Analysis of the Global Burden of Disease study highlights the trends in death and disability-adjusted life years of leukemia from 1990 to 2017

2.1 Data sources

Data on leukemia was collected from the latest version of the GBD study by using the Global Health Data Exchange (GHDx) query tool (http://ghdx.healthdata.org/gbd-results-tool). According to the GBD 2017 online tools instruction, the number and rate of death and DALYs were extracted between 1990 and 2017 based on age, sex, subtype, region, and country, without any inclusion/exclusion criteria. Data on the four main subtypes of leukemia, including acute myeloid leukemia (AML), ALL, CLL, and CML were gathered. The socio-demographic index (SDI) areas were categorized into 5 levels including low, low-middle, middle, high-middle, and high. According to the geographical feature, the world was divided into 21 geographic regions, including East Asia, South Asia, and Western Europe. Data were available for 195 countries/territories, including China, India, and France. The methodologies of the overall GBD 2017 and estimations of disease burden were as described in previous studies [3, 15, 17]. The Human Development Index (HDI) comprises of macro indicators of human development, reflecting the level and availability of health resources. The HDI data was acquired from the United Nations Development Programme (UNDP; http://hdr.undp.org/en/data).

2.2 Statistical analysis

In the formula, α_i denotes the age-specific rates in the i^th age group, w_i denotes the number of persons (or the weight) in the corresponding i^th age subgroup of the selected reference standard population, and A denotes the number of age groups.

It is a widely accepted method that the EAPC describes the magnitude of the ASR trend change in a regression model [18-21], which quantifies the average annual rate of change in ASR over the given interval. A regression line was used to estimate the natural logarithm of the rates, for example, y = α+βx+ε, where y = ln(ASR), and x = calendar year. The EAPC was calculated as 100 × [exp(β)-1], along with the 95% confidence interval (CI) with the linear regression model. An increasing trend was observed when the EAPC value and its 95% CI were > 0. In contrast, a decreasing trend was observed when the EAPC value and its 95% CI were < 0. Others represented that ASR was stable over time. To explore the impact factors of EAPC, the relations between EAPCs and ASR in 1990 and between EAPCs and HDI in 2017, respectively, were estimated using the Pearson correlation analysis in the Statistical Package for Social Sciences (SPSS; version 25.0; SPSS Inc., Chicago, IL, USA). Choropleth maps were drawn using the R program (version 3.6.2).

3.1 Trends in death and DALYs due to leukemia

The global death number of leukemia was 347.58 × 10³ (95% uncertainty interval [UI] = 317.26 × 10³-364.88 × 10³) in 2017, with a 31.22% increase from 1990 to 2017 (Table 1). The age-standardized death rate (ASDR) showed a decreasing trend with an annual average of 1.04% from 1990 to 2017 (EAPC = −1.04; 95% CI = −1.10-−0.99; Table 1 and Figure 1A). The burden of death and DALYs were higher and showed lower decreasing trends in male than in female leukemia patients (Table 1). A dramatic decrease in percentages of absolute death numbers in leukemia was observed in children and adolescents (Supplementary Table S1 and Figure 1B). The trend in ASDR decreased in both sexes and most of the SDI areas and geographic regions (Table 1 and Figure 1A). Among the 195 countries/territories, 126 countries/territories showed decreasing trends in ASDR of leukemia and the largest decreasing one occurred in Bahrain, with an EAPC of −3.01 (95% CI = −3.47-−2.55), followed by Finland, and Australia. The trends of ASDR increased in forty-five countries/territories, and the most pronounced increase was in Georgia (EAPC = 1.73; 95% CI = 1.30-2.16; Supplementary Table S2 and Figure 2A–C).

Globally, the number of DALYs caused by leukemia was 11975.35 × 10³ (95% UI = 10749.15 × 10³-12793.58 × 10³) in 2017. The ASR of DALYs had a decreasing trend from 1990 to 2017 (EAPC = −1.52; 95% CI = −1.59-−1.44; Table 1 and Figure 3A). The most pronounced decreasing percentages in the absolute number of DALYs were noted in children and adolescents (Supplementary Table S1 and Figure 3B). The trends of DALYs generally decreased in both sexes, SDI areas, and geographic regions, particularly in East Asia (EAPC = −2.85; 95% CI = −3.13-−2.57; Table 1 and Figure 3A). Of the 195 countries/territories, 30 countries/territories showed increasing trends in ASR of DALYs from 1990 to 2017, and Jamaica had the largest increase (EAPC = 1.31; 95% CI = 0.87-1.75). 148 countries/territories had decreasing trends in the ASR of DALYs, including Bahrain and China, whose EAPCs were −2.99 (95% CI = −3.41-−2.57) and −2.93 (95% CI = −3.22-−2.64), respectively (Supplementary Table S2 and Figure 4A–C). The trends of death and DALYs due to leukemia were negatively correlated with HDI at a national level (ρ = −0.36 and P = 0.003 for trends of death; ρ = −0.47 and P < 0.001 for trends of DALYs; Figure 5A–B), but not with the ASR in 1990. Similar correlations were also seen in the four main subtypes of leukemia. Overall, the decreasing trends of death and DALYs due to leukemia and its subtypes generally occurred in the countries with high HDI, particularly the CML and CLL subtypes, while increasing trends were more common in countries with low HDI.

