Determinants of delivery practices among Afar pastoralists of Ethiopia

Determinants of delivery practices among Afar pastoralists of Ethiopia

 

Medhanit Getachew Mekonnen^1,&, Kassahun Negash Yalew¹, Jemal Yesouf Umer¹, Muluken Melese¹

 

¹Africa Medical and Research Foundation (AMREF), Ethiopia

 

 

^&Corresponding author
Medhanit Getachew Mekonnen, Africa Medical and Research Foundation (AMREF), Ethiopia

 

 

 

 

More than 90% of maternal deaths in the world occur in Sub-Saharan Africa (SSA) and Asia, whilst women in northern Europe have a 1 in 4,000 likelihood of dying from pregnancy related causes. For those in Africa, the chance is 1 in 16 [1]. Most of these deaths occur because of distant health facilities; lack of transport; high cost of service for mothers and inadequate skilled attendants; poorly motivated staff; inadequate equipment and supplies; and weak referral systems [2].

 

In low and middle income countries, most deliveries take place at home without the assistance of trained attendants [3-5]. Although the annual maternal deaths decreased from 526,300 in 1980 to 342,900 in 2008, only 23 countries have met the Millennium Development Goal (MDG 5) [6]. The MDG 5 is to reduce maternal mortality ratio by 75 per cent by 2015 [7]. The World Health Organization indicated that each year about 3.7 million children die within the first 28 days and close to 9.7 million children die before their fifth birthday [8]. UNICEF stated that progress towards reaching MDG 4, the goal of reducing the under-five mortality rate by two thirds by 2015, is "insufficient" in the Middle East and North Africa, SSA and South Asia [9]. The burden of maternal mortality remains greatest in SSA [10]. Studies have shown that around 20 - 30% of neonatal deaths can be averted if deliveries are attended by skilled birth attendants [11].

 

Home deliveries are common in developing countries. Studies in some countries reported that home deliveries range from 16% in Mumbai (India), 22% in Senegal, 65% in Tanzania to 87.7% in Bangladesh [12-15]. The main reasons given by mothers for home deliveries is distance to health facilities [13,14,16,17]; cost for medical care; delivery complications; low education level and absence of ANC [18,19]. This study was designed to collect data on the rate of home delivery and the determining factors for institutional delivery, among women in rural and urban areas of zone 3 of Afar region.

 

 

Study area

 

This study was conducted in Zone 3 of the Afar regional state of Ethiopia. This Zone is divided into six districts namely Awash, Amibara, Birumodytu, Gewane, Dulessa and Argoba special districts. The total population of Afar regional state is 1,418,000 and the study zone population is nearly 200,000 [20]. Afar regional state is located in the eastern part of Ethiopia and lies in the East African Great Rift Valley. In the west it is bordered by Tigray, Amhara and Oromia regional states; in the south by Somali regional state; in the east by Djibouti and in the north by Eritrea. It is a semi-arid region and more than 90% of the population is pastoralist.

 

Study design and sample size

 

A cross-sectional design was employed for interviewing women, between the ages of 15 - 49 (reproductive age groups) who delivered within the last one year, on determinants of delivery practices.

 

The sample size was calculated using Epi-Info version 3.3 for Windows for single proportion formula. Calculation was based on the assumption of Type I Error of 5%, reported delivery in health facilities of 12.9 % (Afar Regional Health Bureau 2006/2007 report), and a designing effect of 3 with 5% non response rate. The total sample size was 502 women between the ages of 15 - 49 years. A three stage sampling strategy, that is, districts to kebele then household was used to reach the target population.

 

Estimated household size in each district was obtained from the National Population and Housing census report. The number of women to be interviewed (502) was allocated equally to 4 districts; these districts were randomly selected out of 6 districts of the Zone. Then, 12 kebeles (smallest unit of public structure and possibly less than a village) were randomly selected in each district and the allocated households for the district were equally distributed amongst the kebeles (which amounted to about 11 mothers per kebele). In each selected kebele, the centre of the kebele was identified as the starting point and a bottle was spun to determine the direction to be followed. The survey then started from the nearest house to the direction of the bottle top, and continued in the determined direction. Households were surveyed sequentially until the desired number was obtained. All women in the age category of 15 - 49 years, from the sampled households, who gave birth during the last one year, were interviewed using the survey questionnaire.

 

Data collectors who had previous experience of data collection were recruited and trained for two days on how to use the questionnaire. The questionnaire was initially developed in English and translated to Amharic for data collection and translated back to English. It was tested and finally all findings from the pre-test were incorporated into the final form of the questionnaire.

