The Relationship between Social Determinants of Health and Pregnancy Outcomes: A Retrospective Cohort Study in Tabriz

Mahasti Alizadeha, Saeed Dastgirib, Simin Taghavic, Elham Khanlarzadeha, Zhila Khamniana, Mohammad Asghari Jafarabadid, Rana hosseinia, Hossein Jabbari Beyramia,

a: Department of Community Medicine, Faculty of Medicine, Tabriz University of Medical Sciences, Tabriz, Iran.
b: Department of Community Medicine, National Public Health Management Centre (NPMC), Faculty of Medicine, Tabriz University of Medical Sciences, Tabriz, Iran.
c: Obstetrics and Gynecology Department, Al-Zahra Hospital, Tabriz University of Medical Sciences, Tabriz, Iran.
d: Department of Statistics and Epidemiology, Faculty of Health, Tabriz University of Medical Sciences, Tabriz, Iran

Correspondence
Dr. Hossein Jabbari Beyrami National Public Health Management Center, Tabriz University of Medical Sciences, Vali-e-Amr Sq., Tabriz, Iran, IR. Tell: +98411-3364668 Email: hosseinhosseinj@yahoo.com

Received: 2013-12-12
Accepted: 2014-01-20
DOI: 10.13183/jcrg.v3i2.81

Abstract


Purpose: To assess the effects of social determinants of health on pregnancy outcome in rural and urban areas of Tabriz.

Methods: A retrospective cohort study design was used to examine 600 pregnant women who attended antenatal health care services in Tabriz, Iran from August 2012 to the October 2013. They were followed up from the sixth week of gestation to three months after delivery. Socioeconomic indicators and some other measures such as body mass index, family income status, maternal education and occupation, cigarette smoking, depression, intimate partner violence based upon health ministry guidelines and pregnancy outcomes including low birth weight, small for gestational, and preterm delivery, cesarean delivery and miscarriage were assessed.

Results: The mean age of women was 26.1 ± 6.6 years in rural and 27.9 ± 5.7 years in urban areas and the mean duration of pregnancy was 39.03± 1.36 weeks. There was a significant association between maternal education, cesarean section and abortion. BMI played a major role on almost all poor pregnancy outcomes (LBW, PTB, cesarean section, miscarriage) and income status was the first predictor of low birth weight. Socio demographic and behavioral factors were particularly important for predicting miscarriage, preterm labor and low birth weight.

Conclusion: Although residential area (rural versus urban of Tabriz) was associated with preterm labor, low birth weight and cesarean section, only cesarean section had a significant effect after adjusting for confounding factors such as income and education level, age, parity, obstetrical complications. Disparity in distribution of resources in rural and urban area were the main factors that made the pregnancy complicated in our study.

Keywords: Low Birth Weight; Socioeconomic status; Pregnancy outcomes; Preterm labor; Cesarean; Abortion; Small for Gestational Age; Obstetrical complications

© 2014 Swedish Science Pioneers, All rights reserved.


Introduction

Healthy Mother, Healthy Baby is valued hopes and dreams of families of all cultural heritages. National Health goals in several countries round the world prioritize infant [1].

According to WHO report, about 98% of the estimated 529 000 maternal deaths and 5.7 million prenatal deaths occur in developing countries each year. In some countries, a pregnant woman is more than 140 times at risk of dying because of pregnancy problems and more than 50times at risk of pregnancy and delivery complication compared with a woman in a developed country. Every minute more than one woman dye from complication of pregnancy and delivery [2-3].

