Background: Detection and management of high-risk pregnancies, all the way through antenatal care, have been advocated as a high-quality mean of reducing maternal and perinatal morbidity and mortality.

Objectives: This study reviewed the demographic variables, pregnancy and obstetric complications and perinatal outcomes for the years 1999 and 2004 in a rural hospital in KwaZulu-Natal Province, South Africa, with the aim of evaluating trends and gaps that may enhance appropriate strategies for improvement of antenatal care.

Method: A retrospective comparative study, with representative samples of pregnant women, were randomly selected for the respective years 1999 and 2004. Descriptive statistics were calculated depending on measurement scale. A Z-test was carried out to assess the significant difference (p < 0.05) in proportions between pregnancy complications and outcomes of the groups. Multivariate logistic regression analysis was undertaken to determine the significant predictors for outcome variables.

Results: The numbers of pregnancies among young women (< 25 years) increased significantly by 8% (p < 0.05) in the year 2004. Compared with 1999, the reduction in the numbers of pregnancies (1%) among higher parity (parity 5 or more) women in 2004 was remarkable. There were significant reductions of eclampsia, anaemia and post partum haemorrhage. Women with breech presentation were 3.75 times more likely to deliver preterm, and 5.45 times more likely to deliver low birth-weight babies. Similarly, women with pregnancy-induced hypertension were more likely to have preterm (OR = 3.50, 95% CI 2.83; 4.35) and low birth-weight babies (OR = 2.09, 95% CI 1.62; 2.71). Eclampsia was also a risk factor associated with preterm deliveries (OR = 6.14, 95% CI 3.74; 10.09) and low birth-weight babies (OR = 3.40, 95% CI 1.83; 6.28).

Conclusion: This study suggests that further research is needed to find the causes of higher rate of teenage pregnancies and an increase in quality of antenatal care is more important in improving maternal and perinatal health. Training of staff to standard protocol and guidelines on antenatal care and care during delivery, and adherence to it, should be encouraged to improve maternal and child health in South Africa.

Ensuring better maternal and child health through reducing maternal and child mortality is the one of the UN Millennium Development Goals adopted by South Africa's (SA) health system.¹ Detection and management of high-risk pregnancies, all the way through antenatal care, have been advocated as a high-quality means of reducing maternal and perinatal morbidity and mortality. Experience from developing countries has illustrated that making use of this practice has curbed pregnancy complications up to 60%.^2,3 Annually, approximately 530 000 women die worldwide from pregnancy related conditions and over 15% of pregnant women from developing countries experience pregnancy complications, resulting in negative maternal and perinatal outcomes.^4,5 Interventions in controlling the incidence of maternal and perinatal mortality and morbidity, considered effective in the context of developing countries, are not universally provided. A study conducted in Nigeria established that mothers who booked into health facilities for antenatal care and delivered in maternity units, experienced fewer complications and a lower maternal mortality rate than 'unbooked' mothers.⁶

Antenatal care includes, (1) education of pregnant women about pregnancy, labour, childbirth and newborn care, (2) prophylaxis for anaemia, malaria and immunisation against tetanus and (3) identification and management of pregnancy-related conditions such as anaemia, pregnancy-induced hypertension, diabetes and so on. Maternal health education through public health programmes should underline women's knowledge in very specific areas. These include:

• contextualised information about the appropriate age to become pregnant
• understanding 'danger signs' during pregnancy
• family planning
• appropriate 'diet and exercise'
• the minimum gap between two pregnancies to ensure both baby and mother's sound health
• an awareness of HIV/AIDS to improve their chances of a good pregnancy outcome.

Despite all efforts to optimise the positive pregnancy and its outcome nationally and locally, maternal and perinatal mortality remains high in SA. For example, the anticipated maternal mortality ratio for SA was estimated as 147 per 100 000 live births, but it was actually higher (154) for the KwaZulu-Natal (KZN) Province in 2004.⁷ The perinatal mortality rates for KZN and SA were 26.5 and 37.5 per 1000 live births, respectively, in 2006.⁸

Maternal health programme efforts, particularly in the rural areas of the developing countries, are often poorly designed and quite deficient in reaching the target population.⁹ Previously, rights to utilise benefits of maternal health services was a major problem for the rural and Black communities of SA.¹⁰ Following the political transition in 1994, introduction of free maternal and child (< 6 years) healthcare (MCH) in the public health sector was very encouraging. It was reported from two rural areas of KZN (Hlabisa and Ubombo) that over 95% of pregnant women attend health facilities during pregnancy.^{11, 12} However, recent surveys of antenatal patients in Pretoria and in Greytown, KZN, have demonstrated that the majority of women only commence their antenatal care late in their pregnancy (i.e. in the second or third trimester).^13,14 Standard protocol and guidelines on antenatal care and care during delivery have been developed nationally and implemented throughout the country since 2002 to improve maternal and child health in South Africa.¹⁵

