- Email: [email protected]
The effect of prenatal and intrapartum care on the stillbirth rate among women in rural Ethiopia
International Journal of Gynecology and Obstetrics 133 (2016) 164–167
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
The effect of prenatal and intrapartum care on the stillbirth rate among women in rural Ethiopia Karen Ballard ⁎, Zelalem Belete, Hirut Kinfu, Mebkyou Tadesse, Mohammed Amin, Habtamu Atnafu Hamlin Fistula Ethiopia, Addis Ababa Fistula Hospital, Addis Ababa, Ethiopia
a r t i c l e
i n f o
Article history: Received 8 June 2015 Received in revised form 12 September 2015 Accepted 26 January 2016 Keywords: Basic emergency obstetric and newborn care Ethiopia Health facility delivery Prenatal care Satisfaction with care Stillbirth
a b s t r a c t Objective: To determine whether community-based prenatal and intrapartum care in Ethiopia results in a lower stillbirth rate. Methods: Between May and December 2014, a randomly selected sample of women in northern and eastern Ethiopia who had delivered a neonate in the preceding 12 months completed a face-to-face survey about their experiences of maternal services and the fetal outcome for each delivery. The stillbirth rates among women delivering at home and at health facilities were compared. Results: Overall, 4442 women completed surveys. Stillbirth was reported by 42 (1.7%) of the 2437 women who had received prenatal care and 53 (2.8%) of the 1921 women who did not receive prenatal care (P = 0.01). The stillbirth rate was similar among women who delivered in a health center (27/1417 [1.9%]), in a hospital (6/126 [4.8%]), and at home (62/2725 [2.3%]; P = 0.13). However, women experiencing an intrapartum emergency were twice as likely to deliver in a health facility (odds ratio 2.6, 95% confidence interval 2.2–3.0). Satisfaction with health-center care was moderately good (median score 77.5/100). Conclusion: The stillbirth rate was reduced among women receiving prenatal care, although delivering in a health facility did not reduce the risk of stillbirth. Improving the quality of health-center care could lead to their planned use for childbirth, which might reduce stillbirth rates. © 2016 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction For decades, Ethiopian women have been delivering their neonates at home, which is perhaps unsurprising because health centers— originally introduced in 1952 as the backbone of the Ethiopian health service—were slow to appear in the rural areas where the vast majority of people live. In 1974, there were just 93 health centers across the whole country [1], and approximately 20 years later, there were only 451 health centers and 119 hospitals for a population of 77 million [2]. It is only in the past 3–4 years that the number of health facilities has increased, although in 2014, there were still only 3245 health centers and 127 hospitals for a population of almost 86 million [3]. The fairly recent appearance of “widespread” healthcare coverage in Ethiopia makes it easy to understand that health centers have been viewed by women as neither necessary nor customary for childbirth and their subsequent preference for a homebirth [4,5]. Consequently, women often perceive the health facility as a place to seek help only in the event of a complication [6], leading to a persistently low institutional delivery rate of just 16% [5] and a maternal mortality ratio of 671 per 100 000 live births [7]. Likewise, the national
stillbirth and perinatal mortality rates have remained high, at 2.1% and 4.6%, respectively [7]. Encouraging delivery in a health facility is hindered by the suboptimal quality of care that women often receive in facilities [4,8] from poorly skilled staff [9,10]. Indeed, in 2008, only 1.3% of Ethiopian health centers were fully functioning as facilities providing basic emergency obstetric and newborn care (BEmONC) [11]. One study in 2010 [12] revealed that just one-third of hospital-based maternal-care staff had knowledge of the signs of obstructed labor and knew how to treat postpartum hemorrhage. Recognizing the inadequacies of the services, in 2010, the Ethiopian Government prioritized maternal health care, increasing the provision of rural health centers, midwives, surgically trained health officers, and ambulances [13]. However, little is known about whether women have embraced these improved maternal health services and whether the care they experience leads to improvements in fetal outcomes. The aim of the present study was to determine women's satisfaction with the community-based maternal services provided in Ethiopia and whether the use of facility-based prenatal and intrapartum care leads to a reduction in stillbirth rates. 2. Materials and methods
⁎ Corresponding author at: Hamlin Fistula Ethiopia, Addis Ababa Fistula Hospital, PO Box 3609, Addis Ababa, Ethiopia. Tel.: +44 7715 487233. E-mail address: [email protected] (K. Ballard).
A survey-based investigation was undertaken in two regions (Amhara and Oromia) of Ethiopia, a country divided into 11 regions
http://dx.doi.org/10.1016/j.ijgo.2015.09.027 0020-7292/© 2016 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
K. Ballard et al. / International Journal of Gynecology and Obstetrics 133 (2016) 164–167
and with an estimated population of 91 million [14], and where the vast majority of people are subsistence farmers living in rural areas with limited transport. All prenatal, intrapartum and postnatal care received within government facilities is free, although charges are made for items such as gloves, sutures, and medication. The main source of facility-based maternal care is through the government-run health centers, supported by lower-level care from health posts based within villages and staffed by health extension workers (HEWs). The present study was part of a larger investigation that used a household survey of women's maternal experiences to determine whether a maternal health package introduced into eight health centers in 2010–2011 improved the quality of prenatal and intrapartum services when compared with 18 health centers without the package. The maternal health packages were provided by Hamlin Fistula Ethiopia (HFE), a non-governmental organization offering free care for women with vesicovaginal and rectovaginal fistulas, and BSc midwifery training for students recruited from rural areas of Ethiopia. The packages were introduced in three zones of Ethiopia—South Gondar, West Gojjam, and Eastern Harraghe—and consisted of two midwives trained at HFE who were provided with continuous mentoring, accommodation near to the center, telephone credit, any necessary equipment or drugs, and access to an ambulance for referral to higher-level care. The primary outcome measure was stillbirth rate, which we hypothesized would reduce if high-quality prenatal and intrapartum maternal services were available. To calculate the sample size required for the primary household survey, HFE clinicians advised that the maternal package would be a valuable addition to the community-based maternal services if it were capable of bringing about a 50% reduction in the stillbirth rate. According to the Ethiopian Demographic Health Survey (DHS) [7], the stillbirth rate in Amhara and Oromia is 46 per 1000 pregnancies. A design effect of 1.2 was applied to reduce errors due to cluster sampling techniques, meaning that the total sample size required was 4373 women, with a significance level of 5% and 80% power. To select the sample for the survey, four or five health posts linked with each of the 26 health centers were selected according to their distance from the health center, providing one nearby health post, two a medium distance away, and one far away, leading to a total of 116 health posts. Subsequently, 28 000 households served by the health posts were randomly selected using random number tables from HEW household lists to provide a population of 140 000 women. Given the 3.5% annual pregnancy rate for this population [3], a sample of 4900 women was obtained. Allowing for 10% error in data collection, the required sample size would be achieved. The inclusion criterion was having delivered a neonate after a pregnancy lasting 7 months or more in the previous 12 months. The present study uses data collected from all selected households and does not set out to make comparisons between women served by health centers with and without a maternal health package. Data were obtained between May and December 2014. Following ethical approval from the Oromia and Amhara regional health bureaus, and with permission from the heads of relevant districts, HEWs were recruited and trained to administer the survey. The HEWs were recruited from the health posts based within the villages selected for sampling. This was essential because the HEWs were knowledgeable about the location of all households within the village and how to access what are often very remote areas. Additionally, they were acquainted with most of the women in their village and were female, so able to ask questions that could be considered intimate and sensitive in nature. During the training session, 120 households were randomly selected from each HEW's list for surveying. Because the vast majority of women were unable to read or write, the survey was administered verbally and the HEWs transcribed the answers. Before answering, women were read a participant information sheet, and if they agreed to participate, they were required to sign or place a fingerprint on the consent form. The survey focused on women's use of the maternal services for the delivery in the past 12 months, but also asked about the place of birth
165
for all of their deliveries and the outcome for each neonate. The adapted Haddad 20-item scale [15], which has been validated in countries with similar health systems to Ethiopia, was used to assess satisfaction with the maternal care in the health centers among women using these services. A score of 0–100 was calculated, with 0 representing the lowest possible satisfaction with care and 100 representing the highest. The survey was translated into the appropriate languages and piloted. Throughout the data collection period, a team of researchers visited the sites to perform random data accuracy and quality checks. Each HEW was located by phone and a selection of her completed surveys was readministered at the households to check for accuracy of the originally administered survey. At the end of data collection, the HEWs were paid for their completed surveys, with an additional bonus for accurate and complete datasets. Additionally, records for the 26 health centers serving the surveyed women were assessed by members of the research team to determine the number and skill level of staff, availability of services, number of deliveries and outcomes, and the ability to provide BEmONC. Data were entered into SPSS version 20 (IBM, Armonk, NY, USA). Data were not normally distributed and were therefore reported as medians with interquartile ranges (IQRs). Group differences were tested using the Pearson χ2 test and differences in median scores across the four components of satisfaction scale were tested using the Friedman test. P b 0.05 was considered statistically significant. Odds ratios (ORs) with 95% confidence intervals (CIs) were used to estimate the chances of using health facilities for intrapartum care. 3. Results A total of 4442 completed surveys were available for analysis. The median time travelled by foot to reach a health center was 60 min (IQR 40–120). A total of 1670 (37.6%) women delivered over the preceding 12 months in a health facility, with 1417 (31.9%) delivering in a health center, 126 (2.8%) in a hospital, 121 (2.7%) in a health post, and 6 (0.1%) in a private clinic. Overall, 2725 women (61.3%) delivered at home and 45 (1.0%) delivered on the roadside. Place of delivery was not stated by 2 (b 0.1%) women. The 37.6% facility delivery rate in the preceding 12 months was greater than the 9.3% (338/3642) facility delivery rate reported for the previous delivery (P b 0.001). Indeed, the last neonate delivered was 9 times more likely to be delivered in a health facility compared with the last but one neonate (OR 9.5, 95% CI 7.2–12.5). Data reported across the 26 health centers revealed that 8320 (28.4%) of the estimated 29 300 pregnant women served by the health centers had delivered in one of these health centers in the past 12 months. Overall, 2437 (54.9%) women received prenatal care from their nearest health center, 1040 (42.7%) of whom also delivered in the health center. Women receiving prenatal care from a health professional were significantly more likely to deliver in a health facility than were those who had not received such care (OR 3.5, 95% CI 3.0–3.9). A prepartum or intrapartum emergency in the last delivery was reported by 668 (15.0%) women. The main emergency reported was a prolonged labor of more than 24 h (Table 1). More than half the women delivering at hospital, 28.9% delivering in a health post, and 19.5% delivering in a health center had an obstetric emergency. Delivery in a health facility was more common among women who experienced an obstetric emergency than among those who did not (382/668 [57.2%] vs 1261/3774 [33.4%]; P b 0.001). The odds of delivering in a health facility if experiencing an obstetric emergency were 2.6 (95% CI 2.2–3.0). Overall, 33 (0.7%) women reported having a cesarean delivery, all of which were performed in a hospital setting. Of the 8320 deliveries performed in and reported by the 26 health centers, 292 (3.5%) were vacuum assisted. Overall, 95 (2.1%) women reported that their most recent pregnancy ended in a stillbirth. This rate is considerably lower than the
166
K. Ballard et al. / International Journal of Gynecology and Obstetrics 133 (2016) 164–167
Table 1 Prepartum and intrapartum emergencies.a
Any prepartum or intrapartum emergency Prolonged labor (N24 h) Ruptured uterus Eclampsia Prepartum hemorrhage
Delivered in the health post (n = 121)
Delivered at a private clinic (n = 6)
Delivered at the health center (n = 1417)
Delivered at hospital (n = 126)
Delivered at the roadside (n = 45)
668 (15.0)
277 (10.2)
35 (28.9)
0
277 (19.5)
70 (55.6)
9 (20.0)
611 (13.8) 23 (0.5) 30 (0.7) 33 (0.7)
249 (9.1) 12 (0.4) 14 (0.5) 12 (0.4)
31 (25.6) 1 (0.8) 3 (2.5) 2 (1.7)
0 0 0 0
260 (18.3) 7 (0.5) 9 (0.6) 12 (0.8)
63 (50.0) 3 (2.4) 3 (2.4) 6 (4.8)
8 (17.8) 0 1 (2.2) 1 (2.2)
Values are given as number (percentage). Data missing for 2 women.