It is usually accepted that improvements of mother and child health are one of the most important Millennium Development Goals (MDGs) [4]. A woman’s life circumstances play a serious role in determining the health of her baby; known as “social determinants of health” these circumstances include factors such as income and education level, social supports, physical environment and working conditions [5]. Numerous studies have attempted to elucidate pregnant women with low income and education and few social supports could have poorer birth outcomes than pregnant women with higher incomes, education levels and powerful social supports [6]. Women with higher education are more likely to doing healthful behaviors such as; supplementing prenatally with folic acid, seeking early prenatal care, attending prenatal education and exclusively breastfeeding for at least six months. As well, the rates of preterm labor, small for gestational age, stillbirth, infant mortality, smoking, exposure to second-hand smoke and alcohol consumption during pregnancy all decrease once the level of mother’s education increases. Nevertheless, both locally and provincially, women have a higher unemployment rate than men. People with higher incomes generally live longer and healthier than others with lower incomes, means quality and quantity of their life is better [7]. The other previous study has reported that correction of Socioeconomic indicators alone were not associated with reduction in fetal growth or preterm delivery in lower or middle class women and psychological Job demand was alone the weakest predictor for small gestational for age of infant [8]. Education also enhances women’s self-efficacy (confidence in taking freelance and independent choices) and has been associated with other predictors of safe maternity, such as use of contraception programs for birth control, better marital status relationships, and economic independence that all of these illustrating ;once again the close link between biological and social explanations (Grown et al., 2005, Santow, 1995). Lower social class consistently has been related to higher infant mortality rate and lower birth weight in a number of region and nations[9-12]. Racial disparities in obstetrical outcomes are the main problem in developing countries that it has been manifested with consistent two fold increase in the infant mortality rate between African American infants and White infants [13-14].

Based on what mentioned above, it is obvious that understanding the social determinants of health priorities is the first and most important step toward addressing them. Understanding and identifying the determinants of health can help us to know the entry points for action, and also characterizing them are necessary to achieve explicit health goals, including millennium development goals. Nowadays, unfortunately, social determinants of fetal growth and pregnancy adverse outcome have been ignored in developing countries, including Iran. Considering all of these; this paper was conducted to assess SES indicators and other social determinants of health and pregnancy outcome due to maternal characteristics and investigate the role of each determinant in pregnancy outcome.

Methods

Design and sampling

After stratified random sampling in all health centers and private offices in Tabriz, we recruited 20 centers with 600 representative pregnant eligible women in our retrospective cohort study (rural and urban governmental care center and private office) in Tabriz over a 1-year period from August 2012 to October 2013. The eligibility criteria in our cohort was pregnant women in these centers from the 6th week of gestation until 3 months after delivery. The Women lack of this eligibility criteria or unwilling to participate in this survey were excluded from the study. The study was approved by the Research Review Board of Tabriz medical university and carried out under health ethics protocols. All pregnant women having a health record in care center of rural and urban regions of Tabriz or private offices. We used standard questioner of safe pregnancy available in the health ministry of Iran for obtaining data; after all of them received adequate information regarding the study and if they accepted. An interview was done by trained nurses and physicians for missing data and informations that were not available in health record.

Socioeconomic indicators

The selected socioeconomic and other indicators included body mass index, family income status, maternal education, maternal occupation, active and passive smoking, depression, intimate partner violence. Family characteristics based upon health ministry guidelines.

Pregnancy outcome

Outcome of pregnancy included LBW (Low Birth Weight) means infants weighing 2500 g or less, SGA (Small for Gestational Age) mean birth weight below the 10th percentile of weight for a given gestational age according to the Thai standard weight for age, preterm birth means the delivery before 37th gestational week. Furthermore, many other adverse events are discussed such as: Cesarean delivery and miscarriage (20 weeks gestation).

Analysis

The analysis was done using the SPSS 16. The descriptive results were reported as mean ± standard deviation (or median and range) or frequencies and percentages. Multiple linear regressions were used to analyze body mass index. For adjusting confounding, each SES indicator was added to a single independent variable in the model and maternal age, maternal height, parity, past medical history, sex of newborn and number of prenatal care visits were adjusted. The partial F test was used to evaluate the effect of the added variables. Results are presented after controlling the confounder (Adjusted odds ratio). Logistic regression was used to analyze and compare pregnancy outcome and socioeconomic position. P value ≤ 0.05 and power of statistical test 80%were considered respectively.