A few studies, particularly from rural KZN, have shown the demographic profile of pregnant women, along with pregnancy complications and outcomes. This study was undertaken with the objectives of measuring and comparing demographic profiles of pregnant women, specific pregnancy and obstetric complications and indicators for perinatal outcomes for the years 1999 and 2004. Data were collected from a rural hospital in order to measure trends and gaps so that appropriate measures for the improvement of antenatal care could be taken.

Study setting

Study design, sample selection and data collection

The facts about obstetrics and labour that were routinely recorded in the register included:

• presentation of foetus during labour
• plurality
• method of induction
• augmentation of labour
• time of delivery (recorded in hours and minutes)
• mode of delivery (normal vaginal, assisted vaginal delivery using vacuum, forceps or operative caesarean section during delivery)
• complications of delivery (e.g. perineal and/or cervical tear, post-partum haemorrhage [PPH] etc.)
• side-by-side perinatal information, including birth weight and birth outcome of babies (live birth, still birth, Apgar score in 1 min and 5 min).

Attending midwives recorded this information during admission and after delivery. All the midwives working at the labour ward were oriented and received in-service training on filling in the labour ward register, as well as for the compilation of weekly and monthly summaries for the presentation at weekly and monthly perinatal mortality meetings held in the hospital.

Data analysis

Prior written permission was obtained from the hospital policy and ethics committee to make use of the maternity register to conduct the study. No identification of patients or staff was required to present the results.

The mean ages of pregnant women across the two periods were similar (24.61 years), although remarkable differences in age groups and parity were observed (Table 1). The proportions of teenage mothers were considerably higher in 2004 (17%), compared to 1999 (14%). A notable increase of pregnancy (8%) among mothers below 25 years of age was also noted. Women in this age group constituted 50% of the total number of pregnancies in 1999 and this increased to 58% in 2004. There was a momentous reduction (p < 0.05) of pregnancies among higher parity (parity 5 or more) groups for the year 2004 (4.26%), compared to 2004 (5.35%).

Comparison of pregnancy and obstetric complications are given in Table 2. The rates that increased in 2004 were post-term and vacuum deliveries, compared to those for 1999. Those that were significantly lower in 2004, were incidences of eclampsia, the prevalence of anaemia at 36 weeks (or later at the gestational age), induction of labour and PPH. No significant variations were observed regarding the rates of ante partum hemorrhage (APH), PIH, gestational diabetes, multiple pregnancy, preterm delivery, caesarean section delivery (including emergency procedures), augmentation of labour; third-degree perineum tear and retained placenta during the two comparative periods. Table 3 illustrates the perinatal outcomes between the two study periods. Perinatal outcomes were measured in terms of low birth-weight (< 2.5 kg, LBW) at delivery, live birth, still birth, fresh still births (FSBs), macerated still births (MSBs) rates and mean Apgar Scores in 1 min and 5 min. The different perinatal outcome indicators were similar for the two study years.

Multiple logistic regression outputs are shown in the Table 4. In this analysis preterm delivery, LBW babies and FSBs were considered as dependent variables. Independent variables found to be significantly associated with the dependent variables, using the chi-square test, were: age of mothers, breech presentation of foetus, induction and augmentation of labour, PIH, eclampsia, and caesarean, vacuum and forceps deliveries. Breach presentation was a key predictor for preterm and LBW deliveries; a breech presentation is 3.75 times more likely to result in a preterm delivery and 5.45 times more likely to result in the delivery of a LBW baby. A probable reason for such observations could be related to the immaturity of the foetus. Mothers with previous caesarean deliveries were found to be twice as likely (OR = 2.26) to deliver FSBs. Induction of labour was instituted more frequently for preterm pregnancies

(OR = 2.46) and thus resulted in more LBW babies (OR = 2.24) and FSBs (OR = 4.38). Pregnant mothers with PIH were more likely to experience preterm delivery (OR = 3.50) and LBW babies or deliveries (OR = 2.09). Eclampsia during pregnancy was a major risk factor for preterm (OR = 6.14) and LBW (OR = 3.40) deliveries, while forceps deliveries were more likely (OR = 5.79) to deliver FSB babies.

