The global burden of sudden infant death syndrome from 1990 to 2019: a systematic analysis from the Global Burden of Disease study 2019

Summary

Background

Sudden infant death syndrome (SIDS) still remains one of the leading causes of infant death worldwide, especially in high-income countries. To date, however, there is no detailed information on the global health burden of SIDS.

Aims

To characterize the global disease burden of SIDS and its trends from 1990 to 2019 and to compare the burden of SIDS according to the socio-demographic index (SDI).

Design

Systematic analysis based on the Global Burden of Disease (GBD) 2019 data.

Methods

Epidemiological data of 204 countries from 1990 to 2019 were collected via various methods including civil registration and vital statistics in the original GBD study. Estimates for mortality and disease burden of SIDS were modeled. Crude mortality and mortality rates per 100 000 population were analyzed. Disability-adjusted life years (DALYs) and DALY rates were also assessed.

Results

In 2019, mortality rate of SIDS accounted for 20.98 [95% Uncertainty Interval, 9.15–46.16] globally, which was a 51% decrease from 1990. SIDS was most prevalent in Western sub-Saharan Africa, High-income North America and Oceania in 2019. The burden of SIDS was higher in males than females consistently from 1990 to 2019. Higher SDI and income level was associated with lower burden of SIDS; furthermore, countries with higher SDI and income had greater decreases in SIDS burden from 1990 to 2019.

Conclusions

The burden of SIDS has decreased drastically from 1990 to 2019. However, the improvements have occurred disproportionately between regions and SDI levels. Focused preventive efforts in under-resourced populations are needed.

Introduction

Sudden infant death syndrome (SIDS) is defined as the sudden, unexplained death of an infant younger than 1 year diagnosed through a comprehensive investigation including a review of the clinical history, examination of the death scene and autopsy.¹ SIDS is widely believed to be caused by an array of factors including genetic, environmental and sociocultural. Sleeping prone, bed sharing, soft bedding, prematurity, maternal smoking, male sex and some racial variances are known risk factors of SIDS while breastfeeding and use of pacifiers are considered to have protective effects.^2,3 Despite recent advances in understanding the pathophysiology of SIDS, it is currently diagnosed only by exclusion.²

SIDS still remains one of the leading causes of infant death in high-income countries, while its incidence has dramatically decreased starting in the 1990s following risk reduction campaigns and continuous public health efforts.^4,5 In the USA, more than 1900 infants die every year by SIDS, showing ∼0.49 deaths per 1000 live births.² In Italy, the incidence is about 1 out of 1000 live births.⁶ However, the SIDS rate has not decreased since 2006.^2,7 There is a coincidence between the stop in the decrease of SIDS and the stop in the increase of supine sleep.⁸

For making evidence-based care plans and public policies, accurate estimates of global health burden of SIDS are crucial. However, to date, there are few studies detailing the health burden of SIDS and there is no detailed information on it across all countries. A study using Global Burden of Disease, Injuries and Risk Factors Study (GBD) 2017 has reported worldwide SIDS burden as a part of an analysis of child and adolescent health issues. In the study, SIDS was ranked the third cause of disability-adjusted life years (DALYs) in children under 1 year of age in the high-socio-demographic index (SDI) quintile and in the top 10 for all high-income countries.⁹ However, the data on SIDS burden was limited because it did not focus exclusively on SIDS.

The GBD provides a systematic, scientific assessment of published, publicly available and contributed data on diseases and injuries worldwide.¹⁰ This study aims to investigate and provide the most recent comprehensive information on the global SIDS burden. We analyzed the global, regional and national-level data including mortality rates, years of life lost (YLLs) and DALYs of SIDS from the latest GBD research, GBD 2019, and identified changes from 1990 to 2019. Moreover, we investigated SIDS burden in terms of age, sex, income and SDI (a composite indicator of socio-demographic factors including income, education and fertility rate¹⁰).

Methods

Data source

All data were derived from the GBD 2019 Results Database, publicly available through the Global Burden of Disease Collaborative Network website (http://ghdx.healthdata.org). The GBD 2019, performed by the Institute of Health Metrics and Evaluation at the University of Washington and supported by the Bill and Melinda Gates Foundation, provides the most recent and scientific analysis of the descriptive epidemiology of 369 diseases and injuries in 204 countries and territories.¹⁰ Epidemiological data were collected in various ways including systematic reviews of the literature, civil registration and vital statistics, health service visits, surveillance data, survey data and medical claims information. Estimates for incidence, prevalence and disease burden were modeled through three main standardized tools, including (i) Cause of Death Ensemble model (CODEm), an optimized tool for analyzing cause of death by collecting an ensemble of different modeling methods with varying choices of covariates for high predictive validity; (ii) Spatiotemporal Gaussian Process Regression, a model for analyzing and comparing temporal and regional estimates between different groups; (iii) DisMod-MR 2.1, a Bayesian meta-regression tool, used to ensure consistency between epidemiological parameters including incidence, prevalence, remission and mortality by adjusting variations of heterogeneous data sets from different modeling methods.¹⁰ Further details about the GBD 2019 methodology can be found in the literature describing the original GBD projects.^10,11 For estimation of deaths due to SIDS, the standard CODEm modeling method was used.¹⁰

Definitions

Disease burden was estimated through DALYs, a time-based composite measure to quantify overall burden of a given condition defined as the sum of YLL due to premature mortality (YLLs) and years of life lived in states of less-than-optimal health, in other words, years of life with disability (YLDs). Simply put, one DALY represents the loss of 1 year of healthy, normal life. To calculate YLLs, the number of cause-specific deaths was multiplied by standard life expectancy at age of death. YLDs were determined by the product of a prevalence estimate and a disability weight, which represents the level of disability by each health problem.^10,11 Because SIDS is defined by mortality, in SIDS, DALYs equals YLLs.

The level of socioeconomic development of the countries was represented by SDI, a summary measure used since GBD 2017, which is a value between 0.0 and 1.0 determined by three covariates, income per capita, educational attainment and total fertility rate in women under the age of 25 years.¹¹

Generally, in epidemiologic studies, age-standardized rates are used to make comparisons between temporally or regionally distinct populations while accounting for differences in population age structure. However, as the definition of SIDS restricted patient age to infants under 1 year, we mainly used age-specific crude mortality and DALY rates for groups under 1 year of age. Age-standardized mortality and DALY rates are provided separately in the Supplementary material.

Analysis

From GBD Study 2019 Results Database, we extracted the temporal and regional worldwide data for mortality and DALYs of SIDS in terms of age, sex, income and SDI. Analysis of SIDS burden by region was performed by subdividing up to Level 4 regions, which mean the regions of subnational level,¹⁰ to identify the areas that were problematic due to SIDS in more detail. GBD 2019 provides Level 4 analysis of a total of 21 countries, an increase of 5 countries compared to previous GBD 2017¹⁰ (Brazil, China, Ethiopia, India, Indonesia, Iran, Japan, Kenya, Mexico, New Zealand, Norway, Russia, South Africa, Sweden, the UK, the USA, Italy, Nigeria, Pakistan, the Philippines and Poland). For the analysis of DALY rates of SIDS by SDI, the expected DALY rate according to SDI was additionally analyzed using smoothing splines models¹² to determine the shape of association between DALY rate of SIDS and SDI for 21 GBD regions, more focusing on the configuration of dose–response relationships than model fit. Uncertainty intervals (UI) were defined as the 25th and 975th values of 1000 sample draws for each variable. It allows us to confirm the level of the evidence for each data point.

Results

Global and regional burden of SIDS

In 2019, total estimated number of deaths from SIDS worldwide was 27 685 [95% UI, 12 083–60 919] and the global burden of SIDS expressed in DALYs was 2 448 496 [95% UI, 1 068 808–5 387 495] per 100 000 person-years. Crude mortality rate of SIDS was 20.98 [95% UI, 9.15–46.16] per 100 000 infants and crude DALY rate, 1855.13 [95% UI, 809.79–4081.89] per 100 000 infants (Supplementary eTable 1). From 1990 to 2019, both mortality and DALY rates of SIDS showed a steady decline, decreasing by 51% [95% UI, –62% to –30%]. The rate of decline was relatively high from 1990 to 2000, but since then have been slow and steady. Given the absence of YLD in calculating DALY of SIDS, the mortality rate and DALY rate due to SIDS were directly proportional.

Of the 21 GBD regions in the world, Western sub-Saharan Africa (41.80 [95% UI, 6.19–122.3] per 100 000 infants), High-income North America (38.18 [95% UI, 34.17–42.26] per 100 000 infants) and Oceania (32.94 [95% UI, 7.14–104.23] per 100 000 infants) showed the highest mortality rates in 2019 (Table 1). The three lowest mortality rates, on the other hand, were East Asia (5.91 [95% UI, 3.10–8.22] per 100 000 infants), Central Asia (6.63 [95% UI, 4.99–8.76] per 100 000 infants) and Tropical Latin America (7.83 [95% UI, 6.06–10.05] per 100 000 infants), followed by High-income Asia Pacific (8.93 [95% UI, 6.48–11.52] per 100 000 infants). From 1990 to 2019, mortality rates decreased in almost all regions, except for Tropical Latin America (3.55% [95% UI, −28.73% to 40.02%]). Australasia had the highest reduction in mortality rate (–89.25% [95% UI, –91.90% to –85.31%]), whereas Central Latin America was the lowest (–1.33% [95% UI, –34.92 to 42.8]).