13.6% (74/544) stillbirth rate reported for the last but six pregnancy (P b 0.001). The stillbirth rate has declined as the rate of delivery in health institutions has risen (Fig. 1). For neonates born in the 12 months preceding the survey, stillbirths were recorded for 62 (2.3%) of the 2725 home deliveries, 27 (1.9%) of the 1417 deliveries in health centers, and 6 (4.8%) of the 126 hospital deliveries (P = 0.13). Stillbirth was reported by 42 (1.7%) of the 2437 women who received prenatal care and 53 (2.8%) of the 1921 women who did not receive prenatal care (P = 0.01). As previously mentioned, women who experienced an obstetric emergency were more likely to seek help from the health facility. Stillbirth occurred among 22 (3.3%) of the 668 women experiencing an emergency and among 71 (1.9%) of the 3774 women not experiencing an emergency (P = 0.024). Among women who delivered in a health facility, 24 (1.9%) of the 1261 women who did not experience an obstetric emergency had a stillbirth compared with 9 (2.4%) of the 382 women who did experience an emergency (P = 0.58). Although the absolute numbers are too small to determine significant relationships between the rate of institutional stillbirths and complicated deliveries, it is likely that the inevitable delays that occur in reaching a health facility increase the stillbirth rate within the health facilities. The data reported by the 26 health centers revealed a total of 206 (2.5%) stillbirths among the 8320 deliveries that occurred in these centers. The health centers served a total population of 837 131, and there were 49 midwives working across these centers, providing a midwifeto-population ratio of 1:17 084. Nine (34.6%) health centers had running water, 19 (73.1%) had intermittent electricity, and 14 (53.8%) provided BEmONC. Women's satisfaction with the care received when delivering at the health center (n = 1417) was moderately good, with a median score of 77.5 (IQR 72–86). Women were more satisfied with Still birth
access to care (median score 80, IQR 75–85) and healthcare delivery (median 79, IQR 75–88) than they were with the quality of interpersonal care (median 75, IQR 67–88) and the quality of the health facility and its resources (median 75, IQR 67–83; P b 0.001). 4. Discussion More than one-third of women surveyed in the present study had had their most recent delivery in a health facility, reflecting a dramatic increase over the past 2 or 3 years. It is difficult to make direct comparisons with data from the DHS, which refer to women who delivered in the preceding 5 years [7]. Nevertheless, the Ethiopian mini-DHS has shown an increase in institutional delivery rate from 10% in 2011 to 16% in 2014 [5,7], and national data collected directly from health facilities in 2012–2013 [3] established that 23.1% of deliveries were performed by skilled birth attendants. Other data from small, local studies also suggest that there have been substantial improvements in facility delivery rates, with data collected in 2009–2011 reporting an institutional delivery rate of 18.2% [16], and data from 2013 revealing a facility delivery rate of 47% [17]. In addition to the increased facility delivery rate, the present study highlights a concomitant decrease in stillbirth rates. The 2011 Ethiopian DHS reported a 4.8% stillbirth rate in the Oromia region and 4.7% in the Amhara region [7], so the stillbirth rate reported here represents a sizeable reduction, which could be partly explained by the increased use of health facilities for prenatal and intrapartum care. Indeed, it was found that women accessing prenatal care at the health centers had a lower stillbirth rate than did those who did not access such care. However, as in other studies [18,19], it was found that Institutional delivery
40.0% 35.0% 25.0% 20.0% 15.0% 10.0% 5.0%
Previous Deliveries Fig. 1. Percentage of women delivering in a health center and experiencing stillbirth.
Last pregnancy
Last but 1 pregnancy
Last but 4 pregnancy
Last but 5 pregnancy
0.0%
Last but 6 pregnancy
Percentage
30.0%
Last but 2 pregnancy
b
Delivered at home (n = 2725)
Last but 3 pregnancy
a
All participants (n = 4440)b
K. Ballard et al. / International Journal of Gynecology and Obstetrics 133 (2016) 164–167
many women only used the health facilities once an obstetric emergency had arisen, limiting the effectiveness of the care they received. The present finding that women delivering at home had similar stillbirth rates to those delivering at a health center illustrates the limited care that the health centers were able to provide for the unborn fetus. Improvements in the quality of care offered by the health centers will do much to encourage women to seek care during the early stages of labor. Similarly to other findings [8], the present data on satisfaction with the maternal services indicate a need for improvements in interpersonal care, and the awaited results of an in-depth study on disrespect and abuse in maternal health care in Ethiopia [20] should provide the government with valuable direction about ways to best improve the quality of care. The present study also highlighted the need for improvements in the level of care being provided at health centers, with approximately half the facilities being able to provide BEmONC. In 2008, the Ethiopian Government led a national assessment of both basic and comprehensive emergency obstetric and newborn care [11], revealing that only 1.3% of the health centers were full BEmONC providers. Subsequently, a standardized curriculum for BEmONC training was developed [21], increasing the availability of emergency obstetric and neonatal care within health centers to 56% [3]. However, an evaluation of 10 health centers in Addis Ababa in 2013 revealed that only 12% of staff had received BEmONC training and that their overall knowledge and skill in providing emergency care remained poor [22]. Although the sampling methods applied in the present study allow for the findings to be representative of women living in both rural and urban areas, they can reflect only the views and experiences of women living in the two regions studied. Data accuracy is always of concern when using self-reported data. This is of particular importance in Ethiopia, where both women and health workers have received health messages about the importance of delivering in a health facility that could have provoked a degree of moral judgment surrounding home deliveries. However, the delivery and stillbirth rates obtained from the survey data were similar to those retrieved from the health center records. Additionally, the extensive data verification process encouraged data collectors to be accurate. The presence of the research team in the study sites throughout the data collection period and the offer of a financial incentive for high-quality data will also hopefully have contributed to maximizing data quality. In conclusion, the present study has shown that women who access prenatal care have a lower stillbirth rate than do those who do not access such care, but delivery in a health facility does not lower the stillbirth rate. However, many women delay seeking help from the health center until an emergency arises, which is likely to limit the care that health centers can provide. Women's satisfaction with the maternal care they received was moderately good and improvements could motivate them to actively seek care from the facilities rather than relying on them for emergencies only. In turn, this is likely to lead to a reduction in stillbirth rates. Acknowledgments The project was supported by resources from HFE and external funding from Johnson and Johnson. The funders had no role in the study design, data collection and analysis, decision to publish, or the preparation of the manuscript.