This study was confined to pregnant women who delivered at Empangeni Hospital in 1999 and 2004. The idea of conducting such a study presented itself to the authors in 1999 and a period of 5 years was considered appropriate in terms of gathering adequate data to serve the purpose of comparison, therefore 2004 became the year for this research setting. Nonetheless, these data reflected large delivery information of the population of Uthungulu district, because the majority of the deliveries under public health facilities were conducted at Empangeni Hospital. However, during the two study periods, mothers who delivered at home or at private health facilities were not included in this study. Although a vast majority (more than 95%) of pregnant women was known to attend public health facilities for antenatal care in rural areas of KZN, the exclusion of deliveries at home and in private hospitals are considered as a limitation of the present study.^11,12 The proportion of pregnant women who delivered at public health facilities was not known, because maternity care is free of charge in public health facilities and, as such, encourages women to utilise these facilities for deliveries, thus one could expect a higher rate of utilisation by the target population. The retrospective review of records limited the availability of some study variables and consequently led to information bias. For instance, recordkeeping on HIV infection and maternal deaths were not registered properly; thus maternal mortality estimates were not done. In addition, the rate of HIV infection and its effects on perinatal outcomes were not estimated; incomplete and incorrect recording of observations and histories by the attending midwives may also have led to information bias. However, the midwives were trained and oriented with the labour ward register. Thus, it was assumed that such bias would be slight.

It is observed that there has been a significant increase of teenage pregnancies in Empangeni across the two study years (1999 and 2004) and this finding is consistent with the provincial rate for KZN.⁸ The rate of pregnancies among younger mothers (ages < 25 years) has shown a drastic increase of 50% in 1999 to 58% in 2004. The factors which might have contributed to this increase include the following: firstly, there has been a change in sexual behaviour in society, including early initiation in sex, secondly, non-use or poor use of family planning methods by the younger women and, thirdly, a possible tendency to become pregnant in order to obtain social benefits in the form of government-sponsored child support.^16,17 The higher numbers of pregnancies among younger girls may be a result of one of the above-mentioned factors, or a combination of two or more of these factors. In an antenatal care audit, conducted at the same hospital during 2004, it was found that 90% of those pregnant women had no income.¹⁸ Therefore, effective strategies are urgently needed to improve the socio-economic conditions of rural people, particularly pregnant girls who are still in school. Our findings regarding the demographic characteristics of our study subjects were similar to those of another study on the same (pregnant) population, in terms of age and parity.¹⁹

A significant reduction, 5.35% in 1999 to 4.26% in 2004, of higher parity (? 5) women, could have been due to successful adoption or implementation of effective family planning methods by the higher parity and older women and/or rapidly growing urbanisation in South Africa.^20,21 It is believed that urban dwellers, including women, have better access to health facilities, health education and contraception. It is thus assumed that city dwellers aspire to a higher standard of living. It is also assumed that urbanites are generally receptive to, and value, the idea of planning their careers, as well as their families. Women who pursue a higher education tend to delay motherhood until they are older and more established in their careers. As a result, an implication of this trend is that they have smaller families.

A substantial decrease in the prevalence of eclampsia (0.9%), anaemia (5.8%) and PPH, were observed during 2004, although the pattern of changes in pregnancy, obstetrics and perinatal outcomes were very slight. In the developing countries, the prevalence rates of eclampsia are known to vary widely, from 0.5% to 8%.²² It is a common risk because of the following reasons: poverty, a limited education that results in low literacy, lack of knowledge and awareness regarding health issues, and superstitious beliefs and practices, all of which prevent women from seeking medical care and advice during their pregnancy.²¹ These factors may prevail among the rural population to a larger degree in comparison to city dwellers.

Nevertheless, a recent study has shown a very encouraging finding based on an antenatal care audit that confirmed that almost all pregnant women attend antenatal care.¹⁸ The finding also revealed that the majority (88%) received antenatal care from PHC facilities, with an average of six antenatal visits before delivery. More than 95% of the women had routine clinical and laboratory examinations on every visit, as recommended by the national guidelines.¹⁸ Thus, it is assumed that the optimal management of essential hypertension and PIH were the key factors in probable prevention of eclampsia through effective antenatal care. It is likely that the provision of antenatal care by following the standard protocol and guidelines might have contributed to a reduction in the rate of eclampsia. Similarly, lower (though not significant) rates of APH (8%) and PIH (7.2%) were observed in 2004, compared to the findings of 1999 (APH of 6.2% and PIH of 8.8%). To reduce anaemia in pregnancy, prophylaxis with haematinics was given to all pregnant women, who were intermittently screened for the presence of anaemia (e.g. at the booking visit, thereafter at 28 weeks and again at 36 weeks). In addition, appropriate intervention were undertaken during the antenatal period, as required by the national protocol.¹⁵ There was no significant difference as far as multiple pregnancies were concerned, because these were related to biological factors specific to the women and, therefore, no changes were expected. The overall caesarean section delivery rates seemed higher for both periods and corresponded to the provincial rate of KZN.⁸

