1. Introduction

Social determinants of maternal mortality are the biological, physical, socioeconomic, and local cultural context in which women are born, live, work and have children. These factors account for the differential maternal morbidity, mortality and disability within and between the various strata of the population as a result of pregnancy and its related complication in a given place and time. In 2013 alone, an estimated 289,000 women died as a result of pregnancy and its related complications with over 99 percent (286,110) of these deaths accounted by developing countries ^[1]. The Maternal mortality ratio (MMR) for Nigeria is very high and ranked amongst the highest in the world. Official figure put the MMR as 545 maternal deaths per 100,000 live births ranging from 475 to 615deaths per 100,000 live births ^[2]. This figure is far below the 820 deaths per 100,000 live births reported by UNICEF ^[3] and the 1000 deaths per 100,000 live births reported by community based study ^[4]. This is an indication of underestimation, since records of vital events are incomplete and the sensitivity of surveillance on maternal deaths is low in two thirds of countries (Nigeria inclusive) in the WHO African Regional office ^{[5, 6]}. The reason for the fewer cases of maternal deaths in the developed countries is largely as a result of high access and utilization of maternal health services such as ante natal, natal, post-natal and family planning services.

Place of delivery and the availability of skilled health workers are among the major factors that were known to be associated with or influencing the risk of a woman dying as a result of pregnancy and its related complications ^[7]. In Nigeria, only 38% of the estimated 8.3 million annual pregnancies in 2010 were attended by skilled health workers during delivery ^[5]. It is pertinent to note that there exist wide disparities within and between the six geopolitical zones of Nigeria with the North east zone having the worse maternal health indicators ^[2]. Furthermore, Yobe state (study area) has the lowest proportion of deliveries attended by skilled health workers of 9.3% compared to the national and regional average of 38% and 15.5..% respectively ^{[2, 8]}. Considering the fact that more than 70% of the population of Yobe state live in rural and desert areas with poor road network, the proportion of skilled deliveries is likely to be lower than the state average noted above. A plausible reason been that the health facilities in the state lack adequate number of skilled health care workers, equipment’s and essential medications ^{[2, 8, 9]}. This is buttressed by the fact that there are only 45 certified midwives spread across the 523 public health facilities who are supposed to attend to the over half a million pregnancies per year ^[8]. This means that the average workload of a midwife per day will be 50 pregnant women which without doubt, will compromise the time available for health worker – client communication and invariably may result in poor quality of services rendered as was similarly observed in some parts of Sokoto state ^[10].

In order to reposition the state plan, there is the need to identify what are the barriers and challenges at individual, community and health facility levels using the Anderson health behavior model ^[11]. The model has three constructs a) Predisposing characteristics; b) enabling characteristics and c) need characteristics ^[12]. The model provides a systematic approach to look at conditions, processes and relationships at various levels of the society that defines the autonomy of individuals, how it affects the pattern of accessing and utilizing health services and the distribution of health outcomes. The usefulness of the model is underscored by its wide application in research of social and public health importance ^{[13, 14, 15]}. This study aimed to identify whether women’s individual, community and health system factors are associated with the place of delivery in Yobe state.

2. Methodology

Yobe state is one of the 36 states in Nigeria, located in the North eastern part of the country. It shares common boundaries with Niger Republic to the North, Borno state to the east, Bauchi state to the west and Gombe state to the south. The state has an estimated population of 3 million with over 70% living in rural areas and a female and male literacy rate of 12% and 32% respectively ^[8].

This is a descriptive cross sectional study.

The 2008 National Demographic and Health Survey (2008 NDHS) data provides information to assess the influence of social, economic, cultural and political factors on the use of modern delivery services. Respondent’s bio-socio-demographic and economic characteristics were the independent variables and the place of delivery (home or health facility) as the dependent variable.

