Abstract
Background
We examined long-term trends in age-specific and cause-specific maternal mortality and assessed shifts in the relative contributions of direct versus indirect obstetric causes from 1983 to 2023 in Korea.
Methods
In this population-based observational study, we analyzed all registered maternal deaths (International Classification of Diseases-10 O00–O99) among women aged 15–49 years and corresponding live births in Republic of Korea (= South Korea) from 1983 through 2023. Annual age-specific maternal mortality ratios (MMRs) per 100,000 live births were calculated using data from the Korean Statistical Information Service and the Causes of Death Statistics microdata of Statistics Korea. We classified deaths as direct or indirect obstetric causes and further disaggregated direct causes for the period 2009–2023. Proportions of direct versus indirect deaths were evaluated across five intervals (1997–2004, 2005–2009, 2010–2014, 2015–2019, and 2020–2023). Linear or segmented regression analyzed trends in overall MMRs and changes in the share of direct obstetric deaths.
Results
All age groups showed marked declines in MMRs; women aged ≥ 40 years exhibited the greatest absolute reduction (from 249 to 38 per 100,000). The overall MMR ranged from 10.3 to 13.7 per 100,000 since the early 2000s. The proportion of direct obstetric deaths fell by 2.1 percentage points per year before 2011, then rose by 0.9 points annually afterward. From 2009 to 2023, other direct causes accounted for 24.1% of deaths, embolism 21.5%, hemorrhage 21.3%, and indirect causes 19.9%; hypertensive disorders (5.9%), sepsis (4.2%), and abortion complications (3.1%) comprised smaller shares. Indirect causes predominated in maternal deaths among younger women, whereas direct obstetric causes were more common in those aged ≥ 35 years.
Conclusion
Although age-specific MMRs in Korea have steadily declined, the stagnation of the overall rate and continued burden of direct obstetric causes underscore vulnerabilities in the country’s emergency obstetric care infrastructure.
Keywords: Maternal Mortality Ratio, Hemorrhage, Embolism, Sepsis, Sustainable Development Goal, Korea
Graphical Abstract
INTRODUCTION
Reducing maternal mortality has been the primary objective of the maternal health community for decades.1 Global efforts to reduce maternal mortality have predominantly focused on channeling investments into interventions targeting the leading biomedical causes of death, especially during the perinatal period. Historically, major complications for most maternal deaths are hemorrhage, hypertensive disorders of pregnancy, and infections.2 With the demographic change and evolution of healthcare systems, many countries have observed a decline from high to low maternal mortality ratios (MMRs), and an increasing share of indirect causes.3
Republic of Korea (= South Korea) is recognized as one of the countries with lowest MMR. The composition of maternal deaths has begun to diverge from patterns seen in other high-income settings. When national cause-of-death statistics were first published in 1983, the MMR was 23.3 (95% confidence intervals [CIs], 20.0–26.9); it fell to 12.4 (10.0–15.3) in 1993. The MMR remained at similar levels: 11.9 (9.1–15.4) in 2003, 11.7 (8.70–15.4) in 2013, and 11.3 (7.4–16.6) in 2023, indicating no substantial improvement since the 1990s.4 This occurred against a backdrop of rising maternal age and region-wide consolidation of obstetric services implemented with limited efficiency accompanied by declining fertility, all of which may influence both the absolute risk of the healthcare needs for and complications of pregnant women. In this study, we quantify the age-specific maternal mortality incidence from 1983 to 2023, changing contributions of direct and indirect obstetric causes, and trends in cause-specific maternal mortality in Korea from 2009 to 2023. This work will enable tailored interventions that will further reduce preventable deaths and address emerging risks overlooked due to a low-mortality context.
METHODS
Data
We obtained annual counts of maternal deaths and live births in Korea for 1983–2023 from the Korean Statistical Information Service (KOSIS; https://kosis.kr) and the Microdata Integrated Service (MDIS; https://mdis.kostat.go.kr) of Statistics Korea. Cause-of-death codes were assigned according to the Korean Standard Classification of Diseases codes which is based on the International Classification of Diseases (ICD) codes.
From KOSIS, we extracted the total number of maternal deaths (diagnosis codes O00–O99 as the primary cause of death) and live births among women aged 15–49 for each calendar year. To examine detailed cause-of-death patterns and classify deaths into direct versus indirect obstetric causes, we used individual-level Causes of Death Statistics microdata of the same age group from MDIS for 1997–2023. Because the standard MDIS microdata categorizes maternal deaths into only direct and indirect groups, we additionally accessed the Complementary Investigations on Maternal Mortality microdata (MDIS) to obtain granular primary diagnosis codes for all maternal deaths which is available between 2009 and 2023.