Perinatal outcome

Preterm delivery rates were similar during the comparison periods, but at a higher rate (> 11%), compared to developed countries during the same period.²⁴ Preterm delivery is the most important cause of perinatal mortality in the developed world.²⁵ The prevalence of preterm delivery in first-world countries is between 6% and 10%.²⁴ No obvious causes of preterm delivery have yet been established, but several etiological risk factors have been identified as non-obstetrical causes for preterm birth. These include poor socio-economic status, maternal malnutrition and/or smoking, illiteracy, maternal age below 20 years and over 35 years, smoking, and trauma are.^26,27 The obstetric risk factors found to be associated with preterm deliveries are: cervical incompetence, multiple pregnancies, short birth intervals, pre-labour premature rupture of membrane and previous preterm deliveries.^27,28 A number of other medical conditions have also been associated with preterm birth and these include diabetes mellitus, urinary and genital tract infections, HIV infection, and psychological stress.²⁹ Some of these factors were prevalent in our present study population, especially in cases where poor socio-economic conditions and higher rates of HIV infection were present.²³

Low birth-weight delivery rates, though, remained at the same high level for both years (> 13%). However, this rate was lower when compared to many other African countries (e.g. 18% in Tanzania in 1999 and 19% in Kenya in 2004), but higher than in, for instance, Brazil in 1996 (9%), which has a comparable economy to SA.^30,31,32 Well-known factors for LBW delivery are premature delivery and intrauterine growth retardation. Personal (genetic, prior premature birth, age), social, environmental and medical risk factors (e.g. shorter pregnancy intervals, HIV infection, poor maternal nutrition, smoking), have been found to be associated with low birth weight in different parts of the world.²⁵ These factors prevail in South Africa, particularly in rural KZN. In addition, an increasing rate of HIV infection among pregnant women in KZN (20% in 1996 and 37.5% in 2004) has been observed in this study.²³

In South Africa, after the political transition in 1994, access to healthcare services, particularly maternal and child healthcare, became free of charge at the public health facilities in order to remove financial barrier to health services and thereby ensuring better access. As such, one can assume that pregnant mothers attended and received the standard of care, without any prohibiting factors preventing them from seeking the care and support they may need. Higher attendances and improved standard of care thus may have improved overall pregnancy outcomes in the present study population.

The prevalence of pregnancy in teenagers and young women (< 25 years) has been found to be on the increase, but the rate is reduced among higher parity and older groups of women.Most of the pregnancy complications were found to have decreased over time (from 1999 to 2004) whereas the perinatal outcomes remained at the same level. Improvement of maternal and child health initiatives at different levels of public service delivery may have contributed to the effective management of pregnancy complications. Further research is needed to discover simple, yet effective strategies to combat negative perinatal outcomes.

The authors wish to acknowledge the help of the professional nurses in the retrieval of data from the hospital register, the pregnant women for their willingness to take part in the study, Ehsanul Hoque for statistical analysis, and the Empangeni Hospital management team for supporting the study.

1. World Health Organization (WHO). Health and the Millennium Development Goals (530.1). Geneva: WHO; 2005.

2. Backett EM, Davies AM, Petros-Barzavian A. The risk approach in health care, with special reference to natural and child health. WHO Pub Health Papers. 1984;76:113.

3. Nash BT, Leke RJ, Doh AS, Kamdom MJ, Fomulu J, Njikam OM. The risk approach for reducing natural mortality: The Journde experience. Int Journal of Gynecol Obstet. 1991;36:195-201.

4. WHO. The world health report 2005. Make every mother and child count. Geneva: WHO; 2005.

5. WHO. Maternal mortality in 2000: Estimates developed by WHO, UNICEF and UNFFA. Geneva: WHO; 2004.

6. Harrison KA. Child-bearing, health and social priorities: A survey of 22 774 consecutive hospital births in Zaris, northern Nigeria. Br J Obstet Gynaecol. 1985;92:1-118.

7. Medical Research Council and National Department of Health. Saving mothers - Third report on confidential enquiries into maternal deaths in South Africa 2002-2004. Pretoria: Medical Research Council & Department of Health; 2006.