 

To estimate the rural households' asset levels, a wealth ranking was conducted, based on the following five categories, namely poorer (cows/oxen 0, camels 0, sheep/goats less than 5); poor (cows/oxen less than five, camels 0, sheep/goats between 5 - 9); middle (cows/oxen 5 - 9, camels 1 - 5, sheep/goats 10 - 19); rich (cows/oxen 10 - 29, camels 5 - 9, sheep/goats 20 - 49) and richer (cows/oxen 30 and above, camels 10 and above, sheep/goats 50 and above). During the data collection, the number of animals for each category was collected from the sampled households.

 

Data Quality Control and Analysis

 

During data collection, supervisors watched over the data collectors on site and every evening checked the data for accuracy, consistency and completeness. Supervisors also rechecked five percent of the samples in order to cross check the collected data. The data were entered and analyzed in SPSS software (version 16). The association between the socio-demographic and reproductive variables, ANC attendance and delivery place were analyzed using Pearson's Chi-square test. Logistic regression analysis was also used to measure unadjusted and adjusted odds ratios to identify the predictors of health facility delivery.

 

The asset level for the rural dwellers was calculated, taking the average market value of each category of the animal in Birr to get the total asset value of the household. However, during analysis it was found that the middle and rich categories were few; therefore for convenience of the analysis, the data was merged into rich and poor only. The richer, and the rich categories were merged into rich and the others were merged into poor.

 

Ethical Clearance

 

A written permit was obtained from Afar Regional Health Bureau and verbal consent from interviewed mothers.

 

 

Socio-demographic characteristics

 

Out of the sampled 502 mothers who delivered during the last one year, 478 (95%) were interviewed. Out of all those interviewed, 347 (72.7%) belonged to the Afar ethnic group while Amhara accounted for 47 (9.8%); the rest, 83 (17.8%), belonged to various other groups. The majority of women, 468 (97.3%), were married and only 1.7% were single. The socio-demographic characteristics are presented in Table 1.

 

Delivery place

 

Out of the 478 mothers interviewed about the place of their last delivery, 398 (83.3%) had delivered at home and the remaining 16.7%) had delivered in health facilities. Among women who delivered at home 370 (92.5%) of them were assisted by TBAs and about 3.6% were assisted by Health Extension Workers/nurse; a further 3.2% were assisted by a neighbor. Out of all the reasons given as to why they preferred home deliveries, 205 (51.5%) of them said that it was because labour progressed fast and there was no time to reach a health facility. Cultural ceremonies surrounding home deliveries and lack of confidence in health facilities were not reasons for opting for home deliveries (Table 2).

 

As shown in Table 2, there was an association between home delivery and age (p=0.003), educational status and occupation of the husband/wife (mother); p<0.001, and gravidity and parity (p<0.031), but there was no association with the number of wives the husband had (p>0.566). Out of 197 urban mothers interviewed, 49 (24.8%) of them had delivered in health facilities. Moreover, a higher economic status increased the likelihood of health facility delivery; p<0.001 (Table 2). There was no association between economic status and health facility delivery in the rural set up when the rural and urban income level data were analyzed separately viz a viz health facility delivery. Additionally, there was also an association between mothers' educational status and health facility delivery (p<0.001).

 

The choice of TBAs is mostly related to the emergency nature of deliveries. The reasons for mothers' choice of health facility for their delivery were: fear of bad outcome for themselves and the fetus, 42 (45.7%); referral by TBAs, 19 (20.7%); referral by HEWs, 9 (9.8%) and previous risk of adverse outcome during home delivery 16 (17. 4%).

 

Antenatal Care (ANC) Practices

 

Of the 478 women who were asked about their ANC follow up for their last pregnancy, 305 (63.8%) had attended ANC at least once during their pregnancy. For those who did not attend ANC, 48 (50%), as described in Table 3, the main reason for not doing so was mainly related to distance of health facilities and a few, 12 (13%), reported that there was no advantage of ANC.

 

Most of the interviewed mothers, 425 (88.9%), reported that every pregnancy has a risk for the mother and the fetus. As shown in Table 4, there was a significant association between attending ANC and knowing the advantages of ANC (p<0.001), but there was no significant association between knowing pregnancy/delivery danger signs and ANC attendance (p>0.05), except for difficulty of breathing in pregnancy (p=0.03).

 

Concerning the knowledge and skills of HEWs in conducting ANC, 269 (56.3%) of the mothers said that HEWs were superior to TBAs in terms of knowledge and skills; 96 (20.1%) said that TBAs were superior; 64 (13.4%) were not able to compare; 22 (4.6%) said that HEWs and TBAs were not different while 19 (4%) said HEWs could not conduct ANC.

 

Predictors of using health facility delivery

 

A logistic regression model was applied for variables which showed significant association. These variables were gravidity, age of mothers, ANC attendance, occupation and educational status of husband/wife. Education status of wife and ANC attendance among mothers showed significant association from all variables with p=0.036 and p<0.001 levels of significance respectively.