The data base contained 695 women whose had delivered between January 1, 2003 to December 31, 2008. Participants were recruited proportionate to population density, rural/urban ratio and composition from all parts of the state, ^[2]. The sampling strategy was a stratified two stage cluster sampling with the 2006 Nigerian national census enumeration areas as cluster. Sampling at lowest administrative unit is considered ideal as it will strengthen the external validity and generalizability of findings and statistical precision that will demonstrate acceptable effect size ^{[2, 8]}. Details on sample size estimation, sampling strategy and method of data collection are presented in the survey report and other relevant documents ^{[16, 17]}.

The data was analyzed using SPSS version 16. Bivariate Pearson’s Chi-square test statistic and two stages of Multivariate regression analysis were conducted in order to identify factors influencing the utilization of delivery services in Yobe state. Variables that were included the predictive model were either known influencers from previous research and or were found to have significant statistical association from bivariate Pearson Chi-square statistic. The critical value was set at 95% confidence interval. A similar approach was used to identify how religion influence none, partial and complete childhood immunization ^[19]. Furthermore, Fapohunda & Orobaron (2014) have used 2008 NDHS data and similar analytical approach to identify factors associated with women who deliver with no one present ^[20].

Access to the raw data was granted by ORC Macro and ICF International based in Calverton Maryland, USA who are the custodian of the Demographic and Health Survey data.

3. Results

Out of the 695 women aged 15-49 years who were interviewed, only 659 were included in the analysis representing 6.2% missing data. A total of 44 (6.7%) out of the 659 respondents had their last delievery in a health facility with home deliveries accounting for 93.3%.

Table 1: Majority (60.1%) of the respondents are between 20 – 34 years old with a mean age of 26.3 ± 2.4 years.

Table 1. Biological and behavioural characteristics of respondents

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View next table

The extreme of ages (<20 years and >34 years) had the lowest proportion of women whose last delivery was conducted in a health facility accounting 5.7% and 4.1% respectively. Majority of the women are grand multi-para (47.5%) and had a birth interval between the last two deliveries of 24 - 48 months (57.9%). The highest proportion of institutional deliveries (12.8%) was recorded among women having their first delivery.

Table 2: Majority of the women belonged to Kanuri/Baribari, Hausa, and Fulani ethnic groups (57.7%), Muslims (99.1%), lacking any form of autonomy (90.4%) and rural dwellers (73%). The proportion of institutional delivery is higher among urban dwellers (27.2%), Yoruba and Igbo ethnic groups (67%), Christians (50%) and among women having at least some form of autonomy (36.4%) compared to their respective counterparts.

Table 3: Majority of the respondents had no formal education (82.8%) and belonged to the poorest wealth quintile (52.4%), and only 4% and 1.3%, respectively of these women that have delivered in a health facility.

Table 4: Higher proportion of institutional delivery was observed among women living near a health facility (13.4%), availability of skilled health workers in the nearest health facility (63.2%) and those who reported to have received health talks during ANC visits.

Table 2. Socio-cultural characteristics of respondents

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

View next table

Table 3. Economic characteristics of respondents

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

View next table

Table 4. Characteristics of respondents health system environment

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

View next table

Table 5. Association between women’s biological characteristics and place of delivery

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

View next table

Variables that have positive impact on the place of delivery include parity of ≥5 (AOR = 1.88; CI 0.38 – 8.73) (Table 5), Christians (AOR = 6.20; CI 3.24 – 8.41) (Table 6), living in urban areas (AOR = 2.03; CI 0.41 – 10.21) (Table 6), women having full autonomy (AOR = 7.06; CI 3.37 – 13.67) (Table 6), women belonging to the rich wealth quintile (AOR = 3.24; CI 0.23 – 45.71) (Table 7), women who had more than secondary level of education (AOR = 7.41) (Table 7), women whose husband had more than secondary level of education (AOR =12.06; CI 0.19 – 75.73)) (Table 7), distance to the nearest health facility (AOR = 6.31; CI 1.29 – 3.08) (Table 8), availably of skilled health workers (AOR = 3.31; CI 1.15 – 4.49) (Table 8), and access to the media as a source of information on MHS (AOR = 11.71) (Table 8).