Results

A total of 600 pregnant women were included in our data analysis; 200 were from the rural health center; 200 pregnant from urban health center and finally 200 pregnant women from private office. Overall, mean age of the mothers and the mean duration of pregnancy were 27.6 ±5.6 years and 39.03 ±1.36 weeks respectively and 4.6% of term births were LBW and of these3% had been recognized as IUGR babies. The comparison of the number of prenatal care visits in pregnant woman (PNC) in urban; rural and office was shown in Figure 1.

Figure 1. The number prenatal care in rural, urban and private office in Tabriz, 2012-13.

thumblarge

The average of BMI in pregnant women was 26.57 ±5. 6, weight gaining was 1.61±1.65, 4.58±2.53 and 4.30±2.11 respectively in first, second and third trimester. Self-reported cigarette smoking of their husbands was 12%, however, none of pregnant women have mentioned the smoking history before and during the study. The other important socioeconomic indicators were reported in Table 1.

Table 1. Social and health care characteristics of pregnant women referred to rural, urban and private office in Tabriz, 2012-13.

thumblarge

A significant relationship was shown between maternal education and cesarean section and abortion (P≤0.05). Although residential area (rural versus urban of Tabriz) was associated with preterm labor, low birth weight and cesarean section, only caesarean section had a significant effect after adjusting confounding factors such as income, education, age, parity, obstetrical complications (P≥0.05). Pregnancy resulted in 3.7% and 2.4%; miscarriage and LGA of 600 women respectively. Cesarean delivery was reported in 74% of live births. Table 2 has shown the distribution of the pregnancy outcome by maternal socioeconomic characteristics.

Table 2. The Relationship between pregnancy outcome and socioeconomic indicator, Tabriz 2012-13(adjusted odd ratio and confidence interval reported).

thumblarge

The strongest social predictors of poor pregnancy outcomes were BMI and education in mothers with regression analysis. BMI plays a major role in almost all poor pregnancy outcomes (LBW, PTB, Cesarean section, miscarriage). Among the psychosocial and behavioral factors, IPV was an important predictor of PTB.

Notably, social demographic and behavioral factors were the most important predictors in miscarriage; and at the same time BMI and gestational diabetes were primary and secondary predictors. While educational attainment was the strongest predictor of PTB and cesarean section; second hand smoking was associated with preterm labor and low birth weights.

Discussion

This is the important large scale study to document pregnancy complication and perinatal care in Tabriz, Iran. Health outcomes are not alone the conclusion of biology, but are produced by social, economic, and cultural determinants, themselves products of processes that distribute power, privilege, and resources across populations. This distribution is often different, shaped by social gradients based on categories such as wealth, education, ethnic background, or place of residence. The WHO Commission on the Social Determinants of Health has drawn attention to how social position determines the conditions in which people are born, grow, live, work and age, and how these conditions then conclude health inequities between and within countries. Overcoming these health inequities, therefore, requires a social determinants approach based on assumptions that harmful living conditions are changeable, and the political mechanisms that cause unfair and unnecessary social stratification are mutable [6].

The results of this study suggested that among the socioeconomic indicators measured by family income status, maternal education and maternal occupation, the family income was associated with low birth weight (P≤0. 05), but preterm labor was not considered significant (P≥0. 05). Confounding factors that adjusted for multiple logistic regression in the study were: age, short stature, previous history of LBW or preterm delivery, obstetrical complications.

A study in Malaysia found that income was correlated with birth weight. Furthermore, income seemed to modify the risk associated with short interparty intervals, where the risk of fetal malnutrition probability is the highest [15].

In our study, the relationship between socioeconomic status and LBW was different from other studies. Some recent studies in developing countries suggested that low SES was not associated with LBW after adjustment for obstetrical factors and smoking [16-21]; but not all several studies in developing countries showed that poor social status was associated with lack of basic social needs and poor access to health care. Low maternal weight, poor obstetrical history, lack of antenatal care, anemia and hypertension were significant, independent risk factors for both preterm and term LBW infants in India and after adjusting for these factors, SES factors were not associated with these outcomes [22].