3.2 Trends in death and DALYs due to acute myeloid leukemia

Globally, the number of AML deaths increased to 92.98% from 1990 to 2017, and was 99.90 × 10³ (95% UI = 91.28 × 10³-104.58 × 10³) in 2017. The ASDR of AML had an increasing trend from 1990 to 2017 (EAPC = 0.45; 95% CI = 0.37-0.53; Supplementary Table S3 and Figure 1A, C, and D). The percentage changes of the number of deaths remarkably increased in older age groups (> 65 years old) (Supplementary Table S1 and Figure 1B). The increasing trends of ASDR were observed in most SDI areas and geographic regions (Supplementary Table S3 and Figure 1A). The trends of ASDR increased in 132 countries/territories from 1990 to 2017, with the largest increase observed in El Salvador (EAPC = 3.62; 95% CI = 2.96-4.28), followed by Ecuador and Dominican Republic. Thirty-seven countries/territories had decreasing trends of ASDR, particularly observed in Bahrain and Qatar (EAPC were −3.17, 95% CI: −3.65-−2.69, and −2.55; 95% CI = −3.06-−2.03, respectively; Supplementary Table S4 and Figure 6A).

The number of DALYs due to AML was 3221.47 × 10³ (95% UI = 2890.15 × 10³-3438.43 × 10³) in 2017, with a 56.14% increase from 1990 to 2017. The ASR of DALYs had an increasing trend from 1990 to 2017 (EAPC = 0.13; 95% CI = 0.07-0.18; Supplementary Table S5 and Figure 3A, C and D). Based on age groups, the percentages of the absolute number of DALYs significantly increased in older patients (> 65 years old) (Supplementary Table S1 and Figure 3B). The ASR of DALYs showed decreasing and increasing trends in female and male patients, respectively. The trends of DALYs increased in low- and middle-SDI areas and regions (Supplementary Table S5 and Figure 3A). The increasing trends of ASR were observed in 113 countries/territories from 1990 to 2017, and the largest increases were seen in Ecuador and El Salvador, with the respective EAPCs being 3.19 (95% CI = 2.82-3.55) and 3.09 (95% CI = 2.46-3.73). Decreasing trends in the ASR of DALYs were seen in fifty-nine countries/territories, with the largest decrease observed in Bahrain (EAPC = −3.01; 95% CI = −3.44-−2.57); Supplementary Table S6 and Figure 7A).

3.3 Trends in death and DALYs due to chronic myeloid leukemia

Globally, the number of deaths caused by CML was 24.05 × 10³ (95% UI = 22.23 × 10³-26.07 × 10³) in 2017, with a 0.59% decrease from 1990 to 2017. The ASDR of CML decreased at an annual average of 2.76% from 1990 to 2017 (EAPC = −2.76; 95% CI = −2.88-−2.64; Supplementary Table S3 and Figure 1A, C, and D). The percentages in absolute death number of CML decreased in all age groups (Supplementary Table S1 and Figure 1B). The trends of ASDR dramatically decreased in both sexes, SDI areas, and geographic regions, particularly in high SDI areas and the high-income Asia Pacific regions, with the respective EAPCs of −4.58 (95% CI = −4.79-−4.37) and −5.12 (95% CI = −5.30-−4.95; Supplementary Table S3 and Figure 1A). The decreasing trends of ASDR were observed in 164 countries/territories, with the largest decreases seen in Japan and Finland with EAPCs of −5.78 (95% CI = −6.11-−5.45) and −5.62 (95% CI = −6.01-−5.23). Twenty-two countries/territories, including Jamaica and Zimbabwe, had increasing trends of ASDR (Supplementary Table S4 and Figure 6B).

Globally, the number of DALYs due to CML decreased by 10.88% from 1990 to 2017, and was 654.98 × 10³ (95% UI = 594.73 × 10³-712.95 × 10³) in 2017. The ASR of DALYs had a decreasing trend from 1990 to 2017, with an EAPC of −2.84 (95% CI = −2.97-−2.70; Supplementary Table S5 and Figure 3A, C and D). The trends in ASR of DALYs decreased in both sexes, SDI areas, and most geographic regions, and the most significant one was in the high-income Asia Pacific region (EAPC = −5.94; 95% CI = −6.16-−5.73; Supplementary Table S5 and Figure 3A). The ASR of DALYs showed decreasing trends in 171 countries/territories from 1990 to 2017, with the largest decreases seen in Japan and United Kingdom, with EAPCs of −6.69 (95% CI = −7.07-−6.30) and −6.04 (95% CI = −6.52-−5.56). The increasing trends in the ASR of DALYs occurred in fourteen countries/territories, particularly Jamaica (EAPC = 2.60; 95% CI = 2.22-2.99); Supplementary Table S6 and Figure 7B).