The geographic distribution of the mortality and DALY rates of SIDS according to country and sub-country level territories are depicted in Figure 1, Supplementary eFigures S1, S2 and eTables S3, S4. In the late neonatal period (7–27 days after birth), Tlaxcala in Mexico showed the highest mortality rate (176.95 [95% UI, 118.55–256.31] per 100 000 infants), followed by North Kali (168.08 [95% UI, 74.83–330.95] per 100 000 infants) and Papua (164.10 [95% UI, 68.53–313.97] per 100 000 infants), both in Indonesia. In contrast, in the post-neonatal period (28–364 days after birth), Arkansas in USA had the highest mortality rate (107.18 [95% UI, 86.97–128.08] per 100 000 infants), and Sierra Leone (92.06 [95% UI, 9.40–328.81] per 100 000 infants) in Western sub-Saharan Africa, the next. Notably, six other states in the USA, namely West Virginia, North Dakota, Kentucky, Oklahoma, Alaska and Georgia ranked in the top fourth to ninth highest mortality rates in the post-neonatal period, ranging from 82.34 [95% UI, 58.72–113.22] per 100 000 infants in West Virginia to 72.53 [95% UI, 58.62–88.78] per 100 000 infants in Georgia.

Burden of SIDS by sex

Globally, mortality rates of SIDS were slightly higher in males than females in 2019, a trend that has continued from 1990 (Figure 2 and Supplementary eTable 5). When stratified according to GBD region, South Asia and North Africa/Middle East were the only regions with higher mortality rates for females, though the difference in mortality rates between genders was small in North Africa/Middle East. Mortality rates in males ranged from 7.42 [95% UI, 5.14–10.19] per 100 000 male infants in Central Asia to 45.61 [95% UI, 6.31–134.07] per 100 000 male infants in Western sub-Saharan Africa, whereas those in females ranged from 3.22 [95% UI, 1.83–5.18] per 100 000 female infants in East Asia to 37.86 [95% UI, 5.54–116.04] per 100 000 female infants in Western sub-Saharan Africa.

Burden of SIDS according to SDI and income

The disease burden of SIDS in 1990 and 2019 according to SDI quintiles are shown in Table 2 and Supplementary eFigure 5. The percent decrease in mortality and DALY rates from 1990 to 2019 was larger with higher SDI quintile. Mortality and DALY rates in the high SDI quintile decreased the most (–75.91% [95% UI, –79.27% to –71.69%]), and those in the low SDI quintile the least (–43.72% [95% UI, –59.68% to 43.35%]). In 2019, the low SDI quintile had the highest DALY rate of SIDS, whereas the high-middle SDI quintile had the lowest DALY rate. Furthermore, when DALY rates and SDI in each territory were plotted by geographic region, DALY rates in a given GBD region showed a negative correlation with SDI values (Figure 3).

The crude DALY number and rate trends of SIDS according to income level from 1990 to 2019 are shown in Supplementary eFigure 3. Overall, the number of DALY and DALY rates decreased from 1990 to 2019 regardless of income. However, similar to trends according to SDI quintiles, the decrease in mortality rates was largest in the high-income group and lower in other income groups.

Discussion

Although an accumulation of studies on the pathophysiology and risk factors of SIDS has led to an improved understanding of SIDS, it still remains a significant cause of death in the first year of life with racial/ethnic disparities.^3,13–15 As global data on SIDS are lacking apart from scattered systematic reviews on sudden unexpected infant death,^16,17 this study utilized the GBD 2019 data to characterize trends in mortality and disease burden of SIDS from 1990 to 2019 and compared these trends according to income and SDI.

SIDS accounted for 27 685 deaths globally in 2019, which was less than half of the 55 995 deaths in 1990. Following that trend, in most regions, both mortality and DALYs rates of SIDS decreased significantly from 1990 to 2019, with the most remarkable changes occurring in the 1990s. Notably, there were statistically significant changes in High-income Asia Pacific, Australasia, Central, Eastern and Western Europe, Southern Latin America, North Africa/Middle East and High-income North America regions. Australasia had the largest percent decrease in mortality and DALYs rates, followed by Western Europe, both of which had over 85% decrease in SIDS disease burden. This is in accordance with various national statistics describing significant declines in SIDS mortality in similar timeframes after preventive efforts targeting known modifiable external stressors, such as prone positioning, soft bedding, overheating and secondhand smoke exposure.^5,14 A representative example is the ‘Back to Sleep’ (now ‘Safe to Sleep’) campaign, which resulted in drastic decreases in SIDS rates in many North American, European and Australasian countries such as the USA, UK, New Zealand and Norway.^18–21 Furthermore, the slowing down of the decline in SIDS mortality since 2000 could be explained by more sensitive diagnostic classification and reporting of SIDS (i.e. diagnostic shift).^22–24

