International Journal for Quality in Health Care Advance Access originally published online on June 15, 2007
International Journal for Quality in Health Care 2007 19(4):250-256; doi:10.1093/intqhc/mzm023
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Emergency obstetric care in the southernmost provinces of Thailand *
1 Department of Obstetrics and Gynecology, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla 90110, Thailand
2 Obstetrics & Gynecology Division, Hat Yai Hospital, Hat Yai, Songkhla 90110, Thailand
3 Obstetrics & Gynecology Division, Satun Hospital, Satun 91000, Thailand
Address reprint requests to: Tippawan Liabsuetrakul, Epidemiology Unit, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla 90110, Thailand. Tel: +66 74-429754; Fax: +66 74-212900/212903; E-mail: ltippawa{at}yahoo.com
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Objective. To assess the accessibility, utilization and quality of emergency obstetric care in the five southernmost provinces of Thailand.
Methods. A descriptive study was conducted in the five southernmost provinces of Thailand including fifty-six government hospitals and the admitted obstetric women. The accessibility of hospitals that provided emergency obstetric care was assessed over a 3-month period. The utilization and quality of emergency obstetric care were reviewed using the data of obstetric women admitted in a 6-month period. The admitted women with major obstetric complications were identified by the hospital reports. The accuracy of the hospital reports was checked using Lot Quality Assurance Sampling. The accessibility, utilization and quality of emergency obstetric care services, as defined by the United Nations (UN) indicators.
Results. There were 8.4 basic and 1.8 comprehensive emergency obstetric care hospitals per 500 000 population. The proportion of births in hospitals was 89.5%, and the adjusted percentage of women with major obstetric conditions treated in the hospitals was 95.5%. The caesarean section rate was 17.8%, and the case fatality rate caused by postpartum haemorrhage was 0.7%. Delays in seeking, reaching and receiving good care were detected in the causes of maternal deaths. Over-reporting due to double-counting and under-reporting of complications were found in the hospital reports.
Conclusions. Emergency obstetric care in Southern Thailand met UN guidelines; however, there was a problem of delays in seeking treatment in some maternal deaths. Improvement of over- and under-reporting of obstetric conditions in the hospitals is needed.
Keywords: emergency obstetric care, maternal morbidity, maternal mortality, southern Thailand, UN process indicators
Maternal mortality ratio is an important indicator reflecting the quality of maternal health care. In Thailand, there is as yet no good system of vital registration of deaths by cause as in developed countries. Maternal deaths and causes of deaths are reported hierarchically from the county to district then provincial levels, and then registered with the Regional Health Promotion Centre and the Department of Health of the Ministry of Public Health. The maternal health data of Thailand's five southernmost provinces were accumulated and monitored by Regional Health Promotion Centre 12 Yala [1]. In these five provinces, the maternal mortality ratio is higher than the national ratio. In 2002, the maternal mortality ratio in the five southernmost provinces was 54.8 per 100 000 live births and two-thirds of the maternal deaths occurred in hospitals, compared with 23.8 per 100 000 live births nationally.
A project to survey emergency obstetric care in developed and developing countries was launched in 1997 using the United Nations (UN) process indicators as the guidelines. These UN indicators are the number of hospitals that can provide care for emergency obstetric conditions per 500 000 population, the proportion of births in the hospitals, the proportion of women with obstetric complications treated in the hospitals, the caesarean section rate and the case fatality rate [2]. The countries in southeast Asia chosen for participation in this project were Sri Lanka, Nepal, Bhutan, Bangladesh and India [3–7]. Thailand was not studied, and thus data concerning the indicated process indicators were not collected.
Assessment of emergency obstetric care services is essential for all professionals and policy makers involved in maternal health services to allow them to adequately assess the health service situation, not only in Thailand but also other developing countries. The high maternal mortality in the five southernmost provinces of Thailand makes this a prime candidate for evaluating the readiness of government hospitals for emergency obstetric care. This study was thus undertaken to assess the accessibility, utilization and quality of emergency obstetric care using the process indicators in the five southernmost provinces of Thailand.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Design
A cross-sectional study was conducted in the five southernmost provinces of Thailand, Songkhla, Satun, Pattani, Yala and Narathiwat. The study was approved by the Institute Ethics Committee of the Faculty of Medicine, Prince of Songkla University.
Population and sample
The five southernmost provinces of Thailand comprised of 54 districts with a total population of 
3.2 million [8], with 60 government hospitals and 12 private hospitals. The lowest, or basic, level of health facility that offers delivery care is the district hospital. The studied hospitals were 56 government hospitals (93% of such hospitals in the five southernmost provinces), which included 55 district, provincial and regional hospitals authorized by the Ministry of Public Health and a university hospital that offers obstetric services to these provinces. The remaining government hospitals in the study area (7%), which were not studied, were Ministry of Defence hospitals that rarely provide inpatient obstetric services. The studied hospitals cover both urban and rural settings and were either first-level or referral services for obstetric care. The population data were taken from the 2000 population and housing census (Provincial Statistic Office). The available procedures for emergency obstetric care in each hospital were determined during the 3-month period of December 2004 to February 2005. The accessibility was compared on a provincial basis.
To estimate the required study sample size for assessing the utilization and quality of emergency obstetric care within the study provinces for women admitted with major obstetric complications, the sample was calculated based on a proportion of estimated obstetric complications of 15% with a 95% confidence interval, an acceptable error of 2% and variation of births of 30%, resulting in at least 1749 admitted obstetric women from each province being needed. To ensure the same time frame of data collection among provinces, all women admitted due to obstetric conditions between October 2004 and March 2005 were reviewed for assessing utilization and quality of emergency obstetric care.
Variables
The main outcome measures in this study were the UN process indicators of emergency obstetric care for monitoring the availability and use of obstetric services. These indicators are the accessibility of emergency obstetric care, the proportion of births in emergency obstetric care facilities, the met need, the caesarean section rate and the fatality rate. The accessibility of emergency obstetric services is defined as the number of hospitals providing basic or comprehensive obstetric care per 500 000 population in a specific administrative area. Basic emergency obstetric care services consist of the supplying of available parenteral antibiotics, parenteral oxytocics, anticonvulsants, manual removal of placenta, removal of retained products and operative vaginal deliveries. Comprehensive emergency obstetric care consists of the six basic types of care plus blood transfusion and delivery by caesarean section. The minimum acceptable number of available hospitals is four basic hospitals and one comprehensive hospital per 500 000 population.
The proportion of births in hospitals is defined as the number of births in studied hospitals divided by the number of births in a specific administrative area in the same time period. The number of all live births in this study was estimated by two methods. First, we counted the number of births in the studied hospitals plus the number of home-births in the area. Normally, the home-births would be reported and local health centre personnel within the county would visit the mother at home. Second, the number of births in the five provinces registered with the Bureau of Registration Administration of Thailand was checked because only 56 government hospitals were studied, not all hospitals in the five provinces.
The met need is defined as the number of women with major obstetric complications treated divided by the number of women expected to develop complications, normally estimated as 15% of all live births [2]. The number of all live births for calculating the met need was considered as the total number of births in all hospitals and homes in the area, calculated as noted previously. The major obstetric complications were defined as incomplete abortion and complications of abortion, ectopic pregnancy, pregnancy-induced hypertension (pre-eclampsia/eclampsia), antepartum/postpartum haemorrhage, obstructed labour/uterine rupture, amniotic embolism, retained placenta, amniotic fluid infection and postpartum genital tract infections. The caesarean section rate is the number of women who undergo caesarean sections as a percentage of the total number of live births. The case fatality rate is defined as the number of deaths of women due to obstetric complications divided by the number of women with obstetric complications. The cause-specific case fatality rate is identified as the number of deaths to women due to specific complications over the time period per number of women treated for those specific complications in the hospital.
Data collection
Before the study begun, a formal letter requesting permission for the responsible nurses to facilitate information gathering for the study was sent to the directors of the involved hospitals in the study area. All directors approved the study and the facilitating nurses were requested to attend a workshop, which aimed to explain the study and clarify the definition of variables and the method of data collection. After the workshop, the data recording form and the guideline for data collection were modified and then sent to all facilitating nurses. All responsible nurses recorded the data into the collection forms, and if any data were inconsistent or vague, they were contacted and we corrected the data by phone.
Since the data recorded in this study were dependent on the hospital obstetric reports, the quality assurance of these reports was checked to determine whether or not all complications of the pregnant women admitted were reported. In this study, there were 6, 33, 8 and 9 hospitals with levels of 10, 30, 60–90 and > 90 beds, respectively. The sampled hospitals were randomly selected by stratification of the number of hospital beds; therefore, one hospital was sampled from each of the groups of hospitals with 10-, 60–90- and > 90-bed levels, and five hospitals were sampled from the group of hospitals with 30 beds. Randomly sampled medical records were evaluated using a prepared checklist. The records were checked for any abnormal conditions during pregnancy, delivery and postpartum.