Table 6. Association between women’s sociocultural characteristics and place of delivery

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

View next table

Table 7. Association between women’s socioeconomic and place of delivery

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

View next table

The model contained six predictive variables (family wealth quintile, religious affiliation, highest educational attainment, availability of skilled health workers, and distance to health facility). These variables were selected based on the result of the bivariate Pearson Chi square test, binary logistic regression and or known theoretical facts. These variables were found to be statistically significant χ² (16, N=659) 227.774; P< 0.001, indicating that, the model was able to distinguish between participants who delivered at home and those who had delivered in a health facility. The variables accounted for 30.1% (Cox & Snell R Square) and 78.1% (Nagelkerke R Square) of variability among participants, correctly classified 71% of cases and together with Hosmer and Lemeshow goodness of fit test indicated the model being useful (p = 0.59) since the p-value is larger than the alpha level ^[21].

Table 8. Association between women’s enabling factors and Place of delivery

Download as

PowerPoint Slide

Larger image(png format)

• Tables index

Veiw figure

View current table in a new window

View previous table

4. Discussions

Yobe state has not been able to achieve the Millennium Development Goal 5 (MDG 5) as set in the state health strategic plan for 2010 – 2015 ^[8], because as of 2012, the state was reported to have an MMR of about 1100 per 100, 000 live births in 2012 ^[4]. However, while it is important to note that this figure is lower than what was recorded in 2008 (1,549 per 100, 000 live births), it is still higher than the national average of 820 per 100, 000 live births in 2012 ^[3].

In this study, the proportion of institutional delivery is 6.7% which is lower than the figures cited in the state health strategic plan for 2010 – 2015 ^[8] and five folds lower than the National average of 38% as reported by the WHO ^[5].Thus, the state seem to be under performing even within the North east geopolitical zone of Nigeria since the zonal average stands at 15.5% ^[2]. While several studies have reported ethnicity and religion as the fundamental reasons for the underutilization of Maternal Health Services (MHS), however, it must be pointed out that the state has very low number of skilled health workers with only 45 midwives in public facilities to handle more than the annual estimated half a million pregnancies ^{[2, 8]}. Health facilities were largely lacking basic equipment’s and consumables as contained in the state health strategic plan for 2010 – 2015 ^{[2, 8]}. Moreover, states in the North east geopolitical zone were reported to be having the lowest number of health facilities and staffing which is compounded by maldistribution of health resources ^{[9, 22]}. Furthermore, this weakness was recently reported by Lembani and his colleagues (2014), that the insurgence by terrorist Islamic fundamentalist (Boko Haram) in the past 10 years has further resulted in the destructions and closer of many health facilities, with many skilled health workers preferring to work in areas free from the activities of the terrorist group ^[22]. This reinforces the fact that, social and economic services (education, agriculture, commerce, security etc) are equally disrupted which affected geographical and economic accessibility and the utilization of modern health facilities as many has to live in internally displaced camps that could not adequately address the challenges of pregnancy and its related complications. Furthermore, the fact that the vast majority of the respondents were in the poor/poorest wealth quintile (77.4%); had no formal education (82.8%), living in rural areas (73%), has no health insurance (99.8%) and lack any autonomy (90.5%) in terms of ability to independently take decision on health issues, freedom to go out of their matrimonial homes without seeking permission and not gainfully employed will further present as geographical, economic and cultural barriers to utilization of MHS. This scenario has been similarly attributed as one of the major scenarios that affect the utilization of antenatal, natal and post-natal services in other parts of Nigeria ^{[4, 20, 23, 24, 25]} and other low income countries ^[26]. Although ethnicity and religion are plausible reasons for the lack of autonomy which may results in delays to take decision and reaching the hospital in good time, however, the 2008 DHS data is just quantitative and therefore does not provide insights as to what has transpired at home or on the way to health facility and therefore indicate the need for qualitative studies. Moreover, a study in southwestern Nigeria, among the Yoruba ethnic group who are largely Christians, has further indicated that the role of ethnicity and religion is inconsistent with our finding as intentional home deliveries that were conducted by unskilled health workers accounted for nearly 70% of all deliveries reviewed ^[27]. Reasons cited include nearness, lack of transportation, and cost indicating the root causes as factors of social exclusion and economic barriers rather than religion and or ethnicity. Overall, the present study and reasons cited by the study in the south western Nigeria further reinforces the fact that optimal utilization of MHS depends on geographical and economic accessibility ^[27]. It further demonstrate the relevance of the WHO position that the “toxic combination of bad policies, economics, and politics is in large measure, responsible for the fact that a majority of people in the world do not enjoy the good health and that Primary health care, which integrates health in all of government's policies, is the best frame work for doing” ^[28]. Political will and commitment (policy development, implementation and monitoring) influences the distribution of resources and bear on individual, family and community members ability to have access (geographical, economic, and cultural) to qualitative health care services. Economic policies such as liberalization of trade, structural adjustment program, commercialization and privatization of public utilities and outright withdrawal of subsidies on key sectors of the economy (agriculture, education, water and health) resulted in higher unemployment and poverty, thereby affecting individual capacity to access health care services. For instance, studies in Nigeria reported that, there was a higher maternity utilization before the introduction of Structural Adjustment Programme in 1988, when health care was free or highly subsidized and a subsequent reversal when fees were introduced ^{[29, 30]}.

5. Limitations of the Study

Since the study design is cross sectional, it only provided insights on the factors associated with the use of delivery services and not causative agents. The DHS data used for the study does not contain information (qualitative data) on reasons for delays in decision making to seek for modern health services, delays to reach a functional health facility and delay in commencing appropriate services, which could have provided better understanding of the root causes of home deliveries.

6. Conclusions and Recommendations

The use of Maternal Health Services (MHS) is influenced by factors operating at individual, community and health system levels. Family wealth quintile, religious affiliation, highest educational attainment, availability of skilled health workers, and distance to health facility has shown to have positive impact on the use of institutional delivery services. Family wealth income, availability of skilled health worker and proximity to health facility demonstrated significant statistical association with the place of even after controlling for covariates like age, ethnicity, husbands education attainment, parity, religion and frequency of ANC visits.

There is the need to consider engaging other sectors like education in order to increase female literacy rate which in the long term will enhance women autonomy in decision making regarding their health, economic independence, improved health cultural capital and better access and utilization of MHS. In the short term government and partner agencies may consider redistributing health resources based on need. Furthermore, studies on the role of religion and ethnicity on the use of MHS and how based to engage traditional and religious leaders need to be conducted in order to develop local strategies to enhance female autonomy and use of MHS.

Source of Funding or Support

None.

Competing Interest

The authors declare that they have no competing interests.

Authors’ Contribution

The authors have both contributed in the design, analysis and writing of the article.

Acknowledgement

We would like to acknowledge ORC Macro and ICF International Calverton Maryland, USA for granting us the permission to use the 2008 NDHS dataset.

