GRAPHICAL ABSTRACT

Key Message

Despite declining global sudden infant death syndrome (SIDS) incidence, recent nationwide trends and spatiotemporal patterns in Korea remain poorly characterized. This study showed significantly higher SIDS incidence during pandemic (2020–2021) and post-pandemic (2022–2023) periods than during pre-pandemic period (2013–2019). Bayesian spatial modeling revealed progressively stronger geographic clustering, with low-risk clusters in the capital region (Seoul and Gyeonggi-do) and high-risk clusters in southwestern regions (Jeollanam-do). These findings support geographically targeted interventions, including preventive education, breastfeeding promotion, and strengthened regional surveillance, to reduce the growing epidemiological disparities in SIDS risk among Korean infants.

INTRODUCTION

Sudden infant death syndrome (SIDS) is defined as the sudden, unexpected death of an infant younger than 1 year that remains unexplained after a thorough investigation [1]. Public health efforts to reduce SIDS have largely focused on modifiable risk factors, most notably through the “Safe to Sleep” campaign [2]. Although global prevention campaigns have reduced SIDS mortality by 51% since the 1990s, the condition still accounted for 20.9 deaths per 100,000 live births worldwide in 2019 [3,4]. Marked geographic variation persists, with incidence ranging from 5.9 per 100,000 live births in East Asia to 41.8 per 100,000 live births in Western sub-Saharan Africa; consistently lower rates have been reported in Asian populations [3,5,6].

Despite substantial global advances in SIDS research and prevention, SIDS incidence patterns in Korea remain difficult to characterize. The most recent comprehensive nationwide study was conducted in 1997–1998, and subsequent surveillance has been insufficient to monitor contemporary trends in SIDS incidence and geographic distribution [7]. Since then, major changes in healthcare systems and socio-demographic conditions may have influenced SIDS incidence patterns.

Spatial epidemiological approaches, particularly Bayesian spatial modeling, are well suited to identifying geographic clustering of rare events such as SIDS and can help guide targeted public health interventions [8,9]. To date, no study has examined the spatiotemporal epidemiology of SIDS in Korea. Accordingly, this study investigated spatiotemporal variation in SIDS incidence across Korea using a nationwide cardiac arrest registry from 2013 to 2023, with the objectives of characterizing epidemiological trends and identifying priority areas for targeted prevention efforts.

MATERIALS AND METHODS

Study setting

According to the 2023 administrative framework, Korea comprises 6 metropolitan cities (Gwangyeoksi), 1 special city (Teukbyeolsi), 1 special self-governing city (Teukbyeol Jachisi), and 9 provinces (do), including 1 special self-governing province (Teukbyeol Jachido). These upper-level jurisdictions are subdivided into 250 regional administrative districts, which are classified as cities (si), counties (gun), or districts (gu). All analyses were conducted at the administrative-district level.

Data collection

This nationwide retrospective study analyzed data from the Out-of-Hospital Cardiac Arrest Surveillance (OHCAS) database maintained by the Korea Disease Control and Prevention Agency (KDCA). Established in 2006, OHCAS is a national registry and the principal source of standardized data on out-of-hospital cardiac arrest (OHCA) in Korea. The registry incorporates data from the National Fire Agency database and hospital medical records, with rigorous verification performed by KDCA. The primary dataset is derived from emergency medical services (EMS) documentation and includes information on cardiac arrest circumstances, resuscitation timing, and prehospital interventions. EMS providers operate within a government-based, partially dual-dispatch system that uses basic life support fire engines and advanced cardiovascular life support ambulances for suspected OHCA cases. The secondary data component consists of medical record review by trained experts using standardized protocols aligned with internationally recognized OHCA investigation standards, specifically the Utstein-style guidelines and the Resuscitation Outcomes Consortium Project framework [10,11]. The extracted variables included patient demographics, incident location, witness status, bystander cardiopulmonary resuscitation, cardiac arrest etiology, initial electrocardiographic findings, hospital-based management, and outcomes.

Geographic data, including district codes, district names, administrative unit identifiers, polygon boundary coordinates, and area in square kilometers, were obtained from the Statistical Geographic Information Service provided by Statistics Korea (accessible at https://sgis.kostat.go.kr/view/index). Some districts underwent boundary changes during the 11-year observation period. For integrated districts, SIDS case counts and live birth counts were combined across predecessor districts. For subdivided districts, values were equally distributed among the newly created districts based on the number of resulting subdivisions.