Cause of death classification
A maternal death was defined as any death with an underlying ICD-10 code in the O00–O99 range, which were aggregated into a single maternal death category in the death statistics between 1983 and 2023. We identified direct or indirect obstetric causes for the individual-level death record for 1997–2023. We further disaggregated direct obstetric deaths into seven specific groups (abortion, embolism, hemorrhage, hypertensive disorders, pregnancy-related sepsis, and other direct causes) as described by Cresswell et al.3 The ICD codes assigned to each category in the MDIS maternal death database are listed in Supplementary Table 1.
Statistical analysis
Age-specific MMRs were calculated by applying national age-specific fertility rates to the number of maternal deaths in each age stratum (15–24, 25–29, 30–34, 35–39, ≥ 40 years) in order to examine temporal changes. To evaluate secular trends in cause-specific mortality, we divided the study period into five intervals (1997–2004, 2005–2009, 2010–2014, 2015–2019, 2020–2023). The linear trend in the proportions of direct obstetric deaths across these intervals was assessed using the Cochran–Armitage test. Poisson regression analysis was used to test for linear trends in overall number of livebirths, death, and MMR. We performed segmented regression using the segmented package in R to evaluate changes in the annual proportion of direct obstetric causes, identifying points where the trend shifted. For maternal deaths occurring between 2009 and 2023, we computed the proportions and corresponding 95% CIs for each specific cause. All statistical analyzes and figure creation were conducted in SAS version 9.4 (SAS Institute, Cary, NC, USA) and R version 4.3.2 (R Foundation for Statistical Computing, Vienna, Austria).
Ethics statement
This study was approved by the Institutional Review Board of Korea University and the requirement for informed consent was waived (KUIRB-2026-0306-01).
RESULTS
Between 1983 and 2023, all five maternal age groups experienced substantial declines in age-specific MMRs (Fig. 1). In 1983, women aged ≥ 40 years had the highest MMR (249 per 100,000), which declined to 38 per 100,000 by 2023—a larger absolute reduction than in any other age group. Similarly, the MMR for women aged 35–39 years declined from 153 to 19 per 100,000, and for those aged 30–34 years from 44 to 7 per 100,000. Even the youngest group (15–24 years), the lowest risk group, exhibited further reduction from 14 per 100,000 to zero by the most recent period.
Fig. 1. Decrease of annual age-specific MMR and TFR, 1983–2023, Korea.
MMR = maternal mortality ratio, TFR = total fertility rate.
A total of 1,406 maternal deaths occurred in 1997–2023 (Table 1). All of maternal deaths, the proportion of younger women (15–24 and 25–29 years) decreased markedly—from 11.4% to 1.8% and from 29.0% to 9.0%, respectively—while the proportion of maternal deaths among older women (35–39 and ≥ 40 years) increased from 19.1% to 41.4% and from 7.3% to 14.4%, respectively. Among all cases of maternal death, the share of women with only middle-school education decreased (21.7% to 1.8%), and deaths outside the Seoul capital area declined modestly (57.6% to 41.4%). Although the overall MMR showed no significant linear trend across periods (P = 0.683), the proportion of direct obstetric deaths generally decreased from 97.2% in 1997-2004 to 84.7% in 2020–2023. However, the proportion reached its nadir during 2010–2014, then increased in 2015–2019 and further in 2020–2023. Segmented regression identified a joinpoint at 2011. Before 2011, the annual proportion of direct obstetric causes declined by 2.1 percentage points per year. After 2011, the trend reversed, increasing by 0.9 percentage points per year (Supplementary Fig. 1).
Table 1. Individual characteristics of maternal death and proportion of direct obstetric death (%), 1997–2023.