As countries like Australasia, Western Europe and High-income North America, which had the highest mortality rates due to SIDS in 1990, experienced a drastic decline in mortality, the variation according to geographic region, which ranged from 7.56–188.43 per 100 000 infants in 1990, lessened in 2019, ranging from 5.91–41.80 per 100 000 infants. However, some regions, such as Central Latin America, Tropical Latin America and Oceania, experienced very small changes in SIDS rates. In these regions, focused efforts to target the occurrence and mortality of SIDS should be made. Regions which already experienced a large decline in SIDS rates also need interventions, as some of these regions, such as High-income North America and Western sub-Saharan Africa, still had the highest SIDS rates, and disparities in preventive efforts and SIDS mortality may persist in these regions as well.^13–15

Although high SDI countries had the highest burden of SIDS in 1990, they experienced the greatest decline in the burden of SIDS; in 2019, the low SDI countries had the highest mortality rates and DALYs rates of SIDS. When analyzed within GBD regions, SIDS rates correlated negatively with SDI. This trend was replicated when countries were classified by World Bank income quartiles; high SDI countries had the steepest decline in SIDS burden, such that in 2019, SDI burden was largest in countries with low income. This is concordant with the previous literature stating an association between individual low socioeconomic status and SIDS^5,18,25,26; in addition to low economic status being a risk factor, through our data, we can see that countries with the lowest SDI also had the smallest average decrease in SIDS rates.

Although SIDS burden has decreased globally, these differential changes in the incidence of SIDS over the world by region and by socioeconomic status signals a major shift in SIDS epidemiology.^18,21 Studies report that now the majority of SIDS cases occur in high-risk families with multiple risk factors including low socioeconomic status and specific demographic populations,^5,18,25,26 and that the social inequalities are widening.²⁷ As much as preventive campaigns have been effective to date, it is important to address the fundamental causes of health disparities²⁸ and ensure equitable access to these behavioral interventions, which may in turn reduce the disparities.

This study is the first study to assess the global burden of SIDS with the comprehensive data from the GBD study. It was also the first study to examine associations between country-level SDI and income with SIDS burden. However, there are some limitations. First, this study utilized data from the GBD study, but was limited by the data available as in other GBD studies. Namely, there were regions with low availability or quality of incidence and prevalence data. In those regions, statistics relied on predictive values computed from the GBD statistical models. Specifically for this report, since GBD 2017, all deaths assigned to SIDS outside of the countries without sufficient quality of data available were reassigned to neonatal disorders.¹⁰ It means that there was a possibility of selection bias according to the level of quality of data available. Second, there may be some differences between the actual data collected by the governments or other organizations and the data obtained by estimation in this study. It is one of the general limitations of the GBD studies.^9,10,29 For this, the studies using the GBD dataset may be more meaningful in cases where it is hard to obtain actual data. Considering that SIDS also corresponds to one of those,^5,13,27,30 this study clearly has important implications notwithstanding this limitation. Third, associations we found regarding SDI and income were based on aggregate data on the national or sub-national territory level. Therefore, these associations are subject to ecological bias; caution is needed when inferring these associations to the individual level. Finally, as SIDS is a challenging diagnosis in the clinical setting based on exclusion, and considering that it is ambiguous enough to distinguish it from sudden unexplained infant death (SUID), the mortality and DALYs attributed to SIDS may have varied according to country and subject to bias. Despite these limitations, this study is of great significance in that it is the first large-scale study to analyze SIDS globally, which is still an unresolved issue.

In conclusion, the disease burden of SIDS has decreased drastically globally from 1990 to 2019, with the steepest changes occurring from 1990 to 2000 overall. However, the distribution of these changes was not equal, with countries with higher social and developmental status having the greatest reduction in disease rates. Proven preventive measures and community campaigns focused on under-resourced populations are needed to reduce these disparities and further reduce the global burden of SIDS, even in highly developed countries.

Supplementary material

Supplementary material is available at QJMED online.

Funding

This research was not funded by any funding source.

Conflict of interest: None declared.

Data Availability

Data used for the analyses are publicly available via online data source tools (http://ghdx.healthdata.org).

References

Author notes