Analysis
Data were recorded in the EpiData 2.1 program and analysed with the Stata 7.0 program (TX, USA). The names of the provinces were made anonymous in terms of the research results by identifying them only by the codes A to E. Accessibility was determined by the qualified number of hospitals in the province compared with the minimum acceptable levels indicated by the UN process indicators. The actual number of hospitals, where emergency obstetric care was available in this survey, was described as the baseline availability. The indicators for utilization and quality of emergency obstetric care are presented using descriptive statistics.
Since over-reporting from double-counting was detected in some cases of women with major obstetric conditions who were admitted to more than one hospital or admitted more than once in the same hospital, the percentage of double-counting was analysed and adjusted for calculating the number of women with major obstetric complications treated in the hospitals. The quality assurance of hospital reports by Lot Quality Assurance Sampling (LQAS) [9] in eight sampled hospitals was described simply as pass or fail. A ‘failed’ result indicated that an unacceptable reporting error rate of > 5% was detected. If a woman was noted in the admission record as having abnormal conditions and/or being treated for complications, but there was no report of the problem in the diagnosis summary or hospital statistics report, it was defined as under-reporting. The under-reporting rate was calculated as a percentage based on the erroneous number of medical records and the total number of sampled medical records.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionAcknowledgementsReferences |
|---|
The 56 government hospitals included 49 district, 5 provincial, 1 regional and 1 university hospital. The number of hospital beds available in the district hospitals varied from 10 to 200 beds and from 186 to 515 beds in the provincial hospitals. There were
600 and 750 beds available in the regional and university hospitals, respectively. The number of doctors working in each hospital varied depending on the level of hospital and its location. Obstetricians were located in all provincial hospitals, regional hospital and university hospital, but found in only some of the district hospitals in Songkhla, Yala and Narathiwat provinces. In the district hospitals, women are taken care of by general practitioners and midwives throughout their pregnancy, labour, delivery and postpartum periods. Normally, there is at least one doctor working in all district hospitals, but there was one district hospital that did not have a doctor during the period of this survey, and a doctor from a nearby hospital was on call while the hospital searched for a permanent doctor. In Thailand, nurses and midwives are not allowed to perform obstetric procedures in basic emergency obstetric care. The number of doctors in the district hospitals ranged from one to seven with a median of 3 and in the provincial hospitals ranged from three to seven with a median of 5. Most hospitals had midwives and nurses working in the labour room and the postpartum ward. There is a well-managed referral system within the provinces; however, some hospitals located near a provincial boundary are able to refer women to hospitals outside the province. The transportation time within a province is normally an hour or less. Of the 56 hospitals, only one district hospital did not fulfil basic emergency obstetric care requirements due to a lack of services for the removal of placenta and retained products, because there was no available doctor working in the hospital at the time of the survey, as noted above. Twelve of the 56 hospitals (21.4%) could provide comprehensive emergency obstetric care. The baseline available and minimum acceptable number of hospitals, which provided basic or comprehensive emergency obstetric care in each province as found in the study, is shown in Table 1. In the five southernmost provinces, the number of hospitals with basic emergency obstetric care was 8.4 per 500 000 population and with comprehensive emergency obstetric care 1.8 per 500 000 population.
|
Table 2 shows the utilization and quality of emergency obstetric care services. A high percentage of births in the hospitals (89.5%) was demonstrated overall. There were 3093 pregnant women who did not use facility-based delivery care but were delivered at home by a traditional birth attendant in their district. No deaths were recorded in these women. Our study detected over-reporting from double-counting of cases admitted to more than one hospital or admitted more than once in the same hospital in 15.6% of the women with major obstetric complications. After adjustment for the double-counting, the overall rate of met need in the five southernmost provinces of Thailand was 95.5%. The average caesarean section rate in the five southernmost provinces was 17.8%.
|
During the study period, there were eight maternal deaths: five died from direct causes (three from postpartum haemorrhage and two from pregnancy-induced hypertension), two from indirect causes and one from a non-related car accident, giving a cause-specific case fatality rate from postpartum haemorrhage and pregnancy-induced hypertension of 0.7% (3/423) and 0.2% (2/791), respectively. The three maternal deaths from postpartum haemorrhage had associated conditions: one with placenta previa and percreta, one with placental abruption and one with uterine atony and severe pre-eclampsia, and all three cases resulted in disseminated intravascular coagulopathy. The two maternal deaths from pregnancy-induced hypertension were due to one case of severe pre-eclampsia with intracerebral haemorrhage at the left basal ganglion before hospital admission and one of severe pre-eclampsia with hepatic rupture at 31 weeks of pregnancy. Both indirect cause deaths resulted from heart disease: one after her caesarean section and the other at 19 weeks of gestation. One death occurred from a car accident during transfer to another hospital for further treatment of foetal distress. A review of these cases of maternal death found that there were problems of delay in deciding to seek, reaching and getting good care in the area of our study.