 

 

Most mothers in the study community (83.3%) had delivered at home; less than 25% had delivered in health facilities. The main reason given for home deliveries was that the emergency nature of delivery did not allow the mothers to be transported to health facilities. Most mothers live in the rural areas far from health facilities, and there was no transport system. Studies in other countries have also reported similar findings, whereby distance plays the main part in determining health facility delivery [10,13,16,17]. In this study, home ceremonies during delivery were not found to be determinant factors for home deliveries in this study. Traditional birth attendants are still the main attendants whilst there are HEWs and nurses in some of the study kebeles. The findings show that most mothers believe that HEWs are superior to TBAs in terms of skill; however, a significant proportion also prefers TBAs and some are even doubtful about the skills of HEWs. In a previous qualitative study in the same area, participants reported that they had no confidence in HEWs. The reason being that HEWs' training is short; they are young and have not proven themselves like TBAs [21]. The difference in the reasons given between the two studies may be due to social desirability bias in the quantitative study. Considering that the data collectors came from the government, labeling HEWs as "lower than TBAs" may have been seen to be opposing the government initiative. This is an area which needs intervention in order for the community to develop confidence in HEWs assigned to health posts. One possibility is to bring the labouring mother to the health posts along with the TBA so that both the TBA and the HEWs can assist delivery together. This would give a better chance of a clean delivery at a health post and would enable a referral to be organized if the mother needs one.

 

In this study, we found an association between ANC attendance and delivering in health facilities. In the urban areas, delivery is significantly associated with the income of the household, whilst it is not a factor for the rural. This is probably because in the rural settings, whether the mother has money or not, there is no transport and no time to take a labouring mother to health facilities. Again, as reported in our previous study, it was only when the mother failed to deliver at home that the community took her to a health facility (authors' AMREF unpublished data). This does not require a lot of money, as it is a communal responsibility. In health facilities, delivery is free unless the mother needs to go to a referral hospital wherein the family would incur transport costs.

 

Although there is an association between status of education and health facility delivery, ANC and knowing the advantage of ANC, there was no association between delivery in health facilities, and knowing the danger signs in pregnancy/ delivery [18,19]. This could have been due to the distance of health facilities for most mothers as there was no chance even if they wanted to go because there was no transport system to take the mother to health facilities [12-14]. Thus, there is need to design an affordable and appropriate transport system for the area. Rickshaw ambulances can work for Afar as most of the area is flat. Rickshaws are cheap and their maintenance cost is low. The other option is to organize a camel-pulled cart. Ambulances that can be stationed in health centres to transport mothers from health centres to hospitals are crucial. Hospitals are more than 200 kilometres from the study areas.

 

Based on the findings of this study, several recommendations can be made. The first recommendation is to teach mothers on the importance of ANC and institutional delivery. Along with this, developing the competence of HEWs and equipping health posts are important steps in bringing skilled birth attendance nearer to the community. TBAs are still much preferred by the community for good reasons. There is no way for HEWs to get the community's confidence over TBAs in the short term. We need to understand and communicate the concerns of the community. A middle ground strategy of encouraging TBAs to attend all deliveries in health posts, together with HEWs, should be designed. This way, clean delivery can be offered to the mothers. If the community ambulance system, as mentioned above, is attached to health posts, mothers with difficult labours can be referred in time to health centres. In the towns, deliveries still take place at home and there is need to educate the mothers. Ambulance service for health centres will facilitate the transportation of mothers to hospitals.

 

 

This study reports the rate of home delivery and the determinant factors of institutional delivery in Afar region. Majority of deliveries in the study area took place at home mainly assisted by TBAs. Knowledge of danger signs in pregnancy was insignificant for institutional delivery. However educated mothers and mothers who attended ANC had a higher tendency for health facility delivery.

 

Therefore, there is need to educate mothers on the importance of ANC and institutional delivery. Additionally, developing the competence of HEWs and equipping health posts are important steps in bringing skilled birth attendance nearer to the community. As TBAs are still much preferred by the community for good reasons, it may be difficult for HEWs to get the community's confidence over TBAs in the short term. Importantly, the community's concerns should be communicated to local health system's stakeholders. A middle ground strategy of encouraging TBAs to attend all deliveries in health posts, together with HEWs, should be considered. This is likely to result in increased confidence among community members on health facility delivery; and in the long-term result in better maternal outcomes.

 

 

The authors declare no competing interests.

 

 

Medhanit Getachew contributed in the design of the study, participated in the data collection, performed analysis and interpretation of data and drafted the paper and prepared the manuscript. Kassahun Negash, Jemal Yesouf and Muluken Melese contributed in the design, approved the proposal with some revisions, participated in data analysis and revised subsequent drafts of the paper. All authors read and approved the final manuscript.

 

 

We would like to thank Afar Regional Health Bureau for approving this study and to all mothers who responded to our request to be part of this study.