| Indicators | Period | P for trenda | ||||||
|---|---|---|---|---|---|---|---|---|
| 1997–2004 | 2005–2009 | 2010–2014 | 2015–2019 | 2020–2023 | ||||
| Average annual No. of livebirths | 659,764 | 541,971 | 554,864 | 438,507 | 303,273 | < 0.001 | ||
| Average annual No. of death | 66.9 | 51.6 | 62.8 | 37.6 | 27.8 | < 0.001 | ||
| Maternal mortality ratiob | 11.6 (10.6–12.6) | 11.2 (9.9–12.6) | 13.7 (12.2–15.2) | 10.3 (8.8–11.7) | 11 (8.9–13.0) | 0.683 | ||
| Maternal death cases | ||||||||
| Age, yr | ||||||||
| 15–24 | 61 (11.4) | 19 (7.36) | 8 (2.6) | 5 (2.7) | 2 (1.8) | < 0.001 | ||
| 25–29 | 155 (29.0) | 64 (24.8) | 59 (18.8) | 24 (12.8) | 10 (9.0) | < 0.001 | ||
| 30–34 | 178 (33.3) | 95 (36.8) | 121 (38.5) | 82 (43.6) | 37 (33.3) | 0.090 | ||
| 35–39 | 102 (19.1) | 61 (23.6) | 98 (31.2) | 67 (35.6) | 46 (41.4) | < 0.001 | ||
| ≥ 40 | 39 (7.3) | 19 (7.4) | 28 (8.9) | 10 (5.3) | 16 (14.4) | 0.170 | ||
| Education | ||||||||
| ≤ Middle school | 116 (21.7) | 18 (7.0) | 17 (5.4) | 4 (2.1) | 2 (1.8) | < 0.001 | ||
| Living area | ||||||||
| Outside Seoul capital area | 308 (57.6) | 142 (55.0) | 161 (51.3) | 110 (58.5) | 46 (41.4) | 0.025 | ||
| Direct obstetric deathc | 520 (97.2) | 227 (88.0) | 222 (70.7) | 156 (83.0) | 94 (84.7) | < 0.001 | ||
Values are presented as number (%) or maternal mortality ratio (95% confidence interval).
aCalculated by Cochran–Armitage test across the five periods for categorical variables. bPoisson regression was used to assess a linear trend in maternal mortality ratios as recorded in the Microdata Integrated Service database. cDirect obstetric death as recorded in the Microdata Integrated Service database.
Among 783 maternal deaths between 2009 and 2023, where specified causes of death available, the leading causes were other direct obstetric causes (24.1%; 95% CI, 21.1–27.1), embolism (21.5%; 95% CI, 18.6–24.3), hemorrhage (21.3%; 95% CI, 18.5–24.2), and indirect causes (19.9%; 95% CI, 17.1–22.7; Supplementary Fig. 2). Traditionally predominant direct obstetric causes, such as hypertensive disorders (5.9%; 95% CI, 4.2–7.5), sepsis (4.2%; 95% CI, 2.8–5.6), and abortion complications (3.1%; 95% CI, 1.9–4.3), were less common, with overlapping CIs. When stratified by age group, indirect causes were predominant among young women (Fig. 2). For women aged ≥ 35, direct obstetric causes including hemorrhage and embolism were more common.
Fig. 2. Causes in all maternal deaths in Korea from 2009 to 2023 (n = 783), stratified by age group. The classification follows the approach of Cresswell et al.3 The error bars indicate 95% CIs of each proportion. Direct causes are shown in light blue, while the indirect cause is shown in orange.
CI = confidence interval.
DISCUSSION
Between 1983 and 2023, mothers of oldest age group experienced the steepest decline in MMR—falling to approximately one-sixth of the 1983 rate. Although Korea’s overall MMR now ranks among the lowest worldwide, it has ranged from 10.3 to 13.7 since the early 2000s. In contrast to many other countries where deaths from direct obstetric causes have been replaced by indirect causes, Korea has shown a relative rise in direct-cause deaths, particularly postpartum hemorrhage. Direct obstetric causes of death predominated specifically among women aged 35 years and older. These findings underscore the importance of sustaining robust, well-operationalized emergency obstetric care systems, especially in contexts of persistently low fertility.
The relatively high contribution of hemorrhage and embolism in Korean maternal deaths is consistent with the findings of other countries. In Japan, obstetric hemorrhage, brain disease, and amniotic fluid embolism consisted the largest in all maternal deaths.5 However, this pattern partially contradicts findings from or patterns in European countries. A study of maternal deaths in eight European countries similarly reported that direct obstetric causes, including hemorrhage, hypertensive disorders, and sepsis, have become less prominent over the last two decades, whereas cardiovascular disease and mental health conditions account for a substantial and rising fraction of deaths.6
The increased proportion of deaths from indirect causes may reflect both improved management of direct pregnancy and childbirth complications and shifts in underlying risk factors and disease profiles.7,8 In the United States (US), cardiovascular conditions were the leading cause of maternal death between 2013 and 2017, reflecting advanced maternal age, higher multifetal rates, and improved detection.9 Despite dramatic reductions in direct deaths following quality-improvement initiatives, demographic changes such as aging and the epidemiologic transition toward noncommunicable diseases have sustained both direct and indirect maternal mortality in Korea, contributing to the observed plateau in overall MMR. Among countries experiencing a substantial rise in maternal age, Korea recorded one of the highest averages, with the mean age at first birth reaching 32 years in 2021, representing the largest increase observed during that period.10 Globally, the prevalence of hypertension has also risen among individuals aged 20–39 years, likely contributing to a greater risk of obstetric morbidities in pregnant women within this age group.11,12
The persistently high proportion of indirect maternal deaths highlights a critical challenge for extremely low-fertility societies. As birth rates decline, specialized obstetric services consolidate: small and rural maternity units close under financial and workforce pressures, eroding local emergency capacity. In the US and Australia, low reimbursement rates and staffing shortages have precipitated obstetric unit closures, delaying access to lifesaving care in rural areas.13 Although Korea’s universal health coverage ensures broad access, over one-quarter of its administrative districts lack any obstetric facility.14 To avert preventable deaths, strategies proven effective in other near-zero-MMR settings that leverage centralized referral centers with regional emergency teams, quality improvement of obstetric services, and incentives to sustain low-volume units, should be explored.15,16 The implementation of high-risk pregnancy management system that ensures closer antenatal monitoring and early referral protocols especially for pregnant women with advanced age would be also required.17 Several financial and psychological support for those with high-risk factors have been effective in reducing the incidence of obstetric morbidities.18,19
Our findings need caution in interpretation. First, our use of vital-registration data may understate the true burden of maternal death. In Korea, only deaths occurring within 42 days of pregnancy termination are classified as “maternal”; deaths from obstetric causes between 43 and 365 days postpartum are excluded and thus not captured. This could lead to underreporting, particularly of indirect or late-presenting conditions, and bias in cause-specific trends. Second, detailed ICD coding for specific direct obstetric causes was unavailable in the MDIS microdata before 2009. As a result, we could not examine finer etiologic shifts in the conditions under the direct cause category in the earlier decades. Finally, several age and cause-of-death strata contained small counts, producing wide CIs and limiting power to detect subtle temporal changes, especially for infrequent causes.
Although Korea maintains a low overall maternal mortality during the past decades, the persistently high proportion of deaths from direct obstetric causes signals potential vulnerabilities in the healthcare system. Prioritizing and sustaining high-quality emergency obstetric services will be essential to prevent all avoidable maternal deaths.
Footnotes
Funding: This research was supported by the National Research Foundation of Korea grant (2018R1D1A1B07048821 & 2022R1A2C1006364).
Disclosure: The authors have no potential conflicts of interest to disclose.
Data Availability Statement: The data used in this study is publicly available at the websites of the Korean Statistical Information Service (KOSIS; https://kosis.kr) and the Microdata Integrated Service (MDIS; https://mdis.kostat.go.kr) of Statistics Korea.
- Conceptualization & Writing - original draft: Han BR.
- Formal analysis & Supervision: Choe SA.
- Project administration & Provision of technical consultation: Yoon SJ, Min HS.
- Methodology & Writing - review & editing: Oh JW, Seo J, Ahn KH.
SUPPLEMENTARY MATERIALS
Specific diagnostic codes recorded in the MDIS maternal death database assigned to each category described by Cresswell and colleagues
Segmented regression analysis of annual percentage of direct obstetric causes in all maternal deaths from 1997 to 2023, with one joinpoint.
Causes in all maternal deaths in Korea from 2009 to 2023 (n = 783). The classification follows the approach of Cresswell et al.3 The error bars indicate 95% confidence intervals of each proportion. Direct causes are shown in light blue, while the indirect cause is shown in orange.
References
- 1.Vogel JP, Jung J, Lavin T, Simpson G, Kluwgant D, Abalos E, et al. Neglected medium-term and long-term consequences of labour and childbirth: a systematic analysis of the burden, recommended practices, and a way forward. Lancet Glob Health. 2024;12(2):e317–e330. doi: 10.1016/S2214-109X(23)00454-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.World Health Organization. Maternal mortality. [Updated 2025]. [Accessed December 10, 2025]. https://www.who.int/news-room/fact-sheets/detail/maternal-mortality .
