Benchmarking progress in non-communicable diseases: a global analysis of cause-specific mortality from 2001 to 2019

Abstract

Background: Non-communicable diseases (NCDs) have received substantial policy attention globally and in most countries. Our aim was to quantify how much NCD mortality changed from 2010 to 2019 in different countries, especially compared with the preceding decade and with the best-performing country in each region, and the specific NCD causes of death that contributed to change.

Methods: We used data on NCD mortality by sex, age group, and underlying cause of death for 185 countries and territories from the 2021 WHO Global Health Estimates. Our primary outcome was the probability of dying from an NCD between birth and age 80 years in the absence of competing causes of death, and was calculated using age-specific death rates from NCDs and lifetable methods. We calculated change in the probability of death as the difference between values in the final and first year of each period (2001-10 and 2010-19). For 51 countries with high-quality mortality data and 12 countries with large populations within their region, we used the Horiuchi method of decomposition to calculate how much specific causes of death and 5-year age groups contributed towards: (1) increases or decreases in NCD mortality from 2010 to 2019; (2) improvements or deteriorations compared with the preceding decade (2001-10); and (3) differences from the country that had the largest reduction in each region.

Findings: From 2010 to 2019, the probability of dying from an NCD between birth and age 80 years decreased in 152 (82%) of 185 countries for females and in 147 (79%) countries for males; it increased in the remaining 33 (18%) countries for females and 38 (21%) countries for males. The countries where NCD mortality declined for females accounted for 72% of the world female population in 2019, and those where NCD mortality declined for males accounted for 73% of the world male population. NCD mortality declined in all high-income western countries, with Denmark experiencing the largest decline for both sexes and the USA experiencing the smallest decline. Among the largest countries in other regions, NCD mortality declined for both sexes in China, Egypt, Nigeria, Russia, and Brazil, and increased for both sexes in India and Papua New Guinea. On average, females in countries in the central Asia, Middle East and north Africa region had the greatest reduction in NCD mortality followed by those in central and eastern Europe. For males, the largest reduction was among countries in central and eastern Europe, followed by those in central Asia, Middle East and north Africa. The smallest declines were those in the Pacific Island nations. Circulatory diseases were the greatest contributors to declines in NCD mortality from 2010 to 2019 in most countries, with some cancers (eg, stomach and colorectal cancers for both sexes, cervical and breast cancers for females, and lung and prostate cancers for males) also contributing towards lower NCD mortality in 2019 than in 2010 in many countries. Neuropsychiatric conditions and pancