• maternal mortality ratio;
  • intervention;
  • prenatal care;
  • pregnant women
  • taux de mortalité maternelle;
  • intervention;
  • soins prénataux;
  • femmes enceintes
  • tasa de mortalidad materna;
  • intervención;
  • cuidados prenatales;
  • mujeres embarazadas


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Objectives  To understand the utilisation of prenatal care and hospitalised delivery among pregnant Muslim women in Ningxia, China, and to explore the effectiveness of the integrated interventions to reduce maternal mortality.

Methods  Cross-sectional surveys before and after the intervention were carried out. Using multistage sampling, 1215 mothers of children <5 years old were recruited: 583 in the pre-intervention survey and 632 in the post-intervention study. Data on prenatal care and delivery were collected from face-to-face interviews. Maternal mortality ratio (MMR) data were obtained from the local Maternal and Child Mortality Report System.

Results  After the intervention, the MMR significantly decreased (45.5 deaths per 100 000 live births to 32.7 deaths). Fewer children were born at home after the intervention than before the intervention (OR, 0.11; 95% CI, 0.08–0.15). The proportion of women who attended prenatal care at least once increased from 78.2% to 98.9% (OR, 24.55; 95% CI, 11.37–53.12). The proportion of women who had prenatal visit(s) in the first trimester of pregnancy increased from 35.1% to 82.6% (OR, 8.77; 95% CI, 6.58–11.69). The quality of prenatal care was greatly improved. Effects of the intervention on the utilisation of maternal care remained significant after adjusting for education level and household possessions.

Conclusions  The findings suggest that integrated strategies can effectively reduce maternal mortality.

Objectifs:  Comprendre l’utilisation des services de soins prénataux et l’accouchement à hôpital chez les femmes musulmanes enceintes de Ningxia, en Chine et étudier l’efficacité des interventions intégrées visant à réduire la mortalité maternelle.

Méthodes:  Des surveillances transversales avant et après l’intervention ont été réalisées. En utilisant un échantillonnage à plusieurs niveaux, 1215 mères d’enfants de <5 ans ont été recrutées, 583 dans l’enquête pré-intervention et 632 dans l’étude post-intervention. Les données sur les soins prénataux et de l’accouchement ont été recueillies par des interviews de face-à-face. Les données sur le taux de mortalité maternelle (TMM) ont été obtenues à partir du système local de report de la mortalité maternelle et infantile.

Résultats:  Après l’intervention, le TMM a diminué significativement (45,5 à 32,7 décès pour 100.000 naissances vivantes). Moins d’enfants sont nés à la maison après l’intervention qu’avant l’intervention (OR: 0,11; IC95%: 0,08 à 0,15). La proportion de femmes qui ont reçu au moins une fois des soins prénataux a augmenté de 78,2%à 98,9% (OR: 24,55; IC95%: 11,37 à 53,12). La proportion de femmes ayant eu au moins une consultation prénatale dans le premier trimestre de la grossesse est passée de 35,1%à 82,6% (OR: 8,77; IC95%: 6,58 à 11,69). La qualité des soins prénataux a été fortement améliorée. Les effets de l’intervention sur l’utilisation des services de soins de santé maternelle restaient significativement importants après ajustement pour le niveau d’éducation et les biens du ménage.

Conclusions:  Les résultats suggèrent que les stratégies intégrées peuvent efficacement réduire la mortalité maternelle.

Objetivos:  Entender la utilización de cuidados prenatales y partos hospitalarios entre mujeres musulmanas embarazadas en Ningxia, China, y explorar la efectividad de las intervenciones integradas para reducir la mortalidad materna.

Métodos:  Se llevaron a cabo estudios croseccionales antes y después de las intervenciones. Utilizando un muestreo multietapa, se reclutaron 1215 madres de niños <5 años, 583 en el estudio pre-intervención y 632 en el estudio post-intervención. Se recolectaron datos sobre los cuidados prenatales y el parto a partir de entrevistas cara a cara. Los datos sobre la tasa de mortalidad materna (TMM) se obtuvieron del sistema local de vigilancia de mortalidad materno infantil.

Resultados:  Después de la intervención, la TMM disminuyó significativamente (45.5 muertes por 100,000 nacidos vivos a 32.7 muertes). Nacieron menos niños en casa después de la intervención que antes de ella (OR 0.11, 95%IC 0.08 a 0.15). La proporción de mujeres que recibieron cuidados prenatales al menos una vez aumentó del 78.2% al 98.9% (OR 24.55, 95%IC 11.37 a 53.12). La proporción de mujeres que tuvieron visita(s) prenatales durante el primer trimestre del embarazo aumentó del 35.1% al 82.6% (OR 8.77, 95%IC 6.58 a 11.69). La calidad de los cuidados prenatales había mejorado en gran medida. Los efectos de la intervención sobre la utilización de los cuidados maternos continuaban siendo significativos después de ajustar para el nivel de educación y el nivel de riqueza del hogar.

Conclusiones:  Los hallazgos sugieren que las estrategias integradas pueden reducir efectivamente la mortalidad materna.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

The maternal mortality ratio (MMR) is not solely the most fundamental indicator for women and children’s health, but an internationally acknowledged indicator to reflect a country’s social and human development. Through the fifth goal of the eight Millennium Development Goals (MDG5), countries have committed to reducing the MMR by three quarters between 1990 and 2015 (Lozano et al. 2011), but progress towards reaching MDG5 has been slow (Hill et al. 2007; Chatterjee & Paily 2011) and must be accelerated.

China has the largest population in the world. Although the MMR in China decreased from 95 deaths per 100 000 live births in 1990 to 47.7 deaths in 2005 (WHO 1999, 2007), the challenge to accomplish MDG5 persists. According to the annual report of health statistics in China, the MMR in rural areas of China was 53.8 deaths per 100 000 live births in 2005, much higher than that of 25.0 deaths in urban areas (MOH China, 2005). Therefore, promoting maternal health in rural China, especially poor and remote areas, has been top of the agenda of the Chinese government. Ningxia Hui Autonomous Region of China is one of the provinces in this area.

Data published by the annual health statistics of China in 2006 suggested that obstetric haemorrhage was the leading cause of maternal death (50.0%), followed by indirect causes (26.7%) and hypertensive diseases of pregnancy (6.7%) in Ningxia of China (MOH China 2006). At least 60% of all maternal deaths in Ningxia of China were caused by factors that can either be prevented or averted successfully through the provision of essential obstetrical care (Khan et al. 2006; Ronsmans et al. 2006; Mathers et al. 2009). An intervention study in Tanzania indicated that the incidence of post-partum haemorrhage fell significantly from 32.9% to 18.2% with the improved performance of health staff through short-term training (Sorensen et al. 2011). Another study in Tanzania found significant reductions in infections among post-partum women who used clean delivery kits and were managed with the hygienic procedures recommended by WHO (Winani et al. 2007). A systematic review of observational studies showed that haemoglobin levels increased by 11.3 g/l in developing countries, and the prevalence of maternal anaemia dropped from 33.3% to 20.0% after providing iron-folic acid supplementation to pregnant women (Sanghvi et al. 2010).

Other strategies such as community mobilisation, establishing maternity waiting homes (MWH), free access to prenatal care and institutional delivery as well as policy making also contributed to reducing maternal deaths (Campbell et al. 2006). The strategy of comprehensive interventions is effective in reducing maternal deaths, such as the large-scale intervention in reducing maternal mortality in Sri Lanka (Fernando et al. 2003; Prata et al. 2010). Hence, comprehensive interventions were part of the Swedish International Development Cooperation Agency (SIDA) project in Ningxia Hui Autonomous Region of China from 2006 to 2010. To our knowledge, few studies have reported utilisation of maternal healthcare services among Muslim pregnant women. Owing to their religious customs and the disadvantaged socioeconomic status, women in Ningxia may have special needs for maternal healthcare services. Our study aimed to understand the utilisation of prenatal care and hospitalised delivery among pregnant Muslim women in Ningxia, China, and to explore the effectiveness of the integrated interventions to reduce their mortality.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

Study site

Ningxia Hui Autonomous Region in the northwest of China is one of its most economically underdeveloped regions. The four project counties are situated in the south of Ningxia on the Loess Plateau, and most parts are mountainous with difficult access. The total population in the project counties was 1 500 000 in 2009. The dominant ethnic group is Hui/Muslim. The annual per capita income was below the national average (3255 yuan, National Bureau of Statistics of China, in 2005). The number of live births grew from around 22 000 in 2005 to 24 500 in 2010.

Study design

This intervention study used a before-and-after comparison of the same group and took place in three phases: the pre-intervention assessment in 2006, the intervention phase from 2006 to 2010 and the post-intervention assessment in 2010.

Pre-intervention assessment

For the pre-intervention assessment, multistage sampling was used to select townships and villages (Figure 1). First, all townships in the four counties were ranked according to the hospital delivery rate in 2005. The townships were categorised into three strata with almost the same numbers of townships in each stratum, and then one township was randomly selected from each stratum. Then, in each selected township, all villages were ranked according to the distances from the village to the town centre and divided into three strata with almost the same numbers of villages, and then one village was randomly selected from each stratum. Finally, in the selected villages, caregivers who had child(ren) under the age of five by the time when the survey was conducted were selected until the required sample size was achieved.


Figure 1.  The sampling flowchart of this study.

Download figure to PowerPoint

Based on the main project objective of increasing hospitalised delivery in the four counties by 20%, it was calculated that 171 caregivers in each of the four counties, or 19 caregivers in each of the sample villages, needed to be included in this study. Information about the family was collected from the caregivers. The use of prenatal care during the pregnancy of the youngest child was also collected if the caregiver was the mother. The MMR data were obtained from the local Maternal and Child Mortality Report System, which was the China’s official source of statistics on maternal deaths. In each county, the maternal and child health (MCH) institution was responsible for reporting all maternal deaths. One health worker in each village recorded all such events in his or her catchment area. Reports were transmitted monthly to the township MCH worker and quarterly to local MCH institution in each county.


The interventions adopted by the project were based on published strategies (Campbell et al. 2006) and our experiences of UNICEF projects in China. They comprised six major components:

  •  Development of an MCH service guideline package: The package included two technical guidelines for the county MCH workers, two manuals for village doctors and one handbook for pregnant women and mothers with children aged under 5 years. These materials were distributed and used by all health-related facilities, pregnant women and caregivers of children in the project counties.
  •  Development of policies to remove cost obstacles and promote referrals through incentive mechanisms. The project introduced a price control plan for normal delivery and a transport contingency plan including improving emergency referral network and transport subsidy mechanism to promote institutional delivery services. This was evolved into a milestone policy in the whole province –‘four free one care’ policy in 2009 (free hospital delivery, free prenatal and post-natal care, free pre-marital checking, free detection and treatment for neonatal congenital diseases and assisting poor pregnant women with complications).
  •  Multiple communication channels were employed to conduct culturally appropriate health promotion activities, with wide involvement of the women’s union, local communities, schools, public media and government sectors. In addition, mosque-based health education was implemented by taking into account the potential influence of religious leaders on the residents. The coverage of health promotion activities reached 100% of the target population in the project counties.
  •  Periodical training of healthcare providers to improve the quality of care. The project particularly focused on the training of trainers (TOT) and skills transfer through cascade training. Over the entire project implementation period, 16 TOT sessions were conducted by MCH experts, benefiting more than 550 people, and then 56 training sessions were conducted by local health authorities, benefiting over 2460 township MCH workers and village doctors.
  •  Establishment of MWH. The MWH approach included making physical improvements to the existing hospital rooms and provision of living necessities, as well as providing free prenatal care and maternity waiting services. In the project counties, 18 townships were installed with MWH and 2631 pregnant women were benefited from this initiative.
  •  Upgrading critical equipment and facilities. To improve the quality of care, essential obstetrical equipment was procured for the project counties including Hemoglobin Analyzers, Urine Protein Strip, Adult Body-weight Balance and delivery kits for township hospitals.

Post-intervention assessment

The post-intervention assessment was conducted using the same sampling method and the same sample size as the pre-intervention assessment, in the same towns and villages.


Items about prenatal care and delivery in the questionnaire were adopted from the ‘Multiple indicator cluster survey manual’ published by UNICEF (2006). Before implementing the survey, all interviewers had been trained using the same standard. The study was approved by the Ethics Committee of Peking University Health Science Center.

Statistical analysis

Epidata3.0 software was used to input data. SPSS 13.0 for Windows statistical software was used to carry out statistical analysis. Frequencies were calculated, and chi-square tests were used to examine the frequency differences between different groups. The significance level was set at P < 0.05.

Because the multi-stage stratified, non-probability sampling method was employed in the survey, weighting methods were applied to adjust for varying selection probabilities. Weighted data were used in all analysis. The calculation of weights is detailed below.

Probability for selecting each sample:

  • image

i: county i; j: township j; k: village k.

Weighting factor: inline image

Standardised weighting factor: inline image


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

In the pre-intervention survey, of 717 caregivers, 583 (81.3%) were the mothers. In the post-intervention survey, of 757 caregivers, 632 (83.4%) were the mothers of the children. After excluding questionnaires with missing values or mistakes, 554 mothers in the pre-intervention group and 622 in the post-intervention group were analyzed. The groups did not differ significantly with respect to mothers’ age, ethnicity, number of children, gender and age of the youngest child (Table 1). However, mothers in the post-intervention group were better educated than those in the pre-intervention group. More mothers in the post-intervention group had primary (< 0.001) and secondary or higher education (= 0.012). Moreover, more households at post-intervention group than at pre-intervention group had a telephone (< 0.001), a refrigerator (< 0.001) and a colour TV set (< 0.001).

Table 1. Characteristics of mothers and their youngest children in pre- and post-intervention
N % N %
Characteristics of mothers
 Age (years old)
 Maternal education
  Secondary or higher12222.118930.4
 Numbers of children
  1 at least34462.436359.5
 Household characteristics of the mothers
  Have telephone25045.458393.7
  Have refrigerator448.013421.8
  Have colour TV set43378.460196.6
Characteristics of the youngest child
 Age (months)

Table 2 shows the mothers’ utilisation of maternal healthcare services. At the post-intervention survey, the proportion of mothers with at least one prenatal visit was 98.9%, higher than the proportion of 78.2% before the intervention (OR, 24.55; 95% CI, 11.37–53.12). The proportion of those with at least four visits was 70.7% post-intervention, much higher than 22.6% pre-intervention (OR, 8.26; 95% CI, 6.21–10.97). Moreover, the proportion of those with a prenatal visit during the first trimester of pregnancy was 82.6%, statistically significantly higher than the proportion of 35.1% pre-intervention (OR, 8.77; 95% CI, 6.58–11.69). Table 2 also shows the components of prenatal care among mothers who had received prenatal care during their last pregnancy. At the post-intervention survey, 97.4% of women had an ultrasound examination and 95.1% had their blood pressure measured. Around 90% of women had fundal height measured. Women post-intervention were more likely to receive any of these components than those pre-intervention (P <0.05).

Table 2. Utilization of maternal care
VariablesBaselinePost-interventionOR (95%CI)Adjusted OR* (95%CI)
n/N% n/N%
  1. *Adjusted for education level, mobile phone, refrigerator and television possession.

Prenatal care coverage
 At least once433/55478.2615/62298.924.55 (11.37–53.12)10.91 (4.93–24.13)
 4 or more visits98/43322.6435/61570.78.26 (6.21–10.97)8.24 (6.11–11.12)
 5 or more visits56/43312.9335/61554.58.05 (5.83–11.11)8.89 (6.30–12.55)
 Prenatal visit at first trimester152/43335.1508/61582.68.77 (6.58–11.69)8.84 (6.56–11.91)
Place of the first prenatal visit
 Village clinic9/4292.110/6051.60.78 (0.32–1.94)0.98 (0.40–3.19)
 Township hospital147/42934.3226/60537.41.14 (0.88–1.48)1.02 (0.81–1.78)
 Health institution at county or above273/42963.6369/60560.90.89 (0.69–1.15)0.94 (0.82–1.55)
Components of prenatal care
 Weight46/43310.6360/61558.511.87 (8.41–16.77)11.08 (8.27–17.43)
 Blood pressure198/43345.7584/61495.123.10 (15.29–34.91)14.22 (9.21–21.97)
 Blood test56/43312.9410/61466.813.53 (9.75–18.76)12.46 (8.49–18.28)
 Urine test138/43032.1497/61381.19.06 (6.81–12.06)7.90 (5.66–11.02)
 Fundal height154/41337.3551/61589.614.47 (10.44–20.07)12.15 (8.35–17.65)
 B ultrasound377/43287.3598/61497.45.45 (3.07–9.65)5.01 (2.90–10.45)
Delivery information
 Institutional delivery234/53543.7540/62286.88.47 (6.34–11.30)5.69 (4.06–7.99)
 Cesarean section18/2347.748/5408.91.17 (0.66–2.05)1.30 (0.66–2.53)
 Home delivery300/53556.081/62213.00.11 (0.08–0.15)0.17 (0.12–0.24)
 Hygienic delivery147/30049.058/8171.62.62 (1.54–4.47)2.51 (1.46–4.30)

Delivery information is also reported in Table 2. After the intervention, 86.8% of women delivered in a hospital, much higher than the proportion of 43.7% at the pre-intervention survey (OR, 8.47; 95% CI, 6.34–11.30). Although we did not validate self-reported delivery places by individual medical records, we examined hospital statistics in the 4 project counties and found that the number of hospital deliveries in the four county-level hospitals increased from 2126 in 2005 to 8762 in 2009. These data at least partly substantiated the self-reported increase in hospital deliveries. Among women not delivering at hospital, the proportion of hygienic delivery was 71.6% in the post-intervention survey, higher than 49.0% in the pre-intervention survey (OR, 2.62; 95% CI, 1.54–4.47).

The effects of the intervention on the utilisation of maternal care remained significant after adjusting for education level and household possession. Results from logistic regression analysis showed that mothers in the post-intervention group were more likely to use prenatal care services than those in the pre-intervention group (adjusted OR, 10.91; 95% CI, 4.93–24.13) and to choose to give birth in hospital (adjusted OR, 5.69; 95% CI, 4.06–7.99).

As shown in Figure 2, MMR in project areas decreased from 45.5 per 100 000 live births in 2005 to 32.7 per 100 000 live births in 2010. The numbers of maternal deaths fluctuated in these years, from 10 in 2005 to 14 in 2006 and 2007, five in 2008, seven in 2009, and eight in 2010. With the stable number of live births, the increased MMR from 2005 to 2006 was probably due to the improved reporting of maternal deaths when the information system was enhanced by the intervention. After the implementation of the intervention, MMR in the project areas decreased gradually to the level similar to the average MMR in general rural areas in China, 30.1 per 100 000 live births according the newly released Health Statistical Yearbook.


Figure 2.  Maternal deaths per 100 000 live births in project areas.

Download figure to PowerPoint


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References

This pre- and post-intervention observational study showed that an integrated strategy of maternal care components effectively reduced maternal mortality in a poor rural region of China.

A major limitation of our study was that the causal effects of the intervention on the increased utilisation of maternal health service and decreased maternal mortality (Sibbald & Roland 1998) could not be well examined because of the lack of control group and randomisation. However, the current study is the first prospective observational study on the impacts of integrated interventions on the utilisation of prenatal care and institutional delivery services in a Hui/Muslim minority area in China. In the project areas, the most important intervention was to improve the knowledge and behaviours of pregnant women and to promote prenatal care and hospitalised delivery. Therefore, it is highly possible that the improved utilisation of maternal care and reduced maternal deaths in the project areas is attributable to the improved access to and quality of prenatal care and hospital delivery.