167
Conflict of interest The authors have no conflicts of interest.
References [1] Meche H. The Development of Health Services in Ethiopia. In: Koos H, Zein ZA, editors. The Ecology of Health and Disease in Ethiopia. Addis Ababa: Ministry of Health; 1988. p. 57–66. [2] Ministry of Health (Ethiopia). Health and health related indicators. Addis Ababa: Ministry of Health; 2006/7. http://cnhde.ei.columbia.edu/files/5/6/2008-Feb-Fri102213.pdf. Published 2007. Accessed September 7, 2015. [3] Ministry of Health (Ethiopia). Health and health related indicators. Addis Ababa: Ministry of Health; 2012/13. http://www.moh.gov.et/documents/26765/0/Health+ and+Health+Related+Indicators+2005+E.C/1b5b2a9f-a960-4024-8d92-51919 5364023?version=1.0. Published 2014. Accessed September 7, 2015. [4] Shiferaw S, Spigt M, Godefrooij M, Melkamu Y, Tekie M. Why do women prefer home births in Ethiopia? BMC Pregnancy Childbirth 2013;13:5. [5] Central Statistical Agency (Ethiopia). Ethiopia Mini Demographic and Health Survey 2014; Addis Ababa: Central Statistical Agency. http://www.unicef.org/ethiopia/ Mini_DHS_2014__Final_Report.pdf. Published 2014. Accessed September 7, 2015. [6] Bedford J, Gandhi M, Admassu M, Girma A. ‘A normal delivery takes place at home’: a qualitative study of the location of childbirth in rural Ethiopia. Matern Child Health J 2013;17(2):230–9. [7] Central Statistical Agency (Ethiopia). Ethiopia Demographic & Health Survey 2011. Addis Ababa and Calverton, MA: Central Statistical Agency and ICF International; 2012. [8] Belay T, Abdullah M, Mengistu H, Ghebrekidan K, Getaneh E, Mare W. Making pregnancy safer initiative-program impact in selected pilot areas. Ethiop Med J 2007; 45(2):115–22. [9] Tayelgn A, Zegeye DT, Kebede Y. Mothers' satisfaction with referral hospital delivery service in Amhara Region. Ethiop BMC Pregnancy Childbirth 2011;11:78. [10] Hailemichael F, Woldie M, Tafese F. Predictors of institutional delivery in Sodo town, Southern Ethiopia. Afr J Primary Health Care Fam Med 2013;5(1):544. [11] Adamasu K, Haile-Mariam A, Bailey P. Indicators for availability, utilization, and quality of emergency obstetric care in Ethiopia, 2008. Int J Gynecol Obstet 2011; 115(1):101–5. [12] Getachew A, Ricca J, Cantor D, Rawlins B, Rosen H, Tekleberhan A, et al. Quality of care for prevention and management of common maternal and newborn complications: A study of Ethiopia's hospitals. http://www.healthynewbornnetwork.org/sites/ default/files/resources/Ethiopia_QoC_formatted_final.pdf. Published 2011. Accessed September 7, 2015. [13] Ministry of Health (Ethiopia). Health Sector Development Program IV 2010/ 11–2014/15. Addis Ababa: Ministry of Health. http://www.nationalplanningcycles. org/sites/default/files/country_docs/Ethiopia/ethiopia_hsdp_iv_final_draft_2010_2015.pdf. Published 2010. Accessed September 7, 2015. [14] United Nations, Department of Economic and Social Affairs, Population Division. World Population Prospects: The 2012 Revision, Highlights and Advance Tables. Working Paper No. ESA/P/WP.228. http://esa.un.org/unpd/wpp/Publications/Files/ WPP2012_HIGHLIGHTS.pdf. Published 2013. Accessed September 7, 2015. [15] van Duong D, Binns CW, Lee AH, Hipgrave DB. Measuring client-perceived quality of maternity services in rural Vietnam. International J Qual Health Care 2004;16(6): 447–52. [16] Fikre AA, Demissie M. Prevalence of institutional delivery and associated factors in Dodota Woreda (district), Oromia regional state. Ethiopia Reprod Health 2012;9:33. [17] Odo DB, Sifti DM. Institutional delivery service utilization and associated factors among child bearing age women in Goba Woreda. Ethiopia J Gynecol Obstet 2014; 2(4):63–70. [18] Gurmesa T, G/Mariam A. Safe delivery service utilization in Metekel Zone, Northwest Ethiopia. Ethiop J Health Sci 2008;17(4):213–22. [19] Ballard K, Gari L, Mosisa H, Wright J. Provision of individualized obstetric risk advice to increase health facility usage by women at risk of a complicated delivery: a cohort study of women in the rural highlands of West Ethiopia. BJOG 2013;120(8):971–8. [20] Promoting Respectful Maternity Care. News from the Last Ten Kilometers. http:// l10k.jsi.com/Resources/Docs/NewsletterVol3No1.pdf. Published 2014. Accessed September 7, 2015. [21] Federal Ministry of Health (Ethiopia). Basic Emergency Obstetric & Newborn Care (BEmONC). Training manual. Addis Ababa: Federal Ministry of Health; 2013. [22] Mirkuzie AH, Sisay MM, Reta AT, Bedane MM. Current evidence on basic emergency obstetric and newborn care services in Addis Ababa, Ethiopia; a cross sectional study. BMC Pregnancy Childbirth 2014;14:354.