Study population

All OHCA cases recorded in the OHCAS database from January 2013 to December 2023, were screened (n=338,169) (Figure 1). Patients older than 1 year were excluded (n=335,825). Cases with a documented primary cause of cardiac arrest other than SIDS (n=1,435) and cases with missing residential area information (n=25) were also excluded. The final study cohort comprised 884 cases.

Variables and measurements

The analyzed variables included patient demographics—age, sex, and residential location, categorized as metropolitan or non-metropolitan/rural—as well as cardiac arrest circumstances and emergency response parameters. Cardiac arrest characteristics included arrest location (public or non-public), witness status (witnessed or unwitnessed), bystander response (performed or not performed), and first monitored rhythm (shockable or non-shockable). Arrest-to-emergency department (ED) time was defined as the interval from cardiac arrest to ED presentation.

Cardiac arrest etiologies in the OHCAS database are classified using predefined categories, including SIDS. To ensure consistency, only cases coded as SIDS were included in the analysis. SIDS incidence was calculated as the number of SIDS cases per 100,000 live births. To evaluate temporal trends, the 11-year study period was divided into 3 phases: 2013–2019 (pre-pandemic), 2020–2021 (pandemic), and 2022–2023 (post-pandemic). This division was based on the emergence of the coronavirus disease 2019 (COVID-19) pandemic in early 2020 and the reclassification of COVID-19 from a Category 1 to a Category 2 infectious disease in Korea in 2022.

Study outcomes

The study outcomes were temporal and spatial variation in SIDS incidence across all administrative districts in Korea and the identification of high-risk regions.

Statistical analysis

Descriptive statistics are presented as medians with interquartile ranges for continuous variables and as frequencies with percentages for categorical variables. Continuous variables were compared using the Kruskal–Wallis test, whereas categorical variables were compared using the chi-square test or Fisher’s exact test (with Monte Carlo simulation when appropriate).

Temporal changes in SIDS incidence were assessed using Poisson regression models, with SIDS cases as the dependent variable and the logarithm of live births as an offset. First, a 3-period analysis was conducted with study period as the independent variable, and incidence rate ratios (IRRs) and 95% confidence intervals (CIs) were calculated for each period relative to the pre-pandemic reference period. Second, annual IRRs were calculated using 2013 as the reference year to characterize year-specific changes in incidence throughout the observation period. Third, joinpoint regression was performed using a segmented Poisson regression model with year as a continuous predictor to characterize the temporal trend and identify statistically significant change points.

Standardized incidence ratios (SIRs) were calculated for each district as the ratio of observed to expected cases. Expected cases were derived by multiplying the national SIDS rate by the district-specific number of live births. To account for spatial autocorrelation, a Bayesian hierarchical spatial model with a Besag-York-Mollié 2 (BYM2) prior was applied and implemented using the integrated nested Laplace approximation approach. The model assumed a Poisson likelihood with an offset for the logarithm of the expected count. Spatial adjacency was defined using Queen contiguity from a national district shapefile. Posterior means and 95% credible intervals (CrIs) for SIRs were obtained, with separate analyses conducted for each of the 3 periods. Districts without spatial neighbors, including islands, were modeled as independent units. To address multiple comparisons inherent in evaluating district-level SIRs across 250 administrative units, posterior exceedance probabilities (EPs) were calculated from the marginal posterior distributions of the fitted values. Districts were classified as having robustly elevated SIRs when posterior EPs exceeded 0.95; a stricter threshold of 0.99 was also examined.

Global spatial autocorrelation of posterior mean SIRs was assessed using Moran’s I with row-standardized contiguity weights, applied consistently across the 250 districts for all periods with 999 permutations. Temporal changes in clustering (ΔI) were assessed using a paired permutation test with 2,000 iterations by randomly swapping period assignments within each district with a probability of 0.5 [12]. Although grounded in established resampling principles for paired data, this specific application of a permutation procedure to evaluate temporal changes in Moran’s I derived from BYM2 estimates appears, to our knowledge, to be a novel analytical approach developed for this study. Applying this test to BYM2 estimates leveraged spatial shrinkage to partially reduce instability arising from unequal observation windows. One-sided p-values were calculated as the proportion of permuted ΔI values greater than or equal to the observed ΔI.