The most common conditions detected in admitted women during pregnancy were conditions related to abortions (6.6%), anaemia (2.4%), pre-eclampsia and eclampsia (2.1%), hyperemesis gravidarum (1.6%) and antepartum haemorrhage (0.7%). Conditions detected during labour were obstructed labour (6.5%), preterm birth (3.6%) and foetal distress (1.5%) and those during postpartum were postpartum haemorrhage (1.1%), retained placenta (0.7%) and genital tract infection (0.4%). The adjusted prevalence of major obstetric complications was 16.0%.
The quality assurance analysis of hospital reports in eight sampled hospitals were judged as failed. Overall under-reporting varied from 4.4 to 33.3%. The under-reporting of obstetric conditions was detected in 10.1% of the sampled hospitals. The rate of under-reporting from medical or surgical conditions depended on the cases of unreported anaemia, which if included in the statistics gave a total under-reporting rate of 20.8%, but if these cases were not counted, the level of under-reporting was only 13.7%. The details of under-reporting are shown in Table 3.
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionAcknowledgementsReferences |
|---|
The emergency obstetric care in the hospitals in five southernmost provinces of Thailand showed acceptable accessibility, utilization and quality by the criteria of the UN process indicators. Although the figures of these obstetric care services seemed well, it could not answer why the maternal mortality ratio in this area was still higher than other areas of country. The problems of unnecessary deaths related to delays in decision to seek, reaching or getting good care and the hospital reporting system were detected in this study.
Government hospitals in the five southernmost provinces of Thailand provide either basic or comprehensive emergency obstetric care services, at a level greater than the minimum recommendation as the guidelines [2, 10]. To compare our findings in Thailand with other countries, data summarizing previous reports on emergency obstetric care are presented in Table 4. The table shows that both basic and comprehensive emergency obstetric care in India, Nepal and Pakistan were found not to meet the minimum recommendations. In Bhutan, Mozambique, Morocco, Nicaragua, Sri Lanka, Cameroon, Senegal and Uganda, basic emergency obstetric care did not meet the minimum recommendations, whereas comprehensive emergency obstetric care surpassed the recommendations [3, 4, 6, 11, 12]. Most process indicators in our study were at a similar level to the USA [13].
|
Apart from the availability of hospitals providing emergency obstetric care, the promptness of service, available equipment, budget provided and the standards of the health system and care are important for reducing maternal deaths and improving maternal health [14]. A study provided evidence that good standard care decreased the number of maternal deaths [15]. The maternal deaths in the five southernmost provinces of Thailand were notably higher than the national average, indicating that the standard of care for emergency obstetric conditions needs to be improved.
In Thailand, the number of hospitals in any given area depends on the number of districts within the province and also the population, leading generally to no serious problem regarding inequity of distribution of facilities. The registration of vital statistics in Thailand is not organized in a way that allows good monitoring of maternal deaths and causes of death, although registration of births is much more complete because we have the office of Registration Administration to record the registered births in an area. The number of births from our survey was less than the number of registered births by a total of 726 births, which were delivered in other hospitals not included in our study sample. Also in our study, 10% of women delivered at home using a traditional birth attendant, a figure that was smaller than in the year 2000 (20%) [1].
This study found a rate of 95.5% for met need, which was high when compared with other developing countries, and our 16.0% of cases with major obstetric complications was close to the 15% estimation of the World Health Organization [2]. There was some problem in defining the major obstetric complications in postpartum sepsis and complications from abortion, as this was an observational study and the reports of diagnoses were affected by variation of practices among doctors in studied hospitals. There was no postpartum sepsis reported among all delivered women, which is an unlikely occurrence. In this study, we counted the women with postpartum genital infections or amniotic infections and those with incomplete abortion as major obstetric complications, which may have caused an overestimation of major obstetric complications.
The caesarean section and case fatality rates were acceptable. The overall caesarean section rate in the study area was higher than reports for other developing countries [3, 4, 6, 11], but lower than in the USA [13]. Although the UN process indicators were found to be met in the five southernmost provinces of Thailand, the quality assurance assessment of hospital reports revealed that the quality of hospital reports for obstetric conditions needs to be improved. Some of the under-reporting errors were serious, for example, the non-reporting of retained placenta and/or postpartum haemorrhage requiring blood transfusion or hypovolaemic shock and severe pre-eclampsia with or without magnesium sulphate treatment.