Contents lists available at ScienceDirect
International Journal of Gynecology and Obstetrics journal homepage: www.elsevier.com/locate/ijgo
CLINICAL ARTICLE
The effect of prenatal and intrapartum care on the stillbirth rate among women in rural Ethiopia Karen Ballard ⁎, Zelalem Belete, Hirut Kinfu, Mebkyou Tadesse, Mohammed Amin, Habtamu Atnafu Hamlin Fistula Ethiopia, Addis Ababa Fistula Hospital, Addis Ababa, Ethiopia
a r t i c l e
i n f o
Article history: Received 8 June 2015 Received in revised form 12 September 2015 Accepted 26 January 2016 Keywords: Basic emergency obstetric and newborn care Ethiopia Health facility delivery Prenatal care Satisfaction with care Stillbirth
a b s t r a c t Objective: To determine whether community-based prenatal and intrapartum care in Ethiopia results in a lower stillbirth rate. Methods: Between May and December 2014, a randomly selected sample of women in northern and eastern Ethiopia who had delivered a neonate in the preceding 12 months completed a face-to-face survey about their experiences of maternal services and the fetal outcome for each delivery. The stillbirth rates among women delivering at home and at health facilities were compared. Results: Overall, 4442 women completed surveys. Stillbirth was reported by 42 (1.7%) of the 2437 women who had received prenatal care and 53 (2.8%) of the 1921 women who did not receive prenatal care (P = 0.01). The stillbirth rate was similar among women who delivered in a health center (27/1417 [1.9%]), in a hospital (6/126 [4.8%]), and at home (62/2725 [2.3%]; P = 0.13). However, women experiencing an intrapartum emergency were twice as likely to deliver in a health facility (odds ratio 2.6, 95% confidence interval 2.2–3.0). Satisfaction with health-center care was moderately good (median score 77.5/100). Conclusion: The stillbirth rate was reduced among women receiving prenatal care, although delivering in a health facility did not reduce the risk of stillbirth. Improving the quality of health-center care could lead to their planned use for childbirth, which might reduce stillbirth rates. © 2016 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
1. Introduction For decades, Ethiopian women have been delivering their neonates at home, which is perhaps unsurprising because health centers— originally introduced in 1952 as the backbone of the Ethiopian health service—were slow to appear in the rural areas where the vast majority of people live. In 1974, there were just 93 health centers across the whole country [1], and approximately 20 years later, there were only 451 health centers and 119 hospitals for a population of 77 million [2]. It is only in the past 3–4 years that the number of health facilities has increased, although in 2014, there were still only 3245 health centers and 127 hospitals for a population of almost 86 million [3]. The fairly recent appearance of “widespread” healthcare coverage in Ethiopia makes it easy to understand that health centers have been viewed by women as neither necessary nor customary for childbirth and their subsequent preference for a homebirth [4,5]. Consequently, women often perceive the health facility as a place to seek help only in the event of a complication [6], leading to a persistently low institutional delivery rate of just 16% [5] and a maternal mortality ratio of 671 per 100 000 live births [7]. Likewise, the national
stillbirth and perinatal mortality rates have remained high, at 2.1% and 4.6%, respectively [7]. Encouraging delivery in a health facility is hindered by the suboptimal quality of care that women often receive in facilities [4,8] from poorly skilled staff [9,10]. Indeed, in 2008, only 1.3% of Ethiopian health centers were fully functioning as facilities providing basic emergency obstetric and newborn care (BEmONC) [11]. One study in 2010 [12] revealed that just one-third of hospital-based maternal-care staff had knowledge of the signs of obstructed labor and knew how to treat postpartum hemorrhage. Recognizing the inadequacies of the services, in 2010, the Ethiopian Government prioritized maternal health care, increasing the provision of rural health centers, midwives, surgically trained health officers, and ambulances [13]. However, little is known about whether women have embraced these improved maternal health services and whether the care they experience leads to improvements in fetal outcomes. The aim of the present study was to determine women's satisfaction with the community-based maternal services provided in Ethiopia and whether the use of facility-based prenatal and intrapartum care leads to a reduction in stillbirth rates. 2. Materials and methods
⁎ Corresponding author at: Hamlin Fistula Ethiopia, Addis Ababa Fistula Hospital, PO Box 3609, Addis Ababa, Ethiopia. Tel.: +44 7715 487233. E-mail address: [email protected] (K. Ballard).
A survey-based investigation was undertaken in two regions (Amhara and Oromia) of Ethiopia, a country divided into 11 regions
http://dx.doi.org/10.1016/j.ijgo.2015.09.027 0020-7292/© 2016 International Federation of Gynecology and Obstetrics. Published by Elsevier Ireland Ltd. All rights reserved.