In our study the SES and IUGR relationship was significant (P≤0.05), but Kramer et al. study concluded that SES had no effect on mean gestational age at delivery or on IUGR in developed and developing Countries (P≥0. 05).

In contrast with previous studies in Thailand, we found no significant and independent associations between SES indicators and SGA except SHS [19]; the causes of this inconsistency between studies reflected in different study designs, sample size, response rate, self reporting of determinant such as income, education, occupation and receiving care in the private center. However, the data on pregnancy outcome of 560 women who consulted private antenatal clinics did show similar rates for LBW (9.8%) and SGA (2.5%). The low proportion of SGA according to a 25- year old standard indicates that birth weight has increased over time, which may be attributable to better social conditions for most population [21].

The study of Michele Kiely, Ayman A. E. El-Mohandes concluded that the strongest Predictors of poor outcomes were pre-pregnancy BMI, preconception diabetes, employment status, intimate partner violence, and depression. BMI was the first splitter for very PTB, large for gestational age, Cesarean section and perinatal death and employment was the first splitter for miscarriage [24].

Some studies looking at antenatal attendance and social class found that women from manual classes were more likely to book late for antenatal Care and/or make fewer antenatal visits than other Women. All studies reporting on antenatal attendance and ethnicity found that women of Asian origin were more likely to book late for antenatal care than white British women [25].

Early Avqnelle Kirksey investigated that pregnancy weight and maternal intake of animal-source foods were significant positive predictors of the newborn’s orientation and habitual behavior, respectively. Habituation and orientation measures assess the infant’s early ability to process information [26].

A major strength of our study was to collect the data prospectively and cohort design and good response rate. But we had some limitation such as small sample size and lack of follow up in private clinical cases. And self reports about depression, smoking, violence and high risk behavior due to wish bias might lead to not achieve the right information. Maternal health is one global health challenges that exhibits dramatic health inequities. Most of the 385,000 maternal deaths occurring each year are concentrated in poor and vulnerable populations and are all the more tragic for being mostly preventable [27]. Maternal health is receiving increased attention in global health, development and human rights agendas and represents a priority area for accelerating progress on the MDGs. The time is right, therefore, to bring together renewed commitments to addressing maternal health and growing understanding of its intermediary and structural determinants.

Notes

Conflict of interests: The authors declare no conflict of interest.

Acknowledgement

The authors deeply appreciate Dr. Roshanak Naseri for her helpful comments on the final draft of the paper.

References

1. Atrash HK, Johnson K, Adams M, Cordero JF, Howse J J, Preconception Care for Improving Perinatal Outcomes: The Time to ActMarten Child Health 2006; 10: Suppl 13-11. CrossRef.

2. Bakhshian F, Jabbari H, Effectiveness of Health Services for Mothers in Iran Health SystemIran Journal of Nursing 2009; 22: 5843-54.

3. Nachvak SM, Jabbari H, Ostadrahimi A, Djafarian K, Weight Gain During Pregnancy and Birth Weight Outcome In Pregnant Women, Tabriz, IranJournal of Health Research 2012; 1: 11-10.

4. Johnson K, Posner SF, Biermann J, Cordero JF, Atrash HK, Parker CS, Recommendations to improve preconception health and health care--United States. A report of the CDC/ATSDR Preconception Care Work Group and the Select Panel on Preconception CareMMWR Recomm Rep 2006; 55: RR-61-23.

5. Bonnefoy J, Morgan A, Kelly MP, Butt J, Bergman V, Constructing the evidence base on the social determinants of health: A guide. Commission on social determinants of health2007; 2: 8232-245.

6. Discussion Paper: A Social Determinant Approach to Maternal Health. Roles for Development Actors http://www.undp.org/content/Discussion Paper MaternalHealth.

7. Social Determinants of Health for Women of Reproductive Age http://www.simcoemuskokahealthstats.org/Topics/PregnancyAndBefore/HealthDuringPregnancy/SocialDeterminants.aspx.