A 2-sided p-value<0.05 was considered statistically significant unless otherwise specified. All statistical analyses were performed using R version 4.5.0 (R Foundation for Statistical Computing, Vienna, Austria).

Ethics statement

The study protocol was reviewed and approved by the Institutional Review Board (IRB) of Catholic Kwandong University (approval No. IS25RISI0032). Because this study used publicly available, anonymized data, the IRB waived the requirement for informed consent. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology guidelines [13].

RESULTS

Baseline characteristics and clinical information of participants

Table 1 summarizes the baseline characteristics of 884 SIDS cases stratified by study period: pre-pandemic (n=575), pandemic (n=172), and post-pandemic (n=137). The proportion of cases occurring in non-metropolitan or rural areas increased significantly across periods, from 56.9% to 69.2% (p=0.009), as did bystander cardiopulmonary resuscitation performance, from 23.8% to 41.9% (p<0.001), and median arrest-to-ED time, from 38 minutes to 65 minutes (p<0.001). Witness status did not differ significantly across periods (p=0.166).

Temporal trends in sudden infant death syndrome incidence

Overall SIDS incidence fluctuated substantially during the study period, ranging from a nadir of 15.0 per 100,000 live births in 2016 to a peak of 33.4 per 100,000 live births in 2023 (Figure 2). Joinpoint regression identified significant change points at approximately 2014 for total and female infants and 2016 for male infants; after these points, SIDS incidence increased significantly at an annual rate of approximately 8–9% across all groups (total: slope, +0.081; 95% CI, 0.053 to 0.110; male infants: slope, +0.089; 95% CI, 0.046 to 0.133; female infants: slope, +0.080; 95% CI, 0.035 to 0.126; all p<0.001).

SIDS incidence was significantly higher during the pandemic period than during the pre-pandemic period among all infants (IRR, 1.52; 95% CI, 1.28 to 1.80), male infants (IRR, 1.58; 95% CI, 1.26 to 1.96), and female infants (IRR, 1.43; 95% CI, 1.08 to 1.86; all p≤0.01). Incidence was also significantly higher during the post-pandemic period than during the pre-pandemic period among all infants (IRR, 1.34; 95% CI, 1.11 to 1.61) and male infants (IRR, 1.46; 95% CI, 1.14 to 1.83; both p=0.002), but not among female infants (IRR, 1.19; 95% CI, 0.87 to 1.60; p=0.270). The magnitude of these temporal increases did not differ significantly by sex during either the pandemic period (interaction p=0.578) or the post-pandemic period (interaction p=0.299). Annual IRR analysis using 2013 as the reference year showed that incidence was significantly elevated in 2018, 2020, and 2023, whereas incidence in 2021 and 2022 did not differ significantly from that in 2013 (Table 2).

To assess potential diagnostic coding drift, the annual proportion of SIDS among all infant OHCA cases was examined (Supplementary Material 1). This proportion showed significant interannual heterogeneity (chi-square test, p<0.001) and a significant overall temporal trend (Cochran-Armitage test, p=0.005), with non-monotonic fluctuations that were inconsistent with a gradual, unidirectional shift in diagnostic labeling.

Spatiotemporal analysis of sudden infant death syndrome incidence

Bayesian spatial analysis yielded posterior mean SIR ranges of 0.72–1.66 in the pre-pandemic period, 0.62–2.32 in the pandemic period, and 0.82–1.46 in the post-pandemic period (Figure 3). When spatial significance was evaluated using SIRs and 95% CrIs, Pyeongtaek-si, Gyeonggi-do, was the only district with a significantly elevated risk during the pre-pandemic period (posterior mean SIR, 1.60; 95% CrI, 1.01 to 2.42). No districts reached statistical significance during the pandemic or post-pandemic periods. Using EP>0.95, Pyeongtaek-si, Gyeonggi-do, remained significant during the pre-pandemic period (EP=0.98) (Figure 4). Jeju-si, Jeju Special Self-Governing Province, was also identified as a high-risk district during the pandemic period when EP>0.95 was used (EP=0.96; posterior mean SIR, 2.00; 95% CrI, 0.92 to 3.80). Using the stricter EP>0.99 threshold, no districts showed statistical significance in any of the 3 study periods. The spatial distribution of SIR changes between consecutive periods is presented in Supplementary Material 2.

Global spatial autocorrelation increased progressively across the 3 periods. Moran’s I was 0.359 in the pre-pandemic period, 0.654 in the pandemic period, and 0.952 in the post-pandemic period; all 3 values indicated statistically significant spatial clustering (all p=0.001 by Monte Carlo permutation test). Compared with the pre-pandemic period, Moran’s I increased significantly during the pandemic period (ΔI=0.295, 1-sided permutation p<0.001) and the post-pandemic period (ΔI=0.593, 1-sided permutation p<0.001). During the same period, median annual live births per district declined from 1,321 in the pre-pandemic period to 790 in the post-pandemic period, and the median expected SIDS case count per district decreased from 1.968 to 0.452 (Supplementary Material 3).

DISCUSSION

In this nationwide analysis of the OHCAS database from 2013 to 2023, we evaluated SIDS incidence in Korea. We identified distinct spatiotemporal variation in SIDS incidence, characterized by a rising overall burden and substantial geographic heterogeneity across administrative districts. To our knowledge, this is the first comprehensive analysis of nationwide SIDS epidemiology in Korea since the early 2000s. Specifically, our findings showed a significant increase in overall incidence after the pre-pandemic period and identified a southwestern geographic cluster, high-risk rigion. These findings support prioritizing targeted interventions in high-risk areas, including intensified public education on “Safe to Sleep” practices, and underscore the need for strengthened regional surveillance.

The most concerning finding was the sustained increase in SIDS incidence over the study period. Joinpoint regression identified a significant upward inflection in SIDS incidence beginning around 2014–2016, after which incidence increased by approximately 8–9% per year. This increase contrasts sharply with the sustained declines reported in many developed countries after widespread implementation of “Safe to Sleep” campaigns, suggesting that determinants of SIDS incidence in Korea may differ from those in Western populations and warrant further investigation [3,4].

The marked decline in breastfeeding rates in Korea, from 45.6% at 4–6 months in 2007 to 15.4% in 2020, is concerning because breastfeeding is a well-established protective factor against SIDS [14,15]. This sustained decline may have reduced breastfeeding-related protection at the population level. However, because SIDS incidence showed non-monotonic fluctuations rather than a parallel trend, declining breastfeeding alone is unlikely to be the primary driver. Instead, it may function as a contributing contextual factor whose effect on SIDS risk is modified by other concurrent trends, including changes in sleep practices, housing conditions, and pandemic-related disruptions.

The upward trajectory of SIDS incidence beginning in the mid-2010s likely reflects long-term secular trends, including changes in infant sleep practices, increasing co-sleeping prevalence, and changing housing conditions. We also investigated whether these patterns could reflect a gradual shift in diagnostic labeling or ascertainment bias. The non-monotonic fluctuations in the annual proportion of SIDS among infant OHCA cases argue against a consistent drift in diagnostic coding practices, suggesting that the observed temporal patterns likely reflect epidemiological variation rather than ascertainment bias. Nevertheless, because the OHCAS database lacks autopsy findings, misclassification of deaths from unrecognized infections, metabolic disorders, or non-accidental trauma within the broader spectrum of sudden unexpected infant death cannot be entirely excluded; such misclassification could have overestimated true SIDS incidence. Establishing a national autopsy-based surveillance system would support more precise cause-of-death ascertainment and should be prioritized in future research.

Interpretation of the increasing SIDS incidence must also account for Korea’s steep decline in birth rates, with live births approximately halving from 436,000 in 2013 to 230,000 in 2023. This shrinking denominator amplifies stochastic variability in incidence estimates and may partly explain the apparent surge in 2023, which was derived from an increasingly small national birth cohort. Although the BYM2 model reduces this instability through spatial shrinkage, residual volatility cannot be fully eliminated; therefore, the 2023 spike should be interpreted cautiously pending replication in future surveillance data. Future studies should explore alternative denominators, such as registered infant populations or population-based birth registries, and perform sensitivity analyses to examine the robustness of SIR estimates to denominator misspecification, particularly in districts experiencing extreme birth-rate decline or substantial cross-district healthcare utilization.