References

[1]	World Health Organization. World Health Statistics 2014. WHO, Geneva, Switzerland. 2014; p71-114. Available: http://www.who.int/gho/publications/world_health_statistics/2014/en/ [Accessed July 15, 2015].
	In article
[2]	National Population Commission Nigeria and ICF Macro. Nigeria Demographic and Health Survey 2008. Calverton, MD USA: National Population Commission and ICF Macro. 2009; p14-141. Available: http://pdf.usaid.gov/pdf_docs/PNADQ923.pdf. [Accessed March 10, 2014].
	In article
[3]	UNICEF. State of the world’s children 2012. UNICEF, New York, USA. Available: http://www.unicef.org/sowc2012/pdfs/SOWC-2012-Main-Report_EN_21Dec2011.pdf. [Accessed March 10, 2014].
	In article
[4]	Doctor HV, Findley SE, Afenyadu GY. Estimating maternal mortality level in rural northern Nigeria by the sisterhood method. International Journal of Population Research, 2012;5.
	In article		View Article
[5]	World Health Organization. World Health Statistics 2013. WHO, Geneva, Switzerland. 2013; p65-141. Available: http://www.who.int/gho/publications/world_health_statistics/2013/en/. [Accessed March 10, 2014].
	In article
[6]	World Health Organization. Maternal death surveillance and response: Technical guidance. Information for action to prevent maternal death. WHO, Geneva, Switzerland. 2013; p24-61. Available: http://www.who.int/maternal_child_adolescent/documents/maternal_deathsurveillance/en/. [Accessed March 10, 2014].
	In article
[7]	World Health Organization, United Nations Children’s Fund, United Nations Population Fund & The World Bank. Trends in Maternal Mortality: 1990 to 2008 – Estimates developed by WHO, UNICEF, UNFPA, and the World Bank’, WHO, Geneva, Switzerland. 2010. Available: http://whqlibdoc.who.int/publications/2010/9789241500265_eng.pdf. [Accessed July 15, 2014].
	In article
[8]	Yobe State Government. Strategic Health Development Plan 2010 – 2015. Yobe State Ministry of Health. 2010; p12-14. Available: http://www.mamaye.org/sites/default/files/evidence/Yobe%20Naration2.pdf. [Accessed March 5, 2015].
	In article
[9]	Health Reform Foundation of Nigeria. Primary Health Care in Nigeria: 30 years after Alma-Ata. The Nigerian Health review 2007; Health Reform Foundation of Nigeria, Mmm-nat Educational Window Consultant Ltd. 2007.
	In article
[10]	Oche M, Umar AS, Adejoh MA, Isa HO, Ekele BA. Determinants of utilization of maternity services in Gidan Igwe, Sokoto, Nigeria. Sahel Medical Journal, 2010; 13(3):128-134.
	In article
[11]	Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Social Behavior, 1995; 36:1-10.
	In article		View Article  PubMed
[12]	Andersen RM, Newman JF. Societal and individual determinants of medical care utilization in the United States. The MilBank Quarterly, 2005; 83(4):1-28.
	In article		View Article  PubMed
[13]	Caprio S, Daniels SR, Drewnowski A, Kaufman FR, Palinkas LA, Rosenbloom AL, Schwimmer JB. Influence of Race, Ethnicity, and Culture on Childhood Obesity: Implications for Prevention and Treatment a consensus statement of Shaping America's Health and the Obesity Society. Diabetes Care, 2008; 31(11): 2211-2221.
	In article		View Article  PubMed
[14]	Friedlander Y, Manor O, Paltiel O, Meiner V, Sharon N, Calderon R, Hotchner H, et al. Birth weight of offspring, Maternal pre-pregnancy characteristics, and mortality of mothers: the Jerusalem perinatal study cohort. Ann Epidemiol. 2009, 19(2):112-7.
	In article		View Article  PubMed
[15]	Ogden CL, Carroll MD, Curtin LR, Lamb MM, Flegal KM. Prevalence of high body mass index in U.S. children and adolescents, 2007–2008. JAMA, 303(3): 242-249.
	In article		View Article  PubMed
[16]	Smith AK, Ayanian JZ, Covinsky KE, Landon BE, McCarthy EP, Wee CC, Steinman MA. Conducting high-value secondary dataset analysis: An introductory guide and resources. J Gen Intern Med. 2011; 26(8): 920-929.