Another limitation was recall error in the self-reported data in this study (Schmier & Halpern 2004; Fransson et al. 2008). Our data on prenatal care and delivery were based on mothers’ self reports rather than medical records. Under- or over-reporting of recommended practices cannot be excluded. However, major life events, such as pregnancy and the birth of a child, may decrease the recall error (Belli 1998). Our study findings concur with Bhutta’s study that concluded that an intervention promoting preventive maternal and newborn care can result in care-seeking behaviour change for mothers during pregnancy and childbirth and lead to beneficial health outcomes (Bhutta et al. 2011).

We found an evident decrease in maternal mortality in project areas, which could be explained by the several interventions. Establishment of MWH played an important role in promoting delivery at hospital and decreasing maternal mortality as a result. The purpose of MWH is to provide a setting where pregnant women with clinical and social problems can stay near a hospital (WHO 1996). As the beginning of the 20th century, MWH have existed in Northern Europe to serve women in remote geographic areas with few obstetric facilities (WHO 1996; van Lonkhuijzen et al. 2009). In Africa, one of the early experiments with MWH was in Eastern Nigeria. It helped to reduce MMR in hospitals from 10 to <1 per 1000 deliveries (Poovan et al. 1990). Similarly, in Cuba, maternal mortality fell from 118 to 31 per 100 000 live births after the establishment of MWH (van Lonkhuijzen et al. 2009). MWH have been shown to be an effective intervention to improve maternal outcomes in many settings where access to hospitals is limited (Eckermann 2006; Eckermann & Deodato 2008). In the project areas, especially in mountainous areas and remote villages, it is difficult to reach health facilities in an emergency because of extreme weather conditions in winter or the rainy season. According to the pre-intervention survey, a large proportion of pregnant women gave birth at home before the establishment of MWH. However, institutional delivery increased significantly in the project areas after the establishment of MWH.

The economic barrier was another important factor impacting hospital delivery. This includes the cost of delivery services at hospital, the transportation cost from home to health facilities and back and lost labour and production during absences from the village (Richard et al. 2010; Feng et al. 2011). Given the low income in the project areas, the majority of the pregnant women could not afford the transport and other related fees. Hence, a policy was developed to provide a subsidy for delivery at hospital, prenatal visits and post-natal visits for pregnant women. The increased utilisation of these services in the project areas could partly explain the improved maternal health.

Poor-quality obstetric service is a key barrier to reaching MDG5 (Anwar et al. 2009; Li et al. 2011). Therefore, particular attention was given to improve the capacity of maternal health providers by the project. Active learning opportunities including case discussions, hands-on training sessions, interactive workshops and reinforced training sessions were provided. In addition, there is evidence of association between the use of prenatal service and the maternal mortality (Varma et al. 2011). Firstly, there was a negative association between the number of prenatal visits and perinatal death (Vonderheid et al. 2007; Ayoola 2011). The contact established between the pregnant woman and healthcare providers is an ideal opportunity to provide health education on life style, taking advantage of the women’s increased motivation to decrease the risk for the baby. In our study, the proportion of those who had had four or five visits post-intervention was higher than that pre-intervention, indicating that contact opportunities between the pregnant woman and the healthcare providers had been established. Secondly, consensus about the importance of the first prenatal visit is well established (Ochako et al. 2011). The guidelines for prenatal care in China recommended that the first prenatal visit occur before 12 weeks of gestation. Our results showed that after the intervention, 82.6% of pregnant women received their first prenatal visit during the first trimester. It greatly increased the contact opportunities between the pregnant woman and their healthcare providers. Thirdly, the components of prenatal service have a real impact on maternal mortality, especially in developing countries (WHO 2006). The WHO guidelines clearly stated that prenatal care visits should include blood pressure measurement, urine testing, blood testing and weight/height measurement (Kirkham et al. 2005). Our study found that the proportions of women having the above-mentioned components increased to different degrees at post-intervention survey, indicating the enhanced quality of prenatal care after intervention.