K. Ballard et al. / International Journal of Gynecology and Obstetrics 133 (2016) 164–167
and with an estimated population of 91 million [14], and where the vast majority of people are subsistence farmers living in rural areas with limited transport. All prenatal, intrapartum and postnatal care received within government facilities is free, although charges are made for items such as gloves, sutures, and medication. The main source of facility-based maternal care is through the government-run health centers, supported by lower-level care from health posts based within villages and staffed by health extension workers (HEWs). The present study was part of a larger investigation that used a household survey of women's maternal experiences to determine whether a maternal health package introduced into eight health centers in 2010–2011 improved the quality of prenatal and intrapartum services when compared with 18 health centers without the package. The maternal health packages were provided by Hamlin Fistula Ethiopia (HFE), a non-governmental organization offering free care for women with vesicovaginal and rectovaginal fistulas, and BSc midwifery training for students recruited from rural areas of Ethiopia. The packages were introduced in three zones of Ethiopia—South Gondar, West Gojjam, and Eastern Harraghe—and consisted of two midwives trained at HFE who were provided with continuous mentoring, accommodation near to the center, telephone credit, any necessary equipment or drugs, and access to an ambulance for referral to higher-level care. The primary outcome measure was stillbirth rate, which we hypothesized would reduce if high-quality prenatal and intrapartum maternal services were available. To calculate the sample size required for the primary household survey, HFE clinicians advised that the maternal package would be a valuable addition to the community-based maternal services if it were capable of bringing about a 50% reduction in the stillbirth rate. According to the Ethiopian Demographic Health Survey (DHS) [7], the stillbirth rate in Amhara and Oromia is 46 per 1000 pregnancies. A design effect of 1.2 was applied to reduce errors due to cluster sampling techniques, meaning that the total sample size required was 4373 women, with a significance level of 5% and 80% power. To select the sample for the survey, four or five health posts linked with each of the 26 health centers were selected according to their distance from the health center, providing one nearby health post, two a medium distance away, and one far away, leading to a total of 116 health posts. Subsequently, 28 000 households served by the health posts were randomly selected using random number tables from HEW household lists to provide a population of 140 000 women. Given the 3.5% annual pregnancy rate for this population [3], a sample of 4900 women was obtained. Allowing for 10% error in data collection, the required sample size would be achieved. The inclusion criterion was having delivered a neonate after a pregnancy lasting 7 months or more in the previous 12 months. The present study uses data collected from all selected households and does not set out to make comparisons between women served by health centers with and without a maternal health package. Data were obtained between May and December 2014. Following ethical approval from the Oromia and Amhara regional health bureaus, and with permission from the heads of relevant districts, HEWs were recruited and trained to administer the survey. The HEWs were recruited from the health posts based within the villages selected for sampling. This was essential because the HEWs were knowledgeable about the location of all households within the village and how to access what are often very remote areas. Additionally, they were acquainted with most of the women in their village and were female, so able to ask questions that could be considered intimate and sensitive in nature. During the training session, 120 households were randomly selected from each HEW's list for surveying. Because the vast majority of women were unable to read or write, the survey was administered verbally and the HEWs transcribed the answers. Before answering, women were read a participant information sheet, and if they agreed to participate, they were required to sign or place a fingerprint on the consent form. The survey focused on women's use of the maternal services for the delivery in the past 12 months, but also asked about the place of birth
165
for all of their deliveries and the outcome for each neonate. The adapted Haddad 20-item scale [15], which has been validated in countries with similar health systems to Ethiopia, was used to assess satisfaction with the maternal care in the health centers among women using these services. A score of 0–100 was calculated, with 0 representing the lowest possible satisfaction with care and 100 representing the highest. The survey was translated into the appropriate languages and piloted. Throughout the data collection period, a team of researchers visited the sites to perform random data accuracy and quality checks. Each HEW was located by phone and a selection of her completed surveys was readministered at the households to check for accuracy of the originally administered survey. At the end of data collection, the HEWs were paid for their completed surveys, with an additional bonus for accurate and complete datasets. Additionally, records for the 26 health centers serving the surveyed women were assessed by members of the research team to determine the number and skill level of staff, availability of services, number of deliveries and outcomes, and the ability to provide BEmONC. Data were entered into SPSS version 20 (IBM, Armonk, NY, USA). Data were not normally distributed and were therefore reported as medians with interquartile ranges (IQRs). Group differences were tested using the Pearson χ2 test and differences in median scores across the four components of satisfaction scale were tested using the Friedman test. P b 0.05 was considered statistically significant. Odds ratios (ORs) with 95% confidence intervals (CIs) were used to estimate the chances of using health facilities for intrapartum care. 3. Results A total of 4442 completed surveys were available for analysis. The median time travelled by foot to reach a health center was 60 min (IQR 40–120). A total of 1670 (37.6%) women delivered over the preceding 12 months in a health facility, with 1417 (31.9%) delivering in a health center, 126 (2.8%) in a hospital, 121 (2.7%) in a health post, and 6 (0.1%) in a private clinic. Overall, 2725 women (61.3%) delivered at home and 45 (1.0%) delivered on the roadside. Place of delivery was not stated by 2 (b 0.1%) women. The 37.6% facility delivery rate in the preceding 12 months was greater than the 9.3% (338/3642) facility delivery rate reported for the previous delivery (P b 0.001). Indeed, the last neonate delivered was 9 times more likely to be delivered in a health facility compared with the last but one neonate (OR 9.5, 95% CI 7.2–12.5). Data reported across the 26 health centers revealed that 8320 (28.4%) of the estimated 29 300 pregnant women served by the health centers had delivered in one of these health centers in the past 12 months. Overall, 2437 (54.9%) women received prenatal care from their nearest health center, 1040 (42.7%) of whom also delivered in the health center. Women receiving prenatal care from a health professional were significantly more likely to deliver in a health facility than were those who had not received such care (OR 3.5, 95% CI 3.0–3.9). A prepartum or intrapartum emergency in the last delivery was reported by 668 (15.0%) women. The main emergency reported was a prolonged labor of more than 24 h (Table 1). More than half the women delivering at hospital, 28.9% delivering in a health post, and 19.5% delivering in a health center had an obstetric emergency. Delivery in a health facility was more common among women who experienced an obstetric emergency than among those who did not (382/668 [57.2%] vs 1261/3774 [33.4%]; P b 0.001). The odds of delivering in a health facility if experiencing an obstetric emergency were 2.6 (95% CI 2.2–3.0). Overall, 33 (0.7%) women reported having a cesarean delivery, all of which were performed in a hospital setting. Of the 8320 deliveries performed in and reported by the 26 health centers, 292 (3.5%) were vacuum assisted. Overall, 95 (2.1%) women reported that their most recent pregnancy ended in a stillbirth. This rate is considerably lower than the
166
K. Ballard et al. / International Journal of Gynecology and Obstetrics 133 (2016) 164–167
Table 1 Prepartum and intrapartum emergencies.a
Any prepartum or intrapartum emergency Prolonged labor (N24 h) Ruptured uterus Eclampsia Prepartum hemorrhage
Delivered in the health post (n = 121)
Delivered at a private clinic (n = 6)
Delivered at the health center (n = 1417)
Delivered at hospital (n = 126)
Delivered at the roadside (n = 45)
668 (15.0)
277 (10.2)
35 (28.9)
0
277 (19.5)
70 (55.6)
9 (20.0)
611 (13.8) 23 (0.5) 30 (0.7) 33 (0.7)
249 (9.1) 12 (0.4) 14 (0.5) 12 (0.4)
31 (25.6) 1 (0.8) 3 (2.5) 2 (1.7)
0 0 0 0
260 (18.3) 7 (0.5) 9 (0.6) 12 (0.8)
63 (50.0) 3 (2.4) 3 (2.4) 6 (4.8)
8 (17.8) 0 1 (2.2) 1 (2.2)
Values are given as number (percentage). Data missing for 2 women.