 

 

Table 1: Characteristics of respondents for home and institutional delivery

Table 2: Socioeconomic Characteristics of the respondents

Table 3: Reasons given for home delivery

Table 4: The relationship between knowing risk factors of pregnancy/delivery and attending ante-natal care

 

 

1. Starrs Andrew. The safe motherhood action agenda: Priorities for the next decade - Report on the Safe Motherhood Technical Consultation, 18-23 October 1997, Colombo, Sri Lanka


2. Thaddeus Sereen, Maine Deborah. Too far to walk: Maternal mortality in context: Findings from a multidisciplinary literature review. Columbia University School of Public Health: Prevention of Maternal Mortality Program, Center for Population and Family Health, 1990


3. Kyomuhendo Grace. Low Use of Rural Maternity Services in Uganda: Impact of Women's Status, Traditional Beliefs and Limited Resources. Reproductive Health Matters. 2003; 11(21):16-26. This article on PubMed


4. Duong DV, Binns CW, Lee AH. Utilization of delivery services at the primary health care level in rural Vietnam. Soc Sci Med. 2004 Dec;59(12):2585-95. This article on PubMed


5. Mrisho M, Schellenberg JA, Mushi AK et al. Factors affecting home delivery in rural Tanzania. Trop Med Int Health. 2007 Jul;12(7):862-72. This article on PubMed


6. Hogan MC, Foreman KJ, Naghavi M, Ahn SY et al. Maternal mortality for 181 countries, 1980-2008: a systematic analysis of progress towards Millennium Development Goal 5. Lancet. 2010 May 8;375(9726):1609-23. This article on PubMed


7. United Nations. Goal 5: improve maternal health. United Nations Millenium Development Goals Available: http://www.un.org/millenniumgoals/maternal.shtml. Accessed 26 April 2010


8. World Health Organization (2004) Neonatal and perinatal mortality: country, regional, and global estimates. Geneva: World Health Organization


9. UNICEF (2008) The State of the World's Children. Available: http://www.unicef.org/sowc08/docs/sowc08.pdf. Accessed 15 May 2011


10. Ronsmans C, Graham WJ; Lancet Maternal Survival Series steering group. Maternal mortality: who, when, where, and why. Lancet. 2006 Sep 30;368(9542):1189-200. This article on PubMed


11. Darmstadt Gary, Bhutta Zulfiqar, Cousens Simon, Adam Taghreed, Walker Neff, de Bernis Luc. Lancet Neonatal Survival Steering Team: Evidence-based, cost-effective interventions: how many newborn babies can we save? Lancet. 2005; 365(9463): 977-988. This article on PubMed


12. Das S, Bapat U, More NS et al. Prospective study of determinants and costs of home births in Mumbai slums. BMC Pregnancy Childbirth. 2010 Jul 30;10:38. This article on PubMed


13. Faye A, Mariama N, Ibrahim B. Home birth in women who have given birth at least once in a health facility: the contributing factors in developing country. Acta Obstet Gynecol Scand. 2011; 4;


14. Rockers Peter, Wilson Mark, Mbaruku Godfrey, Margaret. Source of antenatal care influences facility delivery in rural Tanzania: a population-based study. Matern Child Health J. 2009; 13(6):879-85. This article on PubMed


15. Wahed T, Moran AC, Iqbal M. The perspectives of clients and unqualified allopathic practitioners on the management of delivery care in urban slums, Dhaka, Bangladesh - A mixed method study. BMC Pregnancy Childbirth. 2010 Sep 7;10:50. This article on PubMed


16. Gabrysch S, Cousens S, Cox J, Campbell OM. The Influence of distance and level of care on delivery place in rural Zambia: A study of linked national data in a Geographic Information System. PLoS Med. 2011 Jan 25;8(1):e1000394. This article on Pubmed Central


17. Gabrysch Sabine, Campbell Oona. Still too far to walk: Literature review of the determinants of delivery service use. BMC Pregnancy and Childbirth. 2009; 9:34. This article on PubMed


18. Paul BK, Rumsey DJ. Utilization of health facilities and trained birth attendants for childbirth in rural Bangladesh: an empirical study. Soc Sci Med. 2002; 54(12):1755-65. This article on PubMed


19. Idris Suleman, Gwarzo UMD, Shehu AU. Determinants of Place of Delivery among Women in a Semi-Urban Settlement in Zaria, Northern Nigeria. Ann of Afric Med. 2006; 5 (2): 68-72


20. Central Statistics Agency (CSA). Population and Housing Census of Ethiopia, United Nations Population Fund (UNFPA), 2008: 59-60


21. Yayehyirad Kitaw, Yemane Ye-Ebiyo, Amir Said et al. Assessment of the Training of the First Intake of Health Extension Workers. Ethiopia Journal of health Development. 2007; 21(3):223-239