Lastly, community mobilisation was another essential strategy (USAID 2007). In Ren’ study, pregnant women of the Hui minority were the most important target population of health education, because Han pregnant women are more likely to receive prenatal care (Ren 2010). Moreover, men were perceived superior to women in the project areas because of the traditional Hui culture. Therefore, the project not only included pregnant women, but also their husbands and family members in the target population of social mobilisation. At the same time, local influential people including village leaders, village doctors and religious leaders (imams) were also encouraged to participate in the health promotion activities to maximise the effectiveness.

In summary, an integrated strategy, including training of obstetricians for obstetric care, establishment of MWH, providing free delivery in hospital, development of a feasible package of MCH care, together with community-based health promotion was effective in reducing maternal mortality.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. References
  • Anwar I, Kalim N & Koblinsky M (2009) Quality of obstetric care in public-sector facilities and constraints to implementing emergency obstetric care services: evidence from high- and low-performing districts of Bangladesh. Journal of Health, Population, and Nutrition27, 139155.
  • Ayoola AB (2011) Reducing the number of antenatal care visits in low-risk pregnancies increases perinatal mortality in low- and middle-income countries; women in all settings prefer the standard visit schedule. Evidence-Based Nursing14, 5556.
  • Belli RF (1998) The structure of autobiographical memory and the event history calendar: potential improvements in the quality of retrospective reports in surveys. Memory6, 386606.
  • Bhutta ZA, Soofi S, Cousens S et al. (2011) Improvement of perinatal and newborn care in rural Pakistan through community-based strategies: a cluster-randomised effectiveness trial. Lancet377, 403412.
  • Campbell OM, Graham WJ & Lancet Maternal Survival Series steering group (2006) Strategies for reducing maternal mortality: getting on with what works. Lancet368, 12841299.
  • Chatterjee A & Paily V (2011) Achieving millennium development goals 4 and 5 in India. BJOG: An International Journal of Obstetrics and Gynaecology118(Suppl. 2), 4759.
  • Eckermann L (2006) Finding a ‘safe’ place on the risk continuum: a case study of pregnancy and birthing in Lao PDR. Health Sociology Review15, 374386.
  • Eckermann E & Deodato G (2008) Maternity waiting homes in Southern Lao PDR: the unique ‘silk home’. Journal of Obstetrics and Gynaecology Research34, 767775.
  • Feng XL, Xu L, Guo Y & Ronsmans C (2011) Socioeconomic inequalities in hospital births in China between 1988 and 2008. Bulletin of the World Health Organization89, 432441.
  • Fernando D, Jayatilleka A & Karunaratna V (2003) Pregnancy – reducing maternal deaths and disability in Sri Lanka: national strategies. British Medical Bulletin67, 8598.
  • Fransson E, Knutsson A, Westerholm P & Alfredsson L (2008) Indications of recall bias found in a retrospective study of physical activity and myocardial infarction. Journal of Clinical Epidemiology61, 840847.
  • Hill K, Thomas K, Abouzahr C et al. (2007) Estimates of maternal mortality worldwide between 1990 and 2005: an assessment of available data. Lancet370, 13111319.
  • Khan KS, Wojdyla D, Say L, Gülmezoglu AM & Van Look PF (2006) WHO analysis of causes of maternal death: a systematic review. Lancet367, 10661074.
  • Kirkham C, Harris S & Grzybowski S (2005) Evidence-based prenatal care: part I. General prenatal care and counseling issues. American Family Physician71, 13071316.