3.4 Trends in death and DALYs due to acute lymphoblastic leukemia

Globally, the number of deaths due to ALL was 52.22 × 10³ (95% UI = 45.97 × 10³-56.73 × 10³) in 2017, with an increase of 40.12% from 1990 to 2017. The EAPCs of ASDR due to ALL demonstrated no significant change during the period 1990–2017 (Supplementary Table S3 and Figure 1A, C, and D). The percentages in absolute death numbers of ALL decreased in the age groups of < 25 years while increasing in other groups (Supplementary Table S1 and Figure 1B). The ASDR showed remarkably decreasing trends in high SDI areas and high-income geographic regions, with the largest increasing trends observed in Andean Latin America and East Asia (Supplementary Table S3 and Figure 1A). The decreasing trends of ASDR occurred in 108 countries/territories from 1990 to 2017, and the largest decrease was found in Ghana and the Czech Republic, with EAPCs of−4.78 (95% CI = −5.71-−3.83) and −4.44 (95% CI = −4.98-−3.90). Sixty-five countries/territories had increasing trends of ASDR, particularly El Salvador and Guatemala (Supplementary Table S4 and Figure 6C).

The global number of DALYs due to ALL increased by 18.56% from 1990 to 2017, and was 2704.69 × 10³ (95% UI = 2375.06 × 10³-2985.23 × 10³) in 2017. The changes in ASR of DALYs had a decreasing trend globally from 1990 to 2017 (EAPC = −0.20, 95%CI:−0.27-−0.13) (Supplementary Table S5 and Figure 3A, C, and D). The ASR of DALYs showed significant decreasing trends in high SDI areas and high-income geographic regions (Supplementary Table S5 and Figure 3A). Forty-nine countries/territories had increasing trends of DALYs from 1990 to 2017, with the largest increases seen in El Salvador and Guatemala, with EAPCs of 4.92 (95% CI = 4.04-5.80) and 4.82 (95% CI = 4.53-5.11). Decreasing trends in the ASR of DALYs occurred in 115 countries/territories, and the most pronounced one was in Ghana (EAPC = −5.63; 95% CI = −6.73-−4.52); Supplementary Table S6 and Figure 7C).

3.5 Trends in chronic lymphocytic leukemia

Globally, the number of deaths due to CLL increased by 70.90% from 1990 to 2017, and was 35.17 × 10³ (95% UI = 33.54 × 10³-36.88 × 10³) in 2017. The ASDR decreased by an average of 0.86% per year from 1990 to 2017 (EAPC = −0.86; 95% CI = −0.94-−0.78) (Supplementary Table S3 and Figure 1A, C, and D). The percentages in absolute death number of CLL decreased in most age groups (Supplementary Table S1 and Figure 1B). The trends in ASDR decreased in the high SDI areas and high-income regions, while increasing in Central Sub-Saharan Africa, East Asia, and Southeast Asia (Supplementary Table S3 and Figure 1A). The increasing trends of ASDR were seen in 116 countries/territories from 1990 to 2017, with the largest increases seen in Jamaica and Georgia, with respective EAPCs of 6.80 (95% CI = 5.92-7.69) and 4.33 (95% CI = 2.98-5.70). The trends of ASDR decreased in fifty-two countries/territories, with the largest decrease seen in the Netherlands (EAPC = −5.39; 95% CI = −6.70-−4.06; Supplementary Table S4 and Figure 6D).

The global number of DALYs due to CLL increased by 56.82% from 1990 to 2017, and was 708.68 × 10³ (95% UI = 666.84 × 10³-755.09 × 10³) in 2017. The ASR of DALYs showed a decreasing trend (EAPC = −0.90; 95% CI = −0.96-−0.84; Supplementary Table S5 and Figure 3A, C, and D). The percentages in absolute number of deaths caused by CLL decreased in most age groups (Supplementary Table S1 and Figure 3B). The trends in ASR of DALYs decreased in high SDI areas and high-income regions, while increasing in low SDI areas and regions (Supplementary Table S5 and Figure 3A). The increasing trends of DALYs were recorded in 115 countries/territories, with the largest increase noted in Jamaica, with EAPC of 6.48 (95% CI = 5.64-7.33). The pronounced decreasing trend in the ASR of DALYs was observed in the Netherlands (EAPC = −5.18; 95% CI = −6.41-−3.93; Supplementary Table S6 and Figure 7D).

ETHICS APPROVAL AND CONSENT TO PARTICIPATE

Not applicable.

CONSENT FOR PUBLICATION

Not applicable.

AVAILABILITY OF DATA AND MATERIALS

The data that support the findings of this study are available from the corresponding author upon reasonable request.

COMPETING INTERESTS

The authors declare that they have no competing interests.

FUNDING

Not applicable.

AUTHORS' CONTRIBUTIONS

ZJO: project administration and drafting.

DFY: Data analysis and validation.

YHL: Data analysis and visualization.

WQH, YZL, MYZ, FFY, HH: Data collection and collation.

QC: supervision and drafting and editing.

All authors read and approved the final manuscript.