A decade of progress providing safe abortion services in Ethiopia: Results of national assessments in 2008 and 2014

BMC Pregnancy and Childbirth, Volume 17, No. 1, Year 2017

Interpretation

Background: Ethiopia has one of the highest maternal mortality ratios in the world (420 per 100,000 live births in 2013), and unsafe abortion continues to be one of the major causes. To reduce deaths and disabilities from unsafe abortion, Ethiopia liberalized its abortion law in 2005 to allow safe abortion under certain conditions. This study aimed to measure how availability and utilization of safe abortion services has changed in the last decade in Ethiopia. Methods: This paper draws on results from nationally representative health facility studies conducted in Ethiopia in 2008 and 2014. The data come from three sources at two points in time: 1) interviews with 335 health providers in 2008 and 822 health care providers in 2014, 2) review of facility logbooks, and 3) prospective data on 3092 women in 2008 and 5604 women in 2014 seeking treatment for abortion complications or induced abortion over a one month period. The Safe Abortion Care Model was used as a framework of analysis. Results: There has been a rapid expansion of health facilities eligible to provide legal abortion services in Ethiopia since 2008. Between 2008 and 2014, the number of facilities reporting basic and comprehensive signal functions for abortion care increased. In 2014, access to basic abortion care services exceeded the recommended level of available facilities providing the service, increasing from 25 to 117%, with more than half of regions meeting the recommended level. Comprehensive abortion services increased from 20% of the recommended level in 2008 to 38% in 2014. Smaller regions and city administrations achieved or exceeded the recommended level of comprehensive service facilities, yet larger regions fall short. Between 2008 and 2014, the use of appropriate technology for conducting first and second trimester abortion and the provision of post abortion family planning has increased at the same time that abortion-related obstetric complications have decreased. Conclusion: Ten years after the change in abortion law, service availability and quality has increased, but access to lifesaving comprehensive care still falls short of recommended levels. Data used for this study come from nationally representative cross-sectional health facility studies conducted in Ethiopia in 2008 and 2014 as part of a national study on abortion incidence and morbidity. Two of the data sources used for determining abortion incidence and morbidity were used for this analysis: a national cross-sectional Health Facility Survey (HFS) and a Prospective Morbidity Survey (PMS). Data collection for the 2008 study occurred between November 2007 and February 2008, and the full methodology is described and published in Gebreselassie et al. [7] and Singh et al. [8]. The 2014 study data were collected from December 2013 to April 2014 and the methodology is described in detail by Gebrehiwot et al. [9, 13]. In both 2008 and 2014, the health facility survey included log-book reviews to collect data on service utilization and quality required for the SAC model. Ethical approval was obtained from the National Health Research Ethics Review Committee at the Ministry of Science and Technology of Ethiopia, Addis Ababa and from the Guttmacher Institute ethical review committee. Informed oral consent was obtained from every participating patient in the PMS and from the health providers in the HFS. To construct the sampling frame, a list of health facilities was obtained from the Ministry of Health in 2008 and the Food, Medicine and Health Care Administration and Control Authority of Ethiopia (FMHACA) in 2014. This list included public hospitals, public health centers, private hospitals, private higher clinics, and NGO reproductive health clinics. Private medium clinics were included in 2014, but not in 2008. In both years, health facilities were eligible for inclusion if they were authorized to provide treatment for abortion complications or induced abortion services, according to the Technical and Procedural Guidelines for Abortion Care [6, 14]. The sampling frame consisted of 896 eligible facilities in 2008 and 4287 facilities in 2014. Multi-stage stratified random sampling was used. The first strata, region (n = 11), and the second, facility type and ownership, were used (33 in 2008 and 77 in 2014) to draw a nationally and geographically representative sample using probability-based sampling. Higher level facilities known to provide abortion and maternity-related services such as hospitals were sampled at a higher fraction, while health centers and medium private facilities were sampled at a relatively lower fraction. Sampling fractions ranged between 0.12 and 1.00 of all eligible facilities. As mentioned above, two data collection methods and tools were used to measure facilities’ capacity to provide safe abortion services both in 2008 and 2014: the Health Facility Survey and the Prospective Morbidity Survey. For the HFS, trained data collectors visited each selected facility and interviewed the most senior abortion care provider or someone knowledgeable about induced abortion and post-abortion care services. Respondents included obstetrician/gynecologists, general practitioners, midwifes, nurses, and other health workers. The HFS interviewers obtained information on the facility infrastructure, the facility’s capacity to provide abortion services, type of abortion care provided and monthly caseload of abortion patients, provider attitudes towards abortion service provision, and the facility’s ability to perform each of the EmOC and SAC signal functions during the previous 3 month period. The HFS interviewers also recorded