  • Li X, Zhu J, Dai L et al. (2011) Trends in maternal mortality due to obstetric hemorrhage in urban and rural China, 1996–2005. Journal of Perinatal Medicine39, 3541.
  • Lozano R, Wang H, Foreman KJ et al. (2011) Progress towards Millennium Development Goals 4 and 5 on maternal and child mortality: an updated systematic analysis. Lancet378, 11391165.
  • Mathers CD, Boerma T & Ma Fat D (2009) Global and regional causes of death. British Medical Bulletin92, 732.
  • MOH China (2005) National Health Statistic Yearbook 2005 (in Chinese). MOH, China, Beijing.
  • MOH China (2006) National Health Statistic Yearbook 2006 (in Chinese). MOH, China, Beijing.
  • Ochako R, Fotso JC, Ikamari L & Khasakhala A (2011) Utilization of maternal health services among young women in Kenya: insights from the Kenya Demographic and Health Survey, 2003. BMC Pregnancy Childbirth11, 1.
  • Poovan P, Kifle F & Kwast B (1990) A maternity waiting home reduces obstetric catastrophes. World Health Forum11, 440445.
  • Prata N, Passano P, Sreenivas A & Gerdts CE (2010) Maternal mortality in developing countries: challenges in scaling-up priority interventions. Women’s Health (London, England)6, 311327.
  • Ren Z (2010) Utilisation of antenatal care in four counties in Ningxia, China. Midwifery27, e260e266.
  • Richard F, Witter S & de Brouwere V (2010) Innovative approaches to reducing financial barriers to obstetric care in low-income countries. American Journal of Public Health100, 18451852.
  • Ronsmans C, Graham WJ & Lancet Maternal Survival Series steering group (2006) Maternal mortality: who, when, where, and why. Lancet368, 11891200.
  • Sanghvi TG, Harvey PW & Wainwright E (2010) Maternal iron-folic acid supplementation programs: evidence of impact and implementation. Food and Nutrition Bulletin31(Suppl. 2), S100S107.
  • Schmier JK & Halpern MT (2004) Patient recall and recall bias of health state and health status. Expert Review of Pharmacoeconomics & Outcomes Research4, 159163.
  • Sibbald B & Roland M (1998) Understanding controlled trials. Why are randomised controlled trials important?BMJ316, 201.
  • Sorensen BL, Rasch V, Massawe S, Nyakina J, Elsass P & Nielsen BB (2011) Advanced life support in obstetrics (ALSO) and post-partum hemorrhage: a prospective intervention study in Tanzania. Acta Obstetricia et Gynecologica Scandinavica90, 609614.
  • UNICEF (2006) Multiple Indicator Cluster Survey Manual 2005: Monitoring the Situation of Children and Women. UNICEF, New York, NY.
  • USAID (2007) Demystifying Community Mobilization: An Effective Strategy to Improve Maternal and Newborn Health. USAID, Baltimore.
  • van Lonkhuijzen L, Stekelenburg J & van Roosmalen J (2009) Maternity waiting facilities for improving maternal and neonatal outcome in low-resource countries. Cochrane Database of Systematic Reviews3, CD006759.
  • Varma GR, Kusuma YS & Babu BV (2011) Antenatal care service utilization in tribal and rural areas in a South Indian district: an evaluation through mixed methods approach. The Journal of the Egyptian Public Health Association86, 1115.
  • Vonderheid SC, Norr KF & Handler AS (2007) Prenatal health promotion content and health behaviors. Western Journal of Nursing Research29, 258276.
  • WHO (1996) Maternity Waiting Homes: A Review of Experiences. WHO, Geneva.
  • WHO (1999) The World Health Report 1999: Making a Difference. WHO, Geneva.
  • WHO (2006) Pregnancy, Childbirth, Postpartum and Newborn Care: A Guide for Essential Practice. WHO, Geneva.
  • WHO, UNICEF, UNFPA, The World Bank (2007) Maternal Mortality in 2005: Estimates Developed by WHO, UNICEF, UNFPA, and the World Bank. WHO, Geneva.
  • Winani S, Wood S, Coffey P, Chirwa T, Mosha F & Changalucha J (2007) Use of a clean delivery kit and factors associated with cord infection and puerperal sepsis in Mwanza, Tanzania. Journal of Midwifery & Women’s Health52, 3743.