The temporary decline in SIDS incidence during the pandemic period may partly reflect the effects of non-pharmaceutical interventions (NPIs), including social distancing, mask use, and enhanced hygiene practices. These measures substantially reduced the circulation of respiratory viral infections, potentially reducing certain SIDS risk-modifying factors, such as co-sleeping with ill caregivers, indoor crowding, and respiratory triggers in vulnerable infants [16]. The rebound in SIDS incidence in 2023 may reflect the epidemiological consequences of NPI relaxation. Zhao et al. [17] demonstrated a sequential resurgence pattern of respiratory viruses after COVID-19 restrictions, with rhinovirus resurging earliest, followed by seasonal coronavirus, parainfluenza virus, respiratory syncytial virus (RSV), and influenza viruses. This viral resurgence may have been associated with increased SIDS incidence. Although causality cannot be inferred, these patterns suggest that pandemic-related changes may have affected SIDS incidence. Burrell et al. [18] reported that, as NPIs eased, respiratory viral transmission increased markedly, particularly for RSV and influenza, with interseasonal epidemics exceeding pre-pandemic levels. The authors attributed this pattern to population-level “immunity debt” after reduced pathogen exposure, in which susceptible infants, many with reduced maternally derived antibody protection, were simultaneously exposed to multiple respiratory pathogens. The relaxation of NPIs and return to pre-pandemic behaviors may have increased infant exposure to respiratory pathogens that had been less common during the pandemic, potentially contributing to the observed 2023 SIDS incidence peak.

The 3-period Bayesian spatial analysis demonstrated a progressive increase in Moran’s I across study periods, indicating that geographically proximate districts became increasingly similar in SIDS risk over time. Notably, the post-pandemic period showed the strongest spatial autocorrelation despite the narrowest SIR range, suggesting that this clustering reflects geographic patterning rather than the influence of extreme outliers. A persistent southwest gradient, with consistently elevated SIRs in Jeollanam-do and surrounding regions, was observed from the pandemic period onward and may reflect pre-existing structural inequalities in infant health that were amplified by pandemic-related disruptions. Southwestern Korea is characterized by higher rural residence rates and substantial disparities in healthcare accessibility. Fewer than 40% of municipalities in non-metropolitan areas such as Jeollanam-do have an equitable distribution of essential medical specialties, including emergency medicine, obstetrics and gynecology, and pediatrics [19]. In addition, the supply of essential primary care clinics in rural and small cities declined continuously during the pandemic period, widening pre-existing regional disparities [20]. These disparities may have made southwestern districts more vulnerable to disruptions in preventive care delivery during the pandemic. From a public health perspective, geographically targeted interventions are warranted in persistently higher-incidence regions, including strengthened parental education on “Safe to Sleep” practices, expanded home-visit nursing programs, and enhanced collaboration between community health centers and tertiary facilities. The physical infrastructure of essential healthcare services in underserved southwestern districts also requires expansion. Establishing additional pediatric and emergency medicine facilities in these regions could reduce structural barriers to specialized care and help address persistent regional disparities in infant health outcomes.

In the pre-pandemic period, Pyeongtaek-si, Gyeonggi-do, was the only district with a significantly elevated SIR. During the pandemic period, Jeju-si, Jeju Special Self-Governing Province, showed a significantly elevated EP, although the posterior mean SIR did not meet conventional significance thresholds. The factors underlying these regional elevations—including local demographic characteristics, healthcare accessibility, socioeconomic conditions, and population structure—could not be determined from the available data. Further studies incorporating district-level contextual data are warranted to clarify the drivers of these regional disparities.

This study has several limitations. First, the OHCAS database does not include infant age in months or autopsy findings. Given the extremely low autopsy rate for infant deaths in Korea, true SIDS incidence may have been overestimated because of potential misclassification [21]. Second, the analysis did not include adjustment for district-level ecological covariates. Accordingly, the reported SIRs are crude estimates, and the observed spatial variation may be influenced by unmeasured confounding. Future studies should establish an integrated longitudinal database that incorporates district-level data on breastfeeding, deprivation indices, financial independence ratios, healthcare accessibility, population density, and urbanization indices to formally test these ecological associations. Third, the unequal observation windows may have yielded less stable SIR estimates in the later periods. Fourth, although BYM2 shrinkage reduces the effective number of independent comparisons by borrowing strength across neighboring districts, multiple-comparison concerns across 250 administrative units cannot be entirely dismissed. Posterior exceedance probabilities were reported as a complementary measure; nonetheless, estimates from districts with small observed counts should be interpreted with caution. Fifth, district-level registered live births may not fully capture the true population at risk in districts with extreme birth-rate decline or substantial cross-district healthcare utilization. Finally, several clinical variables, such as bystander response and initial rhythm, had high rates of missing data.