logbook information on cases of abortion and obstetric complications recorded at each facility for the previous year. Collected information from the facilities’ abortion logbooks included the number of abortion cases, severity of abortion complications, the abortion method used, and provision of post-abortion contraception. Following the method used in Abdella et al. [10] logbook review captured women seeking an induced abortion, and women treated for complications of miscarriage or abortion induced outside the health facility and presenting for treatment. The interviewers also collected logbook data on the number of obstetric complications including hemorrhage, sepsis, prolonged or obstructed labour, complications from abortion, pre-eclampsia or eclampsia, ectopic pregnancy or ruptured uterus. The HFS data were assigned the appropriate facility weight. A total of 335 facilities in 2008 and 822 facilities in 2014 participated in the HFS, with participation rates of 96 and 89% for public hospitals and health centers respectively in 2008, and 98 and 93% respectively in 2014. For the PMS, data were collected prospectively on all women presenting with abortion complications or requests for a legal induced abortion during a consecutive 30-day period in each study facility. Providers in the facilities were trained to record information on each woman requesting a safe induced abortion or seeking PAC either for complications of an unsafe abortion or miscarriage. Data were also recorded on patient demographics, self-reported induction attempts, reproductive history, vital signs, symptoms found on physical examination, abortion symptoms that drew the woman to the facility, and essential elements of the care provided to her. Patients were not interviewed directly on these questions—it was information provided as part of patient intake and treatment. A total of 344 facilities in 2008 and 569 facilities in 2014 participated in the PMS. The PMS included data on 3092 women in 2008 and 5604 women in 2014. More details on the sampling, weighting, data collection tool, and participation rate of the facilities for both the HFS and PMS have been published in Moore et al. [7] and Gebrehiwot et al. [13]. Data from the 2008 and 2014 studies were entered and checked for consistency and completeness using EpiData version 3.1. Datasets from both years were imported into, cleaned, and analyzed both separately and combined using Stata version 13.1. Each of the strata were then weighted for non-response and their probability of selection to achieve national representation. All analyses accounted for the complex sample design and appropriate survey weighting, and are presented by year and by facility type (hospitals or health centers). Missing data were limited, and so analysis was limited to non-missing responses. Analysis techniques conducted and described in Abdella et al. [10] were repeated to reanalyze the 2008 data from public sector facilities only, disaggregating by facility type. The Health Facility Survey sample from both years, unweighted and weighted frequencies as well as weighted proportions, are presented by public facility type in Table 1. Ninety public hospitals and 152 public health centers were included in the sample in 2008; 117 hospitals and 368 health centers were included in the sample in 2014. Distribution of 2008 and 2014 sample in Ethiopia by public facility typea, b aPercentages are weighted to be nationally representative bResults differ slightly from those published in Abdella et al., [10] due to reanalysis Performance of basic and comprehensive SAC signal functions were analyzed descriptively. Weighted frequencies and proportions of public hospitals’ and health centers’ reported performance of each signal function are presented. Basic SAC delivery was calculated including all public facilities, and comprehensive SAC delivery was limited to public hospitals only. Safe abortion indicators and their 95% confidence intervals are presented by facility level using weighted frequencies and percentages from both the HFS and PMS data. The SAC model and estimates of 2008 and 2014 Ethiopia regional population sizes using the latest census and projections [15, 16] were used to calculate recommended levels of basic and comprehensive SAC delivery in both years, by region and nationally. Weighted frequencies of basic and comprehensive SAC delivery were used to calculate the percentage of the recommended level of service delivery achieved for both service types. In both years, log book review data from the HFS were used to calculate obstetric complications, estimates of either the numerator or denominator, were collected only on the HFS; the rest of the indicators come from the PMS which were used to collect client level data on each woman’s presentation, treatment and care. In 2008 only, data on post-abortion contraception was only collected in the HFS. For these indicators, the HFS and PMS datasets were combined, and the HFS estimates were scaled to match the PMS estimate of the scale of difference between variables in the two datasets. The final SAC indicator estimates are presented as percentages of the weighted frequencies along with confidence intervals. Appropriate technology indicators were defined according to WHO recommendations – vacuum aspiration and medical abortion in the first trimester and dilatation and evacuation (D&E) and medical abortion in the second trimester – with the exception of D&E in the second trimester [17]. This decision was made because even by the 2014 study, no clinical trainings on D&E had been provided for abortion providers in Ethiopia, and the supplies for this procedure are not available in the country. Yet, D&E is often used to describe procedures using sharp curettage, a technology which is not recommended by the WHO [17].