8. Pattinson RC, editor. Saving babies 2003-2005: Fifth perinatal care survey of South Africa. Pretoria: University of Pretoria, MRC, CDC; 2007.

9. Bulatao RA, Ross JA. Rating maternal and neonatal health services in developing countries. Bull World Health Org. 2002;60:721-727.

10. Hirschowitz R, Orkin M, De Castro J, et al., editors. A national household survey of health inequalities in South Africa. Washington: Agency for Social Enquiry for the Henry J Family Foundation; 1995.

11. Tsoka JM, Le Seur D, Sharp BL. Maternal health services utilization in Ubombo district. S Afr J Obstet Gynaecol. 2003;9:70-73.

12. Wilkinson D, Cutts F, Ntuli N, Abdul Karrim SS. Maternal and child health indicators in rural South African health district. S Afr Med J. 1997;87:456-459.

13. Tsuari M, Mabale T, Kgobane R, Pattinson RC. Health seeking behavior of pregnant woman. Proceedings of the 17th Priorities in Perinatal Care Conference; 1998 Mar 3-6; Free State, South Africa.

14. Hoque AKM, Kader SB, Hoque E, Mugero C. Prevalence of anaemia in pregnancy at Greytown, South Africa. Trop J Obstet Gynaecol. 2006;23:3-7.

15. Department of Health. Guidelines for Maternity Care in South Africa - A manual for clinics, community health centers and district hospitals. 2nd ed. Department of Health; 2002.

16. Mbizvo MT, Bounduelle MMJ, Chadzuka S, Lindmark G, Nystrom L. Unplanned pregnancies in Harare. What are the social and sexual determinants? Soc Sci Med. 1997;45:937-942.

17. Gillespie D, Ahmed S, Tsui A, Radloff S. Unwanted fertility among the poor an inequity? Bull World Health Organ. 2007;85:1000-1007.

18. Hoque M, Kader SB, Hoque E. Prevalence of anaemia in pregnancy at Uthungulu health district of KwaZulu-Natal, South Africa. SA Fam Practice. 2007;49:16.

19. Hoque M, Hoque E, Kader SB. Audit of antenatal care at a rural district of KZN, South Africa. SA Fam Practice. 2008;50(3):60-65 .

20. Jajeebhoy S. Maternal mortality and morbidity in India: Priorities for social science research. J Fam Welf. 1997;43:31-52.

21. Savitz DA, Kaufman JS, Dole N, Siega-Riz AM, Thorp JM, Kaczor DT. Poverty, education, race and pregnancy outcomes. Ethn Dis. 2004;14:322-329.

22. Begum MR, Begum A, Quadir E, Akhter S, Shamsuddin L. Eclampsia: Still a problem in Bangladesh. Med Gen Med. 2004;6:52.

23. Department of Health. National HIV and syphilis antenatal sero-prevalence survey in South Africa 2004. [document on the Internet]. c2005 [cited 2005 Sep 19]. Available from: http://www.doh.gov.za/docs/reports/2004/hiv-syphilis02.pdf

24. Ezechi OC, Makinde ON, Kalu BE, Nnatu SN. Risk factors for preterm delivery in South Western Nigeria. J Obstet Gynaecol. 2003;23:387-391.

25. Danielian PJ, Hall MH. The epidemiology of prematurity. Curr Obstet Gynecol. 1996;6:133-136.

26. Rawlings JS, Rawlings VB, Read JA. Prevalence of low birth weight and preterm deliveries in relation to the interval between pregnancies among White and Black women. N Engl J Med. 1995;332:69-74.

27. Pitman MC, Steer PJ. Prematurity. Assessment of uterine activity. Curr Obstet Gynecol. 1996;6:137-142.

28. Brocklehurst P, French R. The association between HIV infection and perinatal outcome: A systematic review of the literature and meta-analysis. Br J Obstet Gynaecol. 1998;105:836-48.

29. Moerman ML. Growth of the birth canal in adolescent girls. Am J Obstet Gynecol. 1982;143:528-532.

30. Menendez C, Ordi J, Ismail MR, et al. The impact of placental malaria on gestational age and birth weight. J Infect Dis. 2000;181:1740-1745.

31. Guyatt HL, Snow RW. Impact of malaria during pregnancy on low birth weight in sub-Saharan Africa. Clin Microbiol Rev. 2004;4:760-769.

32. Silva AA, Barbieri MA, Gomes UA, Behiol H. Trends in low birth weight: A comparison of two birth cohorts separated by a 15-year interval in Ribeirao Petro, Brazil. Bull World Health Organ. 1998;76:73-84.