임신출산진료비 바우처 프로그램이 주요 모성 합병증의 발생률과 불평등에 미친 영향: 국민건강보험공단 코호트 자료 연구 = Impact of Voucher Program on the Incidence and Income-based Disparity of Major Obstetric Complications: from the Korean National Health Insurance Cohort Data

다국어 초록 (Multilingual Abstract)

Objective:
Maternal morbidity during pregnancy and childbirth has been a leading contributor to the burden of disease among women aged 15-44 years worldwide. There have been growing evidences which indicate socioeconomic disparities in the obstetric complications. To reduce the occurrence and disparities in the maternal morbidities, a number of demand-side financing interventions were introduced and implemented worldwide. In Korea, a universal voucher program (Goeun-mom card) for antenatal care and birth services was initiated in December 2008 to improve total fertility rate. Since there is no study which addresses the socioeconomic disparities in health of mothers and the effect of the universal voucher scheme on maternal health, the purposes of this study are followings:

First, to identify associated factors and trends of incidence of two major obstetric complications - preeclampsia and postpartum hemorrhage.
Second, to explore the incidence and disparities in the major maternal complications according to household income level.
Third, to evaluate the effect of the universal voucher program on the risk of preeclampsia and postpartum hemorrhage.

Methods:
(1) A population-based cohort data from the Korean National Health Insurance (NHI) Service from 2002 to 2013 was used for analyses. Individuals with live birth in the year were identified by combined diagnostic criteria. The occurrences of preeclampsia or postpartum hemorrhage were determined by the diagnostic code and specific treatment for these conditions. Social determinants such as household income level, being in paid work were included in the multivariable model.
(2) Using regression analysis, annual age-adjusted incidences of preeclampsia and postpartum hemorrhage were calculated, respectively. The temporal change of annual incidences was analyzed with annual percentage change (APC) and average annual percentage change (AAPC) using the Joinpoint Regression Program (Version 4.1.1.4 - February 2015; Statistical Methodology and Applications Branch, Surveillance Research Program, National Cancer Institute).
(3) The study population was divided into three groups (lower, middle and higher) according to the level of relative household income. Based on the household income gradient, inequality indices such as slope index of inequality (SII) and Kunst and Mackenbach relative index of inequality (RIIKM) of predicted incidences were calculated using Health Disparities Calculator (HD*Calc, Version 1.2.4 - October 29, 2013; Division of Cancer Control and Population Sciences, Surveillance Research Program and Applied Research Program, National Cancer Institute).
(4) For the maternal conditions which show significant income-based disparities during the study period, changes after the introduction of voucher program (from 2009 to 2013) were evaluated. To estimate the impact of universal voucher program on the occurrence of major obstetric complications, generalized linear model (GLM) was adopted while controlling for the effect of potential confounding factors including calendar year.

Results:
(1) In the aggregated data of live births in the NHI cohort, the risk of postpartum hemorrhage requiring blood transfusion was higher in lower household income groups than higher income group (P for trend = 0.046). The frequency of preeclampsia was not different among the three groups.
(2) With multivariable analyses, women in lower household income showed higher risk of preeclampsia and postpartum hemorrhage, respectively. For postpartum hemorrhage, being in paid work increased the odds of postpartum hemorrhage. Maternal age between 35 and 39, presence of diabetes, and multiple gestation were observed to increase the risk of preeclampsia. Risk of postpartum hemorrhage was higher in women aged between 15-19 and over 30 years and nulliparity. Occurrence of postpartum hemorrhage was positively associated with induction of labor and Cesarean delivery.
(3) During the study period, there was no significant change in the annual age-adjusted incidence of preeclampsia requiring MgSO4 treatment in all three income groups. On the other hand, annual age-adjusted incidences of postpartum hemorrhage decreased over time (AAPC = -3.58, P = 0.01 for the lower income group; AAPC = -3.86, P = 0.02 for the higher income group).
(4) The estimates of SII’s and RIIKM’s indicated no significant disparities in preeclampsia and/or postpartum hemorrhage among the study population considering each income group share. The RIIKM of the age-adjusted incidences of preeclampsia across the household income gradient showed a generally decreasing trend over time when the study period were divided into three parts (2002-2005, 2006-2009 and 2010-2013). However, the change was not statistically significant. For RIIKM of postpartum hemorrhage, the temporal change was not unidirectional.
(5) When the effects of multiple covariates are controlled, the risks of preeclampsia and postpartum hemorrhage among mothers were reduced after implementation of the voucher program. However, with controlling for effects of calendar year, the risk of preeclampsia was reduced only in 2013. For postpartum hemorrhage, the risk was reduced since 2011. In this multivariable models, the higher risks of preeclampsia and postpartum hemorrhage remained significant in the lower and middle income than in the higher income group.