Although this study has some limitations, it is the first published report on the assessment of emergency obstetric care using process indicators in Thailand. Previous reports have not been attempted, as there are no data on basic and comprehensive hospitals for emergency obstetric care nationally in Thailand or in other sub-regions of the country. The problems discussed in this study should also be useful for other developing countries undertaking similar studies in the future.
In conclusion, the study found good accessibility, high utilization and acceptable quality of emergency obstetric care services in the five southernmost provinces of Thailand. A true assessment of emergency obstetric services was made more difficult by the over- and under-reporting of various conditions, and the hospital reporting system needs to be improved. The correct reporting of obstetric admissions would reveal the true magnitude of complications, leading to better and more appropriate planning. The delays in deciding to seek, reaching or receiving obstetric care, which were related to the causes of maternal deaths, need to be further analysed to find ways to reduce the maternal mortality in this area.
|
Acknowledgements |
|---|
|
TopAbstractMethodsResultsDiscussionAcknowledgementsReferences |
|---|
Funding support for this study was provided by the Institute of Research and Development for Health of Southern Thailand. We appreciate the cooperation and support of the various organizations, hospitals and health personnel, particularly the Faculty of Medicine, Prince of Songkla University, Health Centre 12 Yala, the 56 hospitals in Songkhla, Satun, Pattani, Yala and Narathiwat provinces, the Chief Medical Officers of Songkhla, Satun, Pattani, Yala and Narathiwat Provincial Health Offices and the 56 Hospital Directors, Provincial Maternal and Child Health Teams and the representatives of nurses in the labour rooms who helped to collect the data.
|
Footnotes |
|---|
* An earlier version of this study was presented orally at the XVIII FIGO World Congress of Gynecology and Obstetrics, Kuala Lumpur, Malaysia on 7 November 2006.
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionAcknowledgementsReferences |
|---|
- Maternal and Child Health working group. (2005) Yala: Regional Health Promotion Centre 12 Yala.
- UNICEF/WHO/UNFPA. Guidelines for Monitoring the Availability and Use of Obstetric Services (1997) New York: United Nations children's Fund.
- AMDD working group on indicators. Int J Gynecol Obstet (2003) 80:222–30.[CrossRef][Medline]
- AMDD working group on indicators. Int J Gynecol Obstet (2002) 77:277–84.[CrossRef][Medline]
- Pathak LR, Kwast BE, Malla DS, et al. Process indicators for safe motherhood programmes: their application and implications as derived from hospital data in Nepal. Trop Med Int Health (2000) 5:882–90.[CrossRef][Web of Science][Medline]
- Bailey PE, Paxton A. Program note: using UN process indicators to assess needs in emergency obstetric services. Int J Gynecol Obstet (2002) 76:299–305.[CrossRef][Medline]
- Biswas AB, Das DK, Misra R, et al. Availability and use of emergency obstetric care services in four districts of West Bengal, India. J Health Popul Nutr (2005) 23:266–74.[Web of Science][Medline]
- Economic and Social Statistics Bureau. (2000.) Bangkok: National Statistical Office.
- Lanata CF, Black RE. Lot quality assurance sampling techniques in health surveys in developing countries: advantages and current constraints. World Health Stat Q (1991) 44: 133–9.[Medline]
- Mekbib T, Kassaye E, Getachew A, et al. The FIGO Save the Mothers Initiative: the Ethiopia-Sweden collaboration. Int J Gynecol Obstet (2003) 81:93–102.[CrossRef][Medline]
- Orinda V, Kakande H, Kabarangira J, et al. A sector-wide approach to emergency obstetric care in Uganda. Int J Gynecol Obstet (2005) 91:285–91.[CrossRef][Medline]
- Ali M, Hotta M, Kuroiwa C, et al. Emergency obstetric care in Pakistan: potential for reduced maternal mortality through improved basic EmOC facilities, services, and access. Int J Gynecol Obstet (2005) 91:105–12.[CrossRef][Medline]
- Lobis S, Fry D, Paxton A. Program note: applying the UN process indicators for emergency obstetric care to the United States. Int J Gynecol Obstet (2005) 88:203–7.[CrossRef][Medline]
- Benagiano G, Thomas B. Averting maternal death and disability. Int J Gynecol Obstet (2003) 80:198–203.[CrossRef][Medline]
- Paxton A, Maine D, Freedman L, et al. The evidence for emergency obstetric care. Int J Gynecol Obstet (2005) 88:181–93.[CrossRef][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||