI: 10.1590/S0102-311X2013000600021
https://doi.org/10.1590/S0102-311X201300... ^,3232. Fuster V, Zuluaga P, Colantonio SE, Roman-Busto J. Factors determining the variation in birth weight in Spain (1980-2010). Ann Hum Biol 2013; 40(3): 266-75. DOI: 10.3109/03014460.2013.765034
https://doi.org/10.3109/03014460.2013.76... ^,3434. Raisanen S, Gissler M, Sankilampi U, Saari J, Kramer MR, Heinonen S. Contribution of socioeconomic status to the risk of small for gestational age infants--a population-based study of 1,390,165 singleton live births in Finland. Int J Equity Health 2013; 12: 28. DOI: 10.1186/1475-9276-12-28
https://doi.org/10.1186/1475-9276-12-28... ^,4040. Park MJ, Son M, Kim YJ, Paek D. Social inequality in birth outcomes in Korea, 1995-2008. J Korean Med Sci 2013; 28(1): 25-35. DOI: 10.3346/jkms.2013.28.1.25
https://doi.org/10.3346/jkms.2013.28.1.2... ^,4141. Silvestrin S, Silva CH, Hirakata VN, Goldani AA, Silveira PP, Goldani MZ. Maternal education level and low birth weight: a meta-analysis. J Pediatr 2013; 89(4): 339-45. DOI: 10.1016/j.jped.2013.01.003
https://doi.org/10.1016/j.jped.2013.01.0... ^,4242. Bener A, Saleh NM, Salameh KM, Basha B, Joseph S, Al Buz R. Socio-demographic and consanguinity risk factors associated with low birthweight. J Pak Med Assoc 2013; 63(5): 598-603.^,4343. Aizer A, Currie J. The intergenerational transmission of inequality: maternal disadvantage and health at birth. Science 2014; 344(6186): 856-61. DOI: 10.1126/science.1251872
https://doi.org/10.1126/science.1251872... .

The main strength of this study is the use of a random population-based sample (a 1/3 of the births of São Luís’ residents), which allows for generalization of the results for the general population of live births.

One of the limitations of our study is the percentage of missing values for gestational age, which was attenuated by the imputation of those values. The missing values for income (911) and economic class (300) were also important to mention.

CONCLUSION

In São Luís, from 1997/98 to 2010, there was a drop in IUGR rate and a stability of LBW and PTB rates. Moreover, there was no socioeconomic inequality in PTB and IUGR rates. Regarding LBW, there was little inequality in terms of income, since only those ones born in families with ≤ 1 minimum wage showed a higher LBW rate, and that association was close to the level of significance. Overall, it points to an improvement in health care and/or in health and living conditions, as there was little social inequality according to the three studied perinatal outcomes.

ACKNOWLEDGEMENTS

We thank the interviewers and the mothers who kindly agreed to participate in the study.

REFERENCES

Publication Dates

History