8. Kirksey A, Rahmanifar A, Wachs TD, McCabe GP, Bassily NS, Bishry Z, Determinants of pregnancy outcome and newborn behavior of a semirural Egyptian populationAm J Clin Nutr 1991; 54: 4657-67.

9. Antonovsky A, Bernstein J, Social class and infant mortalitySoc Sci Med 1977; 11: 8-9453-70.

10. Dunn HG, Social aspects of low birth weightCan Med Assoc J 1984; 130: 91131-40.

11. Rodriguez C, Regidor E, Gutierrez-Fisac JL, LBW in Spain associated with sociodemographic factorsJ Epidermal Community Health 1995; 49: 138-42.

12. Joffre M, Social inequalities in low birth weight: Timing of effects and selective mobilitySoc Sci Med 1989; 28: 6613-19.

13. Healy AJ, Malone FD, Sullivan LM, Porter TF, Luthy DA, Comstock CH, Early access to prenatal care: Implications for racial disparity in perinatal mortalityObstetrics and Gynecology 2006; 107: 625-631.

14. Hyatt S, With special feature on medical technologyNational Center for Health Statistics 2009;

15. DaVanzo J, Habicht P, Butz WP, Assessing socioeconomic correlates of birth weight in peninsular Malaysia: ethnic differences and changes over timeSoc Sci Med 1984; 18: 5387-404.

16. Nordstrom ML, Cnattingius S, Effects on birth weights of maternal education, socioeconomic status, and work related characteristicsScand J Public Health 1996; 24: 155-61 CrossRef.

17. Peters TJ, Golding J, Butler NR, Fryer JG, Lawrence CJ, Chamberlain GV, Plus ça change: predictors of birthweight in two national studiesBr J Obstet Gynaecol 1983; 90: 111040-1045.

18. Lumley J, Correy JF, Newman NM, Curran JT, Cigarette smoking, alcohol consumption and fetal outcome in Tasmania 1981–1982Aust N Z J Obstet Gynaecol 1985; 25: 133-40.

19. Bell R, Lumley J, Low birthweight and socioeconomic status, Victoria 1982 to 1986Aust J Public Health 1992; 16: 115-19.

20. Morrison J, Najman JM, Williams GM, Keeping JD, Andersen MJ, Socio-economic status and pregnancy outcome. An Australian studyBr J Obstet Gynaecol 1989; 96: 3298-307.

21. Tuntiseranee P, Olsen J, Chongsuvivatwong V, Limbutara S, Socioeconomic and work related determinants of pregnancy outcome in southern ThailandJournal of epidemiology and community health 1999; 53: 10624-629.

22. Mavalankar DV, Gray RH, Trived CR, Risk factors for preterm and term Low Birthweight in Ahmedabad, IndiaInt J Epidemiol 1992; 21: 2263-72 CrossRef.

23. Kramer MS, Determinants of low birth weight: methodological assessment and meta-analysisBull World Health Organ 1987; 65: 5663-73.

24. Kiely M, El-Mohandes AA, Gantz MG, Chowdhury D, Thornberry JS, El-Khorazaty MN, Understanding the association of biomedical, psychosocial and behavioral risks with adverse pregnancy outcomes among African-Americans in Washington, DCMatern Child Health J 2011; 15: Suppl 1S85-S95 CrossRef.

25. Rowe RE, Garcia J, Social class, ethnicity and attendance for antenatal care in the United Kingdom: a systematic reviewJ Public Health Med 2003; 25: 2113-9.

26. Kirksey A, Rahmanifar A, wachs TD, McCabe GP, Bassily NS, Bishry Z, Determinants of pregnancy outcome and newborn behavior of a semi rural Egyptian populationAm J Clin Nutr 1991; 54: 4657-67.

27. Livingood WC, Brady C, Pierce K, Atrash H, Hou T, Bryant T, 3rdImpact of pre-conception health care: evaluation of a social determinants focused interventionMatern Child Health J 2010; 14: 3382-91 CrossRef.