13.6% (74/544) stillbirth rate reported for the last but six pregnancy (P b 0.001). The stillbirth rate has declined as the rate of delivery in health institutions has risen (Fig. 1). For neonates born in the 12 months preceding the survey, stillbirths were recorded for 62 (2.3%) of the 2725 home deliveries, 27 (1.9%) of the 1417 deliveries in health centers, and 6 (4.8%) of the 126 hospital deliveries (P = 0.13). Stillbirth was reported by 42 (1.7%) of the 2437 women who received prenatal care and 53 (2.8%) of the 1921 women who did not receive prenatal care (P = 0.01). As previously mentioned, women who experienced an obstetric emergency were more likely to seek help from the health facility. Stillbirth occurred among 22 (3.3%) of the 668 women experiencing an emergency and among 71 (1.9%) of the 3774 women not experiencing an emergency (P = 0.024). Among women who delivered in a health facility, 24 (1.9%) of the 1261 women who did not experience an obstetric emergency had a stillbirth compared with 9 (2.4%) of the 382 women who did experience an emergency (P = 0.58). Although the absolute numbers are too small to determine significant relationships between the rate of institutional stillbirths and complicated deliveries, it is likely that the inevitable delays that occur in reaching a health facility increase the stillbirth rate within the health facilities. The data reported by the 26 health centers revealed a total of 206 (2.5%) stillbirths among the 8320 deliveries that occurred in these centers. The health centers served a total population of 837 131, and there were 49 midwives working across these centers, providing a midwifeto-population ratio of 1:17 084. Nine (34.6%) health centers had running water, 19 (73.1%) had intermittent electricity, and 14 (53.8%) provided BEmONC. Women's satisfaction with the care received when delivering at the health center (n = 1417) was moderately good, with a median score of 77.5 (IQR 72–86). Women were more satisfied with Still birth
access to care (median score 80, IQR 75–85) and healthcare delivery (median 79, IQR 75–88) than they were with the quality of interpersonal care (median 75, IQR 67–88) and the quality of the health facility and its resources (median 75, IQR 67–83; P b 0.001). 4. Discussion More than one-third of women surveyed in the present study had had their most recent delivery in a health facility, reflecting a dramatic increase over the past 2 or 3 years. It is difficult to make direct comparisons with data from the DHS, which refer to women who delivered in the preceding 5 years [7]. Nevertheless, the Ethiopian mini-DHS has shown an increase in institutional delivery rate from 10% in 2011 to 16% in 2014 [5,7], and national data collected directly from health facilities in 2012–2013 [3] established that 23.1% of deliveries were performed by skilled birth attendants. Other data from small, local studies also suggest that there have been substantial improvements in facility delivery rates, with data collected in 2009–2011 reporting an institutional delivery rate of 18.2% [16], and data from 2013 revealing a facility delivery rate of 47% [17]. In addition to the increased facility delivery rate, the present study highlights a concomitant decrease in stillbirth rates. The 2011 Ethiopian DHS reported a 4.8% stillbirth rate in the Oromia region and 4.7% in the Amhara region [7], so the stillbirth rate reported here represents a sizeable reduction, which could be partly explained by the increased use of health facilities for prenatal and intrapartum care. Indeed, it was found that women accessing prenatal care at the health centers had a lower stillbirth rate than did those who did not access such care. However, as in other studies [18,19], it was found that Institutional delivery
40.0% 35.0% 25.0% 20.0% 15.0% 10.0% 5.0%
Previous Deliveries Fig. 1. Percentage of women delivering in a health center and experiencing stillbirth.
Last pregnancy
Last but 1 pregnancy
Last but 4 pregnancy
Last but 5 pregnancy
0.0%
Last but 6 pregnancy
Percentage
30.0%
Last but 2 pregnancy
b
Delivered at home (n = 2725)
Last but 3 pregnancy
a
All participants (n = 4440)b
K. Ballard et al. / International Journal of Gynecology and Obstetrics 133 (2016) 164–167
many women only used the health facilities once an obstetric emergency had arisen, limiting the effectiveness of the care they received. The present finding that women delivering at home had similar stillbirth rates to those delivering at a health center illustrates the limited care that the health centers were able to provide for the unborn fetus. Improvements in the quality of care offered by the health centers will do much to encourage women to seek care during the early stages of labor. Similarly to other findings [8], the present data on satisfaction with the maternal services indicate a need for improvements in interpersonal care, and the awaited results of an in-depth study on disrespect and abuse in maternal health care in Ethiopia [20] should provide the government with valuable direction about ways to best improve the quality of care. The present study also highlighted the need for improvements in the level of care being provided at health centers, with approximately half the facilities being able to provide BEmONC. In 2008, the Ethiopian Government led a national assessment of both basic and comprehensive emergency obstetric and newborn care [11], revealing that only 1.3% of the health centers were full BEmONC providers. Subsequently, a standardized curriculum for BEmONC training was developed [21], increasing the availability of emergency obstetric and neonatal care within health centers to 56% [3]. However, an evaluation of 10 health centers in Addis Ababa in 2013 revealed that only 12% of staff had received BEmONC training and that their overall knowledge and skill in providing emergency care remained poor [22]. Although the sampling methods applied in the present study allow for the findings to be representative of women living in both rural and urban areas, they can reflect only the views and experiences of women living in the two regions studied. Data accuracy is always of concern when using self-reported data. This is of particular importance in Ethiopia, where both women and health workers have received health messages about the importance of delivering in a health facility that could have provoked a degree of moral judgment surrounding home deliveries. However, the delivery and stillbirth rates obtained from the survey data were similar to those retrieved from the health center records. Additionally, the extensive data verification process encouraged data collectors to be accurate. The presence of the research team in the study sites throughout the data collection period and the offer of a financial incentive for high-quality data will also hopefully have contributed to maximizing data quality. In conclusion, the present study has shown that women who access prenatal care have a lower stillbirth rate than do those who do not access such care, but delivery in a health facility does not lower the stillbirth rate. However, many women delay seeking help from the health center until an emergency arises, which is likely to limit the care that health centers can provide. Women's satisfaction with the maternal care they received was moderately good and improvements could motivate them to actively seek care from the facilities rather than relying on them for emergencies only. In turn, this is likely to lead to a reduction in stillbirth rates. Acknowledgments The project was supported by resources from HFE and external funding from Johnson and Johnson. The funders had no role in the study design, data collection and analysis, decision to publish, or the preparation of the manuscript.
167
Conflict of interest The authors have no conflicts of interest.
References [1] Meche H. The Development of Health Services in Ethiopia. In: Koos H, Zein ZA, editors. The Ecology of Health and Disease in Ethiopia. Addis Ababa: Ministry of Health; 1988. p. 57–66. [2] Ministry of Health (Ethiopia). Health and health related indicators. Addis Ababa: Ministry of Health; 2006/7. http://cnhde.ei.columbia.edu/files/5/6/2008-Feb-Fri102213.pdf. Published 2007. Accessed September 7, 2015. [3] Ministry of Health (Ethiopia). Health and health related indicators. Addis Ababa: Ministry of Health; 2012/13. http://www.moh.gov.et/documents/26765/0/Health+ and+Health+Related+Indicators+2005+E.C/1b5b2a9f-a960-4024-8d92-51919 5364023?version=1.0. Published 2014. Accessed September 7, 2015. [4] Shiferaw S, Spigt M, Godefrooij M, Melkamu Y, Tekie M. Why do women prefer home births in Ethiopia? BMC Pregnancy Childbirth 2013;13:5. [5] Central Statistical Agency (Ethiopia). Ethiopia Mini Demographic and Health Survey 2014; Addis Ababa: Central Statistical Agency. http://www.unicef.org/ethiopia/ Mini_DHS_2014__Final_Report.pdf. Published 2014. Accessed September 7, 2015. [6] Bedford J, Gandhi M, Admassu M, Girma A. ‘A normal delivery takes place at home’: a qualitative study of the location of childbirth in rural Ethiopia. Matern Child Health J 2013;17(2):230–9. [7] Central Statistical Agency (Ethiopia). Ethiopia Demographic & Health Survey 2011. Addis Ababa and Calverton, MA: Central Statistical Agency and ICF International; 2012. [8] Belay T, Abdullah M, Mengistu H, Ghebrekidan K, Getaneh E, Mare W. Making pregnancy safer initiative-program impact in selected pilot areas. Ethiop Med J 2007; 45(2):115–22. [9] Tayelgn A, Zegeye DT, Kebede Y. Mothers' satisfaction with referral hospital delivery service in Amhara Region. Ethiop BMC Pregnancy Childbirth 2011;11:78. [10] Hailemichael F, Woldie M, Tafese F. Predictors of institutional delivery in Sodo town, Southern Ethiopia. Afr J Primary Health Care Fam Med 2013;5(1):544. [11] Adamasu K, Haile-Mariam A, Bailey P. Indicators for availability, utilization, and quality of emergency obstetric care in Ethiopia, 2008. Int J Gynecol Obstet 2011; 115(1):101–5. [12] Getachew A, Ricca J, Cantor D, Rawlins B, Rosen H, Tekleberhan A, et al. Quality of care for prevention and management of common maternal and newborn complications: A study of Ethiopia's hospitals. http://www.healthynewbornnetwork.org/sites/ default/files/resources/Ethiopia_QoC_formatted_final.pdf. Published 2011. Accessed September 7, 2015. [13] Ministry of Health (Ethiopia). Health Sector Development Program IV 2010/ 11–2014/15. Addis Ababa: Ministry of Health. http://www.nationalplanningcycles. org/sites/default/files/country_docs/Ethiopia/ethiopia_hsdp_iv_final_draft_2010_2015.pdf. Published 2010. Accessed September 7, 2015. [14] United Nations, Department of Economic and Social Affairs, Population Division. World Population Prospects: The 2012 Revision, Highlights and Advance Tables. Working Paper No. ESA/P/WP.228. http://esa.un.org/unpd/wpp/Publications/Files/ WPP2012_HIGHLIGHTS.pdf. Published 2013. Accessed September 7, 2015. [15] van Duong D, Binns CW, Lee AH, Hipgrave DB. Measuring client-perceived quality of maternity services in rural Vietnam. International J Qual Health Care 2004;16(6): 447–52. [16] Fikre AA, Demissie M. Prevalence of institutional delivery and associated factors in Dodota Woreda (district), Oromia regional state. Ethiopia Reprod Health 2012;9:33. [17] Odo DB, Sifti DM. Institutional delivery service utilization and associated factors among child bearing age women in Goba Woreda. Ethiopia J Gynecol Obstet 2014; 2(4):63–70. [18] Gurmesa T, G/Mariam A. Safe delivery service utilization in Metekel Zone, Northwest Ethiopia. Ethiop J Health Sci 2008;17(4):213–22. [19] Ballard K, Gari L, Mosisa H, Wright J. Provision of individualized obstetric risk advice to increase health facility usage by women at risk of a complicated delivery: a cohort study of women in the rural highlands of West Ethiopia. BJOG 2013;120(8):971–8. [20] Promoting Respectful Maternity Care. News from the Last Ten Kilometers. http:// l10k.jsi.com/Resources/Docs/NewsletterVol3No1.pdf. Published 2014. Accessed September 7, 2015. [21] Federal Ministry of Health (Ethiopia). Basic Emergency Obstetric & Newborn Care (BEmONC). Training manual. Addis Ababa: Federal Ministry of Health; 2013. [22] Mirkuzie AH, Sisay MM, Reta AT, Bedane MM. Current evidence on basic emergency obstetric and newborn care services in Addis Ababa, Ethiopia; a cross sectional study. BMC Pregnancy Childbirth 2014;14:354.