This paper has attempted to assess inequities in access to basic maternal health interventions with a view to identifying constraints that may impede progress towards the MDG 5 target of reducing maternal mortality. The findings demonstrate that in the absence of targeted interventions, the achievement of MDG 5 target will be difficult.
The findings indicate significant spatial inequalities in the utilization of basic maternal health interventions. This is in line with previous studies [2, 5]. Use of skilled providers for child delivery services is much below the national average in three regions: Kavango, Kunene and Ohangwena. The rate in other four regions is also less than the national average of 81.4%, although it is better than the three regions mentioned above. Thus, it is evident that resource allocation decisions based on the national average figures may not lead to appropriate targeting of the scarce maternal health resources and that the regions who need it more may not receive what is due to them to achieve the target of reducing maternal mortality.
The inverse relationship between delivery by skilled attendants and maternal mortality has been well-established . Thus, to reduce the increasing levels of maternal mortality ratio and expedite progress towards the MDG 5 target, it is necessary to focus all efforts and interventions towards those regions that have worse-off indicators in terms of access to maternal health interventions.
The regions that have lower coverage of the basic maternal interventions are also the ones that have low human development index. Hence, to improve maternal health and promote equity, it is necessary to launch a multi-sectoral action that also addresses the social determinants of health such as poverty and levels of education in line with the recommendations of the Commission on Social Determinants of Health.
The rate of caesarean section in the region with the highest rate (Khomas) is 13 times more than that of Caprivi with the lowest rate. Three regions - Caprivi, Kunene and Ohangwena - have rates that are less than 5%. In contrast three regions - Erongo, Khomas and Omaheke - have caesarean section rates well above 15%. This indicates that there are regions of under-coverage as well as region where there is over-provision of caesarean section. Although there is a debate, a population-based rate of 5-15% has been considered as the acceptable level for cesarean section to ensure the best outcomes for mothers and children . The proportion of deliveries by cesarean sections in a geographical area is a measure of access to and use of obstetric emergency care for averting maternal and neonatal deaths. Therefore, it is evident that there is limited access to comprehensive emergency obstetric care in a third of the regions mentioned that requires urgent action.
Education-related inequalities in the rate of cesarean section and delivery by skilled providers are more pronounced that those of antenatal care by skilled providers. There is an excess utilization of cesarean section among women with post-secondary education. This is more than twice the threshold for the acceptable limit of cesarean section. Almost all women with post-secondary education are delivered by skilled providers compared to about half of those with no education. Women who are primary school complete use skilled providers for delivery 60% more than those with no education. Women's education influences health outcomes through a variety of channels including health-seeking behaviors and earning opportunities. In Namibia, income poverty has a very wide variability by educational groupings. The 2003/2004 household income and expenditure survey indicates that while the incidence of poverty among those without formal education is 41.4%, it is only 0.5% among those with tertiary education . These may influence the demand for maternal health services. As stated above, it is therefore necessary to improve mother's level of education and its correlates in order to bridge the inequity and improve uptake of essential maternal health interventions.
A significant rural-urban inequality in the use of interventions is also observed. Delivery by cesarean section, antenatal care in a private facility, antenatal care by a doctor, delivery by a doctor, delivery in a private facility and delivery by a skilled attendant demonstrate inequalities in favour of the rich. Use of private facilities and doctors for antenatal care and delivery and cesarean section is lowest in rural areas. In contrast, rural women use more of delivery by traditional birth attendant, delivery by a relative/other, delivery at home. However, antenatal care by a skilled provider does not show any significant inequalities between urban and rural areas, which may indicate that although mothers access health facilities for ANC, only a proportion of these offer delivery services. Women in urban areas also make more use of delivery services in public facilities compared to those in rural areas. This may perhaps be linked to the levels of poverty and educational status, which also manifest rural-urban disparity.
The concentration curves and indices indicate significant pro-wealthy inequalities in antenatal care and delivery by skilled providers; delivery in public and private health care facilities; delivery by cesarean section; and post-natal checkup. In contrast concentration curves and indices for antenatal care and delivery at home have a pro-poor bias.
Addressing the poor-rich inequalities in delivery care by skilled attendants is essential for achieving the MDGs for maternal health . However, the low rate of delivery by skilled personnel in some of the regions, among the less educated, rural areas and the less wealthy is of concern, as the groups where most of the improvements in maternal health are expected have a limited access. This will lead to a slow progress in halting the current trend of increasing maternal mortality ratio and achievement of the MDG 5 targets.
The case of cesarean section is also another area of concern, as conditions that require comprehensive emergency obstetric care are major causes of maternal mortality. Under-provision among the poor, rural areas, less educated and regions with low human development index and over-provision among the wealthiest, more educated, urban areas and regions with high human development signal the need for re-allocation and targeting of the available resources in order to make a significant contribution to the reduction of maternal mortality in line with the MDG 5 target.
The inequities observed in this study may be explained by demand and supply side factors [3, 23]. For example mother's education may affect health-seeking behaviour and together with household wealth may also constrain the demand for services. The NDHS 2006-07 indicates that almost all women pay for delivery mainly in cash and to a lesser extent in kind. About 86% of women who had live births in the five years preceding the survey paid in cash . For the majority (85%), the payment was less than 50 Namibian dollars (about US$ 7 at the then prevailing exchange rate). Adding to this the indirect costs that the women and those accompanying them are likely to incur (e.g. transport cost), payment for deliveries could be a barrier to use of delivery services by trained providers for those poorest segments of the population. It is therefore worthwhile to revisit the policy of charging women for delivery services and possibly make blanket exemptions in those regions where the poverty levels/HDI are the lowest in order to increase uptake of interventions caused by demand side factors. It should also be noted that the revenue generated from these payments for delivery is a very negligible fraction of the total government expenditure on health  and that it doesn't play a significant role in terms of revenue generation.
The increasing trend in the maternal mortality ratio may also be related to the supply of services. In 2005/06, only 11.8% of health facilities provided comprehensive emergency obstetric care, which in addition to those services under basic emergency obstetric care, includes blood transfusion and the provision of caesarean section. Furthermore, no health centres provided basic emergency obstetric care .
With per capita expenditure on health of US$ 276 in 2006, the country is in a better off position than most countries in the African region, where the average per capita expenditure on health in 2006 was US$ 58 . Therefore it is necessary to address possible allocative inefficiency, where resources may be allocated in purchasing the non-optimal mix of inputs and/or producing the non-optimal mix of outputs. The non-provision of basic emergency obstetric care at the health centre level may, among other things, imply the presence of regulatory frameworks that do not allow junior level health workers to provide the basic services (signal functions). To halt the upward trend in maternal mortality ratio, bottlenecks related to policy and regulatory frameworks have to be addressed and issues of possible task-shifting be explored.
Inequality in the distribution of health care inputs is also one of the important factors that may contribute to the limited access to essential maternal health services, consequently leading to increase in the maternal mortality ratio. The Namibia health and social services system review indicated a health worker density of 3 per 1,000 population , which is well above the estimated minimum level of health workforce density of 2.5 per 1000 population required to achieve 80% coverage of immunization and delivery by skilled attendants . However, a breakdown of this national figure indicates that while the health workforce density in the private sector was 8 per 1000 (high density), in the public sector it was only 2 (low density), which is below the threshold stated above.
Explaining inequities to maternal health interventions in terms of demand and supply fits well with the three delays model . This model proposes three barriers to accessing maternal health services: (i) delay in decision to seek care; (ii) delay in getting to the facility; and (iii) delay in getting the appropriate care once at the facility. The first two delays are demand-side barriers, which may be affected by mother's education, household wealth and community-level factors such as the levels of poverty, which also have a bearing on the intra-household resource allocation and inequities. It can be discerned that the causes of inequities in the utilization of maternal health interventions are those that may explain the delays model. However, this study recommends that the causes of the inequities be identified using a decomposable concentration index  in order to target resources at the root causes of the inequities.
In the presence of inequities, achievement of the MDG and other national and international targets becomes elusive. The segment of society that has more need is left out, thus impeding progress towards the cherished goals. Some of the access gradients observed, including educational status and geographical location, lie outside the health sector. Hence, addressing inequities in access to maternal health services should not only be seen as a health systems issue. The social determinants of health have to be tackled through multi-sectoral approaches in line with the principles of Primary Health Care  and the recommendations of the Commission on Social Determinants of Health . The above-mentioned factors that are outside the health sector can be tackled through action points within the three overarching recommendations of the Commission on Social Determinants of Health . To this end, it is necessary to assess the contribution of each of the above determinants to the overall inequality in access to the various maternal health interventions.