	In article		View Article  PubMed
[17]	Szklo M, Nieto FJ. Understanding a Lack of Validity: Bias in Epidemiology: Beyond the basics (3rd ed.). Sudbury, MA: Jones and Bartlett. 2014; p8-24.
	In article
[18]	United State Agency for International Development & ICF Macro. Guide to DHS Statistics: Demographic and Health Survey Methodology. Calverton, Maryland, USA: USAID and ICF Macro. 2006. Available: http://dhsprogram.com/pubs/pdf/DHSG1/Guide_to_DHS_Statistics_29Oct2012_DHS1pdf. [Accessed April 10, 2015].
	In article
[19]	Antai D. Faith and child survival: The role of religion in childhood immunization in Nigeria. Journal of Biosocial Science, 41, 57-76.
	In article		View Article  PubMed
[20]	Fapohunda, B., & Orobaron, N. (2014). Factors influencing the selection of delivery with no one present in Northern Nigeria: implications for policy and programs. Int J Womens Health., 2014; 6: 171-183.
	In article		View Article  PubMed
[21]	Field D. Discovering statistics using SPSS (2nd Ed). Chapter, 8, “Logistics regression” Thousand Oaks, California; Sage Publications ltd. 2009; p. 264-268.
	In article
[22]	Lembani M, Mohammed A, Abdulwahab A, Garba A, de Pinho H, Delobelle P, Zarowsky C, Ager A. Health Systems Resilience: A Systems Analysis -A Case Study of Health Service Provision in Yobe State, Nigeria in the Context of the Boko Haram Insurgency. Final Report, October 2014. Rebuild Consortium and Mailman School of Public Health, Columbia University, USA. Available: http://www.rebuildconsortium.com/news/documents/HealthSystemsResilience_YobeCaseStudy_ FinalizedReport.pdf. [Accessed March 5, 2015].
	In article
[23]	Babalola S, Fatusi A. Determinants of use of maternal health services in Nigeria – looking beyond individual and household factors. BMC Pregnancy and Childbirth, 2009; 9, 43.
	In article		View Article  PubMed
[24]	Doctor HV, Bairagi R, Findley SE, Helleringer S, Dahiru T. Northern Nigeria, Maternal, Newborn, and Child Health Program: Selected analysis from population based baseline study. The Open Demographic Journal, 4: 11-21.
	In article
[25]	Fawole AO. Predictors of maternal mortality in institutional deliveries in Nigeria. Afr Health Sci. 12(1): 32-40.
	In article		PubMed
[26]	Coast E, McDaid D, Leone T, Pitchforth E, Matthews Z, Lemmi V, Jones E. What are the effects of different models of delivery for improving maternal and infant health outcomes for poor people in urban areas in low income and lower middle income countries? Department for International Development, London. 2012. Available: http://eprints.lse.ac.uk/41908/1/What_are_the_effects_of_different_models_of_delivery_for_improving _maternal_and_infant_outcomes_for_poor_people_in_urban_areas_in_low_income_and_lower_middle_ income_countries_%28LSERO%29.pdf. [Accessed March 5, 2015].
	In article
[27]	Adelaja LM. A survey of home delivery and newborn care practices among women in a suburban area of western Nigeria. Obstet Gynecol. 2011; 983542, 9.
	In article		View Article
[28]	World Health Organization. Inequities are killing people on a “grand scale”: The Reports WHOs commission on health inequalities, News Release WHO/29, August 28, WHO, Geneva, Switzerland. 2008. Available: http://www.who.int/mediacentre/news/releases/2008/pr29/en/index.html. [Accessed April 25, 2014].
	In article
[29]	Otolorin EO, Marinho AO, Ojengbede O, Odukoya AO. Palmer CK. Maternity Care Monitoring. A constant at two Levels of Health Care delivery in Ibadan Nigeria, Int. J. Gynaec and Obstet. 1988; 26:367.
	In article		View Article
[30]	Ekwempu CC, Maine D, Olorukoba MB, Esseien ES, Kisseka MN. Structural Adjustment and Health in Africa –The Lancet, 1990; 336: 56-57.
	In article		View Article