- 3.Cresswell JA, Alexander M, Chong MYC, Link HM, Pejchinovska M, Gazeley U, et al. Global and regional causes of maternal deaths 2009-20: a WHO systematic analysis. Lancet Glob Health. 2025;13(4):e626–e634. doi: 10.1016/S2214-109X(24)00560-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Korean Statistical Information Service. Number of Deaths and Mortality Rates by Cause of Death (237 Categories), Sex, and 5-Year Age Groups. Daejeon, Korea: Statistics Korea; 2025. [Google Scholar]
- 5.Hasegawa J, Sekizawa A, Tanaka H, Katsuragi S, Osato K, Murakoshi T, et al. Current status of pregnancy-related maternal mortality in Japan: a report from the Maternal Death Exploratory Committee in Japan. BMJ Open. 2016;6(3):e010304. doi: 10.1136/bmjopen-2015-010304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Diguisto C, Saucedo M, Kallianidis A, Bloemenkamp K, Bødker B, Buoncristiano M, et al. Maternal mortality in eight European countries with enhanced surveillance systems: descriptive population based study. BMJ. 2022;379:e070621. doi: 10.1136/bmj-2022-070621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kassebaum NJ, Bertozzi-Villa A, Coggeshall MS, Shackelford KA, Steiner C, Heuton KR, et al. Global, regional, and national levels and causes of maternal mortality during 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014;384(9947):980–1004. doi: 10.1016/S0140-6736(14)60696-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Storm F, Agampodi S, Eddleston M, Sørensen JB, Konradsen F, Rheinländer T. Indirect causes of maternal death. Lancet Glob Health. 2014;2(10):e566. doi: 10.1016/S2214-109X(14)70297-9. [DOI] [PubMed] [Google Scholar]
- 9.Collier AY, Molina RL. Maternal mortality in the United States: updates on trends, causes, and solutions. Neoreviews. 2019;20(10):e561–e574. doi: 10.1542/neo.20-10-e561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Organisation for Economic Co-operation and Development (OECD) SF2.3: Age of Mothers at Childbirth and Age-Specific Fertility. Paris, France: OECD and OECD KOREA Policy Centre; 2023. [Google Scholar]
- 11.Kearney PM, Whelton M, Reynolds K, Muntner P, Whelton PK, He J. Global burden of hypertension: analysis of worldwide data. Lancet. 2005;365(9455):217–223. doi: 10.1016/S0140-6736(05)17741-1. [DOI] [PubMed] [Google Scholar]
- 12.Choe SA, Min HS, Cho SI. The income-based disparities in preeclampsia and postpartum hemorrhage: a study of the Korean National Health Insurance cohort data from 2002 to 2013. Springerplus. 2016;5(1):895. doi: 10.1186/s40064-016-2620-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Smith JG, Brown KK. Rural hospital and obstetric unit closures as social determinants of racial and ethnic maternal health disparities: a scoping review. J Adv Nurs. 2024;80(8):3059–3071. doi: 10.1111/jan.16005. [DOI] [PubMed] [Google Scholar]
- 14.Min HS, Kim S, Kim S, Lee T, Kim SY, Ahn HS, et al. Is limited access to obstetric services associated with adverse birth outcomes? A cross-sectional study of Korean national birth data. BMJ Open. 2022;12(5):e056634. doi: 10.1136/bmjopen-2021-056634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Holmer H, Oyerinde K, Meara JG, Gillies R, Liljestrand J, Hagander L. The global met need for emergency obstetric care: a systematic review. BJOG. 2015;122(2):183–189. doi: 10.1111/1471-0528.13230. [DOI] [PubMed] [Google Scholar]
- 16.Fladeby LKL, Raunedokken M, Fonkalsrud HE, Hvidtjørn D, Lukasse M. Midwives’ experiences of using the Obstetric Norwegian Early Warning System (ONEWS): a national cross-sectional study. Eur J Midwifery. 2021;5:10. doi: 10.18332/ejm/134510. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kim DS, Kim YT, Kim TH. Maternal Age and Adverse Pregnancy Outcomes in Korea: A Comprehensive Approach to Prenatal Care. [Updated 2011]. [Accessed November 15, 2025]. [DOI]
- 18.Choe SA, Min HS, Cho SI. Decreased risk of preeclampsia after the introduction of universal voucher scheme for antenatal care and birth services in the Republic of Korea. Matern Child Health J. 2017;21(1):222–227. doi: 10.1007/s10995-016-2112-z. [DOI] [PubMed] [Google Scholar]
- 19.Milgrom J, Schembri C, Ericksen J, Ross J, Gemmill AW. Towards parenthood: an antenatal intervention to reduce depression, anxiety and parenting difficulties. J Affect Disord. 2011;130(3):385–394. doi: 10.1016/j.jad.2010.10.045. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Specific diagnostic codes recorded in the MDIS maternal death database assigned to each category described by Cresswell and colleagues
Segmented regression analysis of annual percentage of direct obstetric causes in all maternal deaths from 1997 to 2023, with one joinpoint.
Causes in all maternal deaths in Korea from 2009 to 2023 (n = 783). The classification follows the approach of Cresswell et al.3 The error bars indicate 95% confidence intervals of each proportion. Direct causes are shown in light blue, while the indirect cause is shown in orange.