In summary, SIDS incidence was higher and spatial clustering was more pronounced during the pandemic and post-pandemic periods than during the pre-pandemic period. The observed increase in SIDS incidence and widening regional disparities underscore the need for geographically targeted interventions, including safe sleep education, breastfeeding promotion, and expanded home-visit nursing programs, particularly in high-risk rigion. In addition, improved surveillance systems that incorporate detailed socioeconomic data and autopsy findings would strengthen future epidemiological monitoring and research. Taken together, these findings support sustained vigilance and renewed SIDS prevention efforts in Korea.

Data availability

Supplementary materials

NOTES

Conflict of interest

The authors have no conflicts of interest to declare for this study.

Funding

This research was supported by the research promotion project of Catholic Kwandong University International St. Mary’s Hospital.

Acknowledgements

The authors acknowledge the contributions of the Korea Disease Control and Prevention Agency investigators.

Author contributions

Conceptualization: Cha M, Jeon S, Yun S. Data curation: Cha M. Formal analysis: Cha M. Funding acquisition: Cha M. Methodology: Jeon S, Cha M. Project administration: Cha M. Visualization: Jeon S, Cha M. Writing – original draft: Jeon S, Cha M. Writing – review & editing: Jeon S, Cha M, Yun S.

Figure 1.

Flowchart of the participant selection process. OHCA, out-of-hospital cardiac arrest; SIDS, sudden infant death syndrome.

Figure 2.

Annual incidence of sudden infant death syndrome cases per 100,000 live births.

Figure 3.

Spatiotemporal distribution of the posterior mean standardized incidence ratios (SIRs) for sudden infant death syndrome across 3 study periods (A) 2013-2019, (B) 2020-2021, and (C) 2022-2023.

Figure 4.

Spatiotemporal distribution of the posterior exceedance probabilities for sudden infant death syndrome across 3 study periods (A) 2013-2019, (B) 2020-2021, and (C) 2022-2023.

REFERENCES

• 1. Willinger M, James LS, Catz C. Defining the sudden infant death syndrome (SIDS): deliberations of an expert panel convened by the National Institute of Child Health and Human Development. Pediatr Pathol 1991;11:677-684. https://doi.org/10.3109/15513819109065465ArticlePubMed
• 2. de Luca F, Hinde A. Effectiveness of the ‘Back-to-Sleep’ campaigns among healthcare professionals in the past 20 years: a systematic review. BMJ Open 2016;6:e011435. https://doi.org/10.1136/bmjopen-2016-011435ArticlePubMedPMC
• 3. Park S, Han JH, Hwang J, Yon DK, Lee SW, Kim JH, et al. The global burden of sudden infant death syndrome from 1990 to 2019: a systematic analysis from the Global Burden of Disease study 2019. QJM 2022;115:735-744. https://doi.org/10.1093/qjmed/hcac093ArticlePubMed
• 4. Goldstein RD, Trachtenberg FL, Sens MA, Harty BJ, Kinney HC. Overall postneonatal mortality and rates of SIDS. Pediatrics 2016;137:e20152298. https://doi.org/10.1542/peds.2015-2298Article
• 5. Erck Lambert AB, Parks SE, Shapiro-Mendoza CK. National and state trends in sudden unexpected infant death: 1990-2015. Pediatrics 2018;141:e20173519. https://doi.org/10.1542/peds.2017-3519ArticlePubMedPMC
• 6. Manfredini M, Perrone S, Ardenghi A, Lavezzi AM, Beretta V, Scarpa E, et al. The changing epidemiology of sudden infant death syndrome: a 15-year overview comparing Italian and European data. Health Sci Rep 2025;8:e70599. https://doi.org/10.1002/hsr2.70599ArticlePubMedPMC
• 7. Ha M, Yoon SJ, Lee HY, Goh UY, Kim CH, Lee YS. Estimation of the incidence of sudden infant death syndrome in Korea: using the capture-recapture method. Paediatr Perinat Epidemiol 2004;18:138-142. https://doi.org/10.1111/j.1365-3016.2003.00544.xArticlePubMed
• 8. Doan TN, Wilson D, Rashford S, Ball S, Bosley E. Spatiotemporal variation in the risk of out-of-hospital cardiac arrests in Queensland, Australia. Resusc Plus 2021;8:100166. https://doi.org/10.1016/j.resplu.2021.100166ArticlePubMedPMC
• 9. Riebler A, Sørbye SH, Simpson D, Rue H. An intuitive Bayesian spatial model for disease mapping that accounts for scaling. Stat Methods Med Res 2016;25:1145-1165. https://doi.org/10.1177/0962280216660421ArticlePubMed
• 10. Morrison LJ, Nichol G, Rea TD, Christenson J, Callaway CW, Stephens S, et al. Rationale, development and implementation of the Resuscitation Outcomes Consortium Epistry-Cardiac Arrest. Resuscitation 2008;78:161-169. https://doi.org/10.1016/j.resuscitation.2008.02.020ArticlePubMedPMC
• 11. Grasner JT, Bray JE, Nolan JP, Iwami T, Ong ME, Finn J, et al. Cardiac arrest and cardiopulmonary resuscitation outcome reports: 2024 update of the Utstein Out-of-Hospital Cardiac Arrest Registry template. Resuscitation 2024;201:110288. https://doi.org/10.1016/j.resuscitation.2024.110288ArticlePubMed
• 12. Good P. Permutation, parametric and bootstrap tests of hypotheses. 3rd ed. New York: Springer; 2005, https://doi.org/10.1007/b138696
• 13. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370:1453-1457. https://doi.org/10.1016/S0140-6736(07)61602-XArticlePubMedPMC
• 14. Kang M, Choi EK, Lee JM, Shin HJ, Lee WR, Shin SM, et al. Trends and determinants in breastfeeding among Korean infants (2007-2021): a nationwide study using the National Health Screening Program for Infants and Children. Clin Exp Pediatr 2025;68:772-780. https://doi.org/10.3345/cep.2025.00857ArticlePubMedPMC
• 15. Moon RY, Carlin RF, Hand I; Task Force on Sudden Infant Death Syndrome and the Committee on Fetus and Newborn. Sleep-related infant deaths: updated 2022 recommendations for reducing infant deaths in the sleep environment. Pediatrics 2022;150:e2022057990. https://doi.org/10.1542/peds.2022-057990ArticlePubMed
• 16. Chow EJ, Uyeki TM, Chu HY. The effects of the COVID-19 pandemic on community respiratory virus activity. Nat Rev Microbiol 2023;21:195-210. https://doi.org/10.1038/s41579-022-00807-9ArticlePubMedPMC
• 17. Zhao C, Zhang T, Guo L, Sun S, Miao Y, Yung CF, et al. Characterising the asynchronous resurgence of common respiratory viruses following the COVID-19 pandemic. Nat Commun 2025;16:1610. https://doi.org/10.1038/s41467-025-56776-zArticlePubMedPMC
• 18. Burrell R, Saravanos G, Britton PN. Unintended impacts of COVID-19 on the epidemiology and burden of paediatric respiratory infections. Paediatr Respir Rev 2025;53:3-13. https://doi.org/10.1016/j.prrv.2023.07.004ArticlePubMed
• 19. Park J, Jeon B, Lee E. Regional disparities in the distribution of public and private healthcare facilities in South Korea. PLoS One 2025;20:e0330090. https://doi.org/10.1371/journal.pone.0330090ArticlePubMedPMC
• 20. Jin HY, Ju J. Deepening regional disparities in primary health care during COVID-19 in South Korea. Rural Remote Health 2024;24:8612. https://doi.org/10.22605/RRH8612Article
• 21. Yoo SH, Kim AJ, Kang SM, Lee HY, Seo JS, Kwon TJ, et al. Sudden infant death syndrome in Korea: a retrospective analysis of autopsy-diagnosed cases. J Korean Med Sci 2013;28:438-442. https://doi.org/10.3346/jkms.2013.28.3.438ArticlePubMedPMC

Figure & Data

References

Citations

Citations to this article as recorded by