Conclusions:
(1) The temporal trend of the major maternal complications could be different according to the level of household income. In Korea, the incidence of postpartum hemorrhage decreased in the lower and higher income groups during the past decade.
(2) The relative inequality indices for the three maternal morbidities showed no unidirectional change during the study period. Considering the differential trend of postpartum hemorrhage according to household income, attempts to monitor and reduce the disparities in maternal health across different economic status would be necessary.
(3) Women who delivered in the post-voucher period showed lower risk of preeclampsia and postpartum hemorrhage, respectively. This may suggest that a voucher program for antenatal care and birth services covering all mothers would be associated with the lower risk of maternal complications even in the setting of national health insurance system.

목차 (Table of Contents)

• CHAPTER 1. INTRODUCTION 1
• 1-1. Public Health Implication of Maternal Health 1
• 1-2. Income-based Differentials in Maternal Conditions during Pregnancy and Delivery 3
• 1-3. Voucher Program and Inequalities in Health 4
• 1-4. Introduction of the Universal Voucher Program for Antenatal Care and Birth Services in the Republic of Korea (Goeun-mom card) 6
• 1-5. Study Objectives 7
• CHAPTER 2. STUDY MATERIALS 9
• 2-1. Data sources 9
• 2-2. Study setting 10
• 2-2-1. Identification of live birth cases in the NHI database 10
• 2-2-2. Outcome and explanatory variables 11
• CHAPTER 3. SOCIAL DETERMINANTS MAJOR OBSTETRIC COMPLICATIONS IN KOREA 14
• 3-1. Background 14
• 3-2. Methods 15
• 3-3. Result 18
• 3-4. Discussion 22
• CHAPTER 4. TREND OF ANNUAL INCIDENCE OF THE MAJOR OBSTETRIC COMPLICATIONS FROM 2002 TO 2013 IN KOREA 24
• 4-1. Background 24
• 4-2. Methods 26
• 4-3. Results 28
• 4-4. Discussion 33
• CHAPTER 5. INCOME-BASED DISPARITIES IN THE MAJOR MATERNAL COMPLICATIONS IN KOREA 38
• 5-1. Background 38
• 5-2. Methods 39
• 5-3. Results 40
• 5-4. Discussion 44
• CHAPTER 6. EFFECT OF THE UNIVERSAL VOUCHER PROGRAM (GOEUN-MOM CARD) ON THE INCIDENCE OF PREECLAMPSIA AND POSTPARTUM HEMORRHAGE 47
• 6-1. Background 47
• 6-2. Methods 49
• 6-3. Results 52
• 6-4. Discussion 60
• CHAPTER 7. OVERALL DISCUSSION AND CONCLUSION 64
• 7-1. Different temporal trend of incidences of maternal complication according to household income 64
• 7-2. Income-based disparities in the incidence of maternal complication 65
• 7-3. Different temporal trend of incidences and disparities of preeclampsia 66
• 7-4. Implications for maternal health-related policy and future researches 67
• REFERENCES 70
• 국문 초록 82
• Appendix 86
• Appendix 1. Procedure codes for delivery mode, determination of nulliparity and multiple gestation 86
• Appendix 2. Dianosis-related Group (DRG) code for cesarean delivery cases (in Korean) 88
• Appendix 3. Treatment codes for MgSO4 to determine the presence of preeclampsia (in Korean) 89