Global, regional, and national prevalence and mortality burden of sickle cell disease, 2000-2021: a systematic analysis from the Global Burden of Disease Study 2021

Abstract

Background: Previous global analyses, with known underdiagnosis and single cause per death attribution systems, provide only a small insight into the suspected high population health effect of sickle cell disease. Completed as part of the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2021, this study delivers a comprehensive global assessment of prevalence of sickle cell disease and mortality burden by age and sex for 204 countries and territories from 2000 to 2021.

Methods: We estimated cause-specific sickle cell disease mortality using standardised GBD approaches, in which each death is assigned to a single underlying cause, to estimate mortality rates from the International Classification of Diseases (ICD)-coded vital registration, surveillance, and verbal autopsy data. In parallel, our goal was to estimate a more accurate account of sickle cell disease health burden using four types of epidemiological data on sickle cell disease: birth incidence, age-specific prevalence, with-condition mortality (total deaths), and excess mortality (excess deaths). Systematic reviews, supplemented with ICD-coded hospital discharge and insurance claims data, informed this modelling approach. We employed DisMod-MR 2.1 to triangulate between these measures-borrowing strength from predictive covariates and across age, time, and geography-and generated internally consistent estimates of incidence, prevalence, and mortality for three distinct genotypes of sickle cell disease: homozygous sickle cell disease and severe sickle cell β-thalassaemia, sickle-haemoglobin C disease, and mild sickle cell β-thalassaemia. Summing the three models yielded final estimates of incidence at birth, prevalence by age and sex, and total sickle cell disease mortality, the latter of which was compared directly against cause-specific mortality estimates to evaluate differences in mortality burden assessment and implications for the Sustainable Development Goals (SDGs).

Findings: Between 2000 and 2021, national incidence rates of sickle cell disease were relatively stable, but total births of babies with sickle cell disease increased globally by 13·7% (95% uncertainty interval 11·1-16·5), to 515 000 (425 000-614 000), primarily due to population growth in the Caribbean and western and central sub-Saharan Africa. The number of people living with sickle cell disease globally increased by 41·4% (38·3-44·9), from 5·46 million (4·62-6·45) in 2000 to 7·74 million (6·51-9·2) in 2021. We estimated 34 400 (25 000-45 200) cause-specific all-age deaths globally in 2021, but total sickle cell disease mortality burden was nearly 11-times higher at 376 000 (303 000-467 000). In children younger than 5 years, there were 81 100 (58 800-108 000) deaths, ranking total sickle cell disease mortality as 12th (compared to 40th for cause-specific sickle cell disease mortality) across all causes estimated by the GBD in 2021.

Interpretation: Our findings show a strikingly high contribution of sickle cell disease to all-cause mortality that is not apparent when each death is assigned to only a single cause. Sickle cell disease mortality burden is highest in children, especially in countries with the greatest under-5 mortality rates. Without comprehensive strategies to address morbidity and mortality associated with sickle cell disease, attainment of SDG 3.1, 3.2, and 3.4 is uncertain. Widespread data gaps and correspondingly high uncertainty in the estimates highlight the urgent need for routine and sustained surveillance efforts, further research to assess the contribution of conditions associated with sickle cell disease, and widespread deployment of evidence-based prevention and treatment for those with sickle cell disease.

Funding: Bill & Melinda Gates Foundation.

Figure 1

Flowchart showing prevalence, cause-specific mortality, and total sickle cell disease mortality estimation process Overview of the estimation process for sickle cell disorders for both cause-specific mortality on the basis of the International Classification of Diseases definition of one cause per death and total sickle cell disease mortality. Shapes differentiate input data, disease measures, processes, and estimates, whereas colours distinguish different modelling tools. The four steps are highlighted: (1) sickle cell disease-specific data on disease epidemiology were inputted into DisMod-MR 2.1 models to generate internally consistent incidence, prevalence, and overall mortality; (2) causes of death data processed across GBD were run through a series of CODEm to generate mortality estimates for total haemoglobinopathies (parent GBD cause) for all locations and for data rich locations only, for each of the child causes, including sickle cell disease, thalassaemias, G6PD deficiency, and other haemoglobinopathies; (3) the mortality results from each of the four subcauses were scaled to 100%; and (4) these mortality results were used to divide total haemoglobinopathies CODEm results into cause-specific mortality results for sickle cell disease and other subcauses. CODEm=Causes of Death Ensemble model. G6PD=glucose-6-phosphate dehydrogenase.

Figure 2

Data availability map showing types of input data source measures present in each country for sickle cell disease DisMod-MR 2.1 models Each colour represents a different combination of birth incidence, age-specific prevalence, and survival or mortality data that were used as inputs for DisMod-MR 2.1 models for each of the three estimated genotypes of sickle cell disease. The outputs of these models include estimated birth incidence, age-specific prevalence, and total sickle cell disease mortality for each GBD location, age group, sex, and year (see appendix for genotype specific counts [pp 95–178] and for genotype specific rates [pp 207–90]). The specific sources are in the appendix (pp 15–22) and online at the Global Health Data Exchange; contributing counts of prevalence and mortality measures from each genotype are also in the appendix (p 50).

Figure 3

Maps of total sickle cell disease rates per 100 000 population (A) Birth incidence. (B) All-age prevalence. (C) All-age cause-specific mortality. (D) All-age total sickle cell disease mortality among males and females combined in 2021. See the appendix (p 63) for all measures in 2000.

Figure 4

Time trends of sickle cell disease cause-specific mortality (top) and total sickle cell disease mortality (bottom) as a cause fraction of total mortality within the super-region (A) and rate within the super-region (B), by age group Solid lines are coloured by super-region and display the trends of sickle cell disease mortality measures between 2000 and 2021. The shaded area represents the 95% uncertainty interval.

Figure 5

Ranking of total and cause-specific sickle cell disease mortality in children younger than 5 years compared with other GBD causes in 2021, and comparison of cause-specific mortality rates to total sickle cell disease mortality rates in individuals aged 5–14 years and 15–49 years Total under-5 mortality deaths of children with sickle cell disease ranked together with sickle cell disease cause-specific deaths and the leading 20 causes of death globally for each super-region (A) and ranked as a proportion of all-cause under-5 mortality by country (B). The 2021 cause-specific mortality rates are plotted along the x-axis compared with total sickle cell disease mortality rates plotted along the y-axis for the 5–14 years age group (C) and 15–49 years age group (D). *Sickle cell disease cause-specific deaths are not included in the 20 leading causes but are included for comparison to total sickle cell disease. See the appendix (p 65) for rankings by country in 2000.