AI Digest

Study Justification:

– Ethiopia has one of the highest maternal mortality ratios in the world, with unsafe abortion being a major cause.
– The study aimed to measure the changes in availability and utilization of safe abortion services in Ethiopia over the past decade.
– The findings can inform policy and decision-making to further improve access to safe abortion services and reduce maternal mortality.

Study Highlights:

– Rapid expansion of health facilities eligible to provide legal abortion services in Ethiopia since 2008.
– Access to basic abortion care services exceeded the recommended level in 2014, with more than half of regions meeting the recommended level.
– Comprehensive abortion services increased from 20% of the recommended level in 2008 to 38% in 2014.
– Use of appropriate technology for conducting first and second trimester abortion and provision of post-abortion family planning has increased.
– Abortion-related obstetric complications have decreased.

Study Recommendations:

– Despite improvements, access to comprehensive care still falls short of recommended levels.
– Efforts should be made to ensure that larger regions have the recommended level of comprehensive service facilities.
– Continued training and provision of appropriate technology for abortion procedures should be prioritized.
– Post-abortion contraception services should be further strengthened.

Key Role Players:

– Ministry of Health in Ethiopia
– National Health Research Ethics Review Committee
– Guttmacher Institute ethical review committee
– Health care providers (obstetrician/gynecologists, general practitioners, midwives, nurses, etc.)
– Facility administrators and managers

Cost Items for Planning Recommendations:

– Training programs for health care providers on safe abortion procedures and post-abortion contraception
– Procurement of appropriate technology and supplies for abortion procedures
– Implementation of quality assurance measures for safe abortion services
– Monitoring and evaluation activities to assess the impact of the recommendations

Strength of Evidence

The strength of evidence for this abstract is 8 out of 10.
The evidence in the abstract is strong, as it is based on nationally representative health facility studies conducted in Ethiopia in 2008 and 2014. The data sources include interviews with health providers, review of facility logbooks, and prospective data on women seeking treatment for abortion complications or induced abortion. The study methodology is described in detail, including the sampling frame, data collection methods, and analysis techniques. However, to improve the evidence, the abstract could provide more information on the sample size and representativeness of the study population, as well as the statistical significance of the findings. Additionally, it would be helpful to include information on any limitations or potential biases in the study.

Abstract

Data used for this study come from nationally representative cross-sectional health facility studies conducted in Ethiopia in 2008 and 2014 as part of a national study on abortion incidence and morbidity. Two of the data sources used for determining abortion incidence and morbidity were used for this analysis: a national cross-sectional Health Facility Survey (HFS) and a Prospective Morbidity Survey (PMS). Data collection for the 2008 study occurred between November 2007 and February 2008, and the full methodology is described and published in Gebreselassie et al. [7] and Singh et al. [8]. The 2014 study data were collected from December 2013 to April 2014 and the methodology is described in detail by Gebrehiwot et al. [9, 13]. In both 2008 and 2014, the health facility survey included log-book reviews to collect data on service utilization and quality required for the SAC model. Ethical approval was obtained from the National Health Research Ethics Review Committee at the Ministry of Science and Technology of Ethiopia, Addis Ababa and from the Guttmacher Institute ethical review committee. Informed oral consent was obtained from every participating patient in the PMS and from the health providers in the HFS. To construct the sampling frame, a list of health facilities was obtained from the Ministry of Health in 2008 and the Food, Medicine and Health Care Administration and Control Authority of Ethiopia (FMHACA) in 2014. This list included public hospitals, public health centers, private hospitals, private higher clinics, and NGO reproductive health clinics. Private medium clinics were included in 2014, but not in 2008. In both years, health facilities were eligible for inclusion if they were authorized to provide treatment for abortion complications or induced abortion services, according to the Technical and Procedural Guidelines for Abortion Care [6, 14]. The sampling frame consisted of 896 eligible facilities in 2008 and 4287 facilities in 2014. Multi-stage stratified random sampling was used. The first strata, region (n = 11), and the second, facility type and ownership, were used (33 in 2008 and 77 in 2014) to draw a nationally and geographically representative sample using probability-based sampling. Higher level facilities known to provide abortion and maternity-related services such as hospitals were sampled at a higher fraction, while health centers and medium private facilities were sampled at a relatively lower fraction. Sampling fractions ranged between 0.12 and 1.00 of all eligible facilities. As mentioned above, two data collection methods and tools were used to measure facilities’ capacity to provide safe abortion services both in 2008 and 2014: the Health Facility Survey and the Prospective Morbidity Survey. For the HFS, trained data collectors visited each selected facility and interviewed the most senior abortion care provider or someone knowledgeable about induced abortion and post-abortion care services. Respondents included obstetrician/gynecologists, general practitioners, midwifes, nurses, and other health workers. The HFS interviewers obtained information on the facility infrastructure, the facility’s capacity to provide abortion services, type of abortion care provided and monthly caseload of abortion patients, provider attitudes towards abortion service provision, and the facility’s ability to perform each of the EmOC and SAC signal functions during the previous 3 month period. The HFS interviewers also recorded logbook information on cases of abortion and obstetric complications recorded at each facility for the previous year. Collected information from the facilities’ abortion logbooks included the number of abortion cases, severity of abortion complications, the abortion method used, and provision of post-abortion contraception. Following the method used in Abdella et al. [10] logbook review captured women seeking an induced abortion, and women treated for complications of miscarriage or abortion induced outside the health facility and presenting for treatment. The interviewers also collected logbook data on the number of obstetric complications including hemorrhage, sepsis, prolonged or obstructed labour, complications from abortion, pre-eclampsia or eclampsia, ectopic pregnancy or ruptured uterus. The HFS data were assigned the appropriate facility weight. A total of 335 facilities in 2008 and 822 facilities in 2014 participated in the HFS, with participation rates of 96 and 89% for public hospitals and health centers respectively in 2008, and 98 and 93% respectively in 2014. For the PMS, data were collected prospectively on all women presenting with abortion complications or requests for a legal induced abortion during a consecutive 30-day period in each study facility. Providers in the facilities were trained to record information on each woman requesting a safe induced abortion or seeking PAC either for complications of an unsafe abortion or miscarriage. Data were also recorded on patient demographics, self-reported induction attempts, reproductive history, vital signs, symptoms found on physical examination, abortion symptoms that drew the woman to the facility, and essential elements of the care provided to her. Patients were not interviewed directly on these questions—it was information provided as part of patient intake and treatment. A total of 344 facilities in 2008 and 569 facilities in 2014 participated in the PMS. The PMS included data on 3092 women in 2008 and 5604 women in 2014. More details on the sampling, weighting, data collection tool, and participation rate of the facilities for both the HFS and PMS have been published in Moore et al. [7] and Gebrehiwot et al. [13]. Data from the 2008 and 2014 studies were entered and checked for consistency and completeness using EpiData version 3.1. Datasets from both years were imported into, cleaned, and analyzed both separately and combined using Stata version 13.1. Each of the strata were then weighted for non-response and their probability of selection to achieve national representation. All analyses accounted for the complex sample design and appropriate survey weighting, and are presented by year and by facility type (hospitals or health centers). Missing data were limited, and so analysis was limited to non-missing responses. Analysis techniques conducted and described in Abdella et al. [10] were repeated to reanalyze the 2008 data from public sector facilities only, disaggregating by facility type. The Health Facility Survey sample from both years, unweighted and weighted frequencies as well as weighted proportions, are presented by public facility type in Table 1. Ninety public hospitals and 152 public health centers were included in the sample in 2008; 117 hospitals and 368 health centers were included in the sample in 2014. Distribution of 2008 and 2014 sample in Ethiopia by public facility typea, b aPercentages are weighted to be nationally representative bResults differ slightly from those published in Abdella et al., [10] due to reanalysis Performance of basic and comprehensive SAC signal functions were analyzed descriptively. Weighted frequencies and proportions of public hospitals’ and health centers’ reported performance of each signal function are presented. Basic SAC delivery was calculated including all public facilities, and comprehensive SAC delivery was limited to public hospitals only. Safe abortion indicators and their 95% confidence intervals are presented by facility level using weighted frequencies and percentages from both the HFS and PMS data. The SAC model and estimates of 2008 and 2014 Ethiopia regional population sizes using the latest census and projections [15, 16] were used to calculate recommended levels of basic and comprehensive SAC delivery in both years, by region and nationally. Weighted frequencies of basic and comprehensive SAC delivery were used to calculate the percentage of the recommended level of service delivery achieved for both service types. In both years, log book review data from the HFS were used to calculate obstetric complications, estimates of either the numerator or denominator, were collected only on the HFS; the rest of the indicators come from the PMS which were used to collect client level data on each woman’s presentation, treatment and care. In 2008 only, data on post-abortion contraception was only collected in the HFS. For these indicators, the HFS and PMS datasets were combined, and the HFS estimates were scaled to match the PMS estimate of the scale of difference between variables in the two datasets. The final SAC indicator estimates are presented as percentages of the weighted frequencies along with confidence intervals. Appropriate technology indicators were defined according to WHO recommendations – vacuum aspiration and medical abortion in the first trimester and dilatation and evacuation (D&E) and medical abortion in the second trimester – with the exception of D&E in the second trimester [17]. This decision was made because even by the 2014 study, no clinical trainings on D&E had been provided for abortion providers in Ethiopia, and the supplies for this procedure are not available in the country. Yet, D&E is often used to describe procedures using sharp curettage, a technology which is not recommended by the WHO [17].

Materials

N/A

Innovations

Based on the provided information, here are some potential innovations that can be used to improve access to maternal health:

1. Telemedicine: Implementing telemedicine services can help overcome geographical barriers and provide remote access to healthcare professionals. This can be particularly useful in rural areas where access to maternal health services is limited.

2. Mobile health (mHealth) applications: Developing mobile applications that provide information and resources related to maternal health can empower women with knowledge and enable them to make informed decisions about their healthcare. These apps can also provide reminders for prenatal care appointments and medication schedules.

3. Community health workers: Training and deploying community health workers who can provide basic maternal health services, education, and support in underserved areas can help bridge the gap in access to healthcare.

4. Task-shifting: Expanding the roles and responsibilities of healthcare workers, such as nurses and midwives, to include certain aspects of maternal health care can help alleviate the shortage of skilled healthcare professionals.

5. Mobile clinics: Setting up mobile clinics that travel to remote areas can bring essential maternal health services closer to communities that lack access to healthcare facilities.

6. Public-private partnerships: Collaborating with private healthcare providers to improve access to maternal health services can help leverage their resources and expertise.

7. Health financing schemes: Implementing innovative health financing schemes, such as community-based health insurance or conditional cash transfer programs, can help reduce financial barriers to accessing maternal health services.

8. Quality improvement initiatives: Implementing quality improvement initiatives in healthcare facilities can enhance the overall quality of maternal health services, ensuring that women receive safe and effective care.

9. Health education and awareness campaigns: Conducting targeted health education and awareness campaigns can help increase knowledge about maternal health issues and promote the importance of seeking timely care.

10. Strengthening referral systems: Improving the coordination and effectiveness of referral systems between primary healthcare facilities and higher-level hospitals can ensure that women receive appropriate and timely care, especially in emergency situations.

It is important to note that the specific context and needs of each community should be taken into consideration when implementing these innovations to ensure their effectiveness and sustainability.

AI Innovations Description

Based on the information provided, the recommendation to improve access to maternal health in Ethiopia is to continue expanding the availability and utilization of safe abortion services. The study shows that there has been a rapid expansion of health facilities eligible to provide legal abortion services since 2008. However, while there has been an increase in access to basic abortion care services, access to comprehensive abortion care still falls short of recommended levels.

To address this, it is recommended to focus on increasing the availability of comprehensive abortion services, particularly in larger regions where access is currently limited. This can be achieved by training more healthcare providers in comprehensive abortion care and ensuring that facilities have the necessary equipment and supplies to provide these services. Additionally, efforts should be made to improve the provision of post-abortion family planning to ensure that women have access to contraception after receiving abortion care.

Overall, the recommendation is to continue building on the progress made in the last decade by further expanding and improving access to safe abortion services, with a particular focus on comprehensive care. This will help reduce maternal mortality and improve maternal health outcomes in Ethiopia.

AI Innovations Methodology

Based on the provided description, here are some potential recommendations to improve access to maternal health:

1. Increase the number of health facilities providing safe abortion services: The study found that there has been a rapid expansion of health facilities eligible to provide legal abortion services in Ethiopia since 2008. However, access to comprehensive care still falls short of recommended levels. Increasing the number of health facilities that can provide safe abortion services can help improve access to maternal health.

2. Improve the quality of abortion care services: While the availability of abortion care services has increased, it is important to ensure that the quality of care provided is also improved. This can be done through training programs for healthcare providers, ensuring the availability of appropriate technology for conducting abortions, and promoting the provision of post-abortion family planning.

3. Address regional disparities in access to comprehensive abortion care: The study found that smaller regions and city administrations achieved or exceeded the recommended level of comprehensive service facilities, while larger regions fell short. Efforts should be made to address these regional disparities and ensure that all regions have access to comprehensive abortion care services.

To simulate the impact of these recommendations on improving access to maternal health, a methodology could be developed as follows:

1. Define the indicators: Identify key indicators that can measure the impact of the recommendations on improving access to maternal health. These indicators could include the number of health facilities providing safe abortion services, the quality of care provided, and the availability of comprehensive abortion care services in different regions.

2. Collect baseline data: Gather data on the current status of access to maternal health in Ethiopia, including the number of health facilities providing safe abortion services, the quality of care provided, and the availability of comprehensive abortion care services in different regions. This data can be obtained through surveys, interviews, and reviews of health facility records.

3. Develop a simulation model: Create a simulation model that can estimate the potential impact of the recommendations on improving access to maternal health. This model should take into account factors such as population size, healthcare infrastructure, and regional disparities.

4. Input data and run simulations: Input the baseline data into the simulation model and run simulations to estimate the impact of the recommendations. This can be done by adjusting the variables related to the number of health facilities, quality of care, and regional disparities, and observing the changes in the indicators.

5. Analyze results: Analyze the results of the simulations to determine the potential impact of the recommendations on improving access to maternal health. This can be done by comparing the simulated values of the indicators with the baseline values.

6. Refine and validate the model: Refine the simulation model based on the results and feedback from experts in the field. Validate the model by comparing the simulated values with real-world data, if available.

7. Communicate findings: Present the findings of the simulation study in a clear and concise manner, highlighting the potential impact of the recommendations on improving access to maternal health. This can be done through reports, presentations, or other communication channels.

By following this methodology, policymakers and healthcare providers can gain insights into the potential impact of different recommendations on improving access to maternal health and make informed decisions to address the challenges in this area.

Author

Dima Health

Statistics:

Citations: 24

Identifiers:

DOI: 10.1186/s12884-017-1266-z

Research Areas:

Food Security, Health System and Policy, Maternal and Child Health, Quality of Care, Sexual and Reproductive Health, Social Determinants, Technology and Innovations, Workforce

Study Countries:

Ethiopia

Study Design:

Cohort Study, Cross Sectional Study, Grounded Theory

Study Approach:

Quantitative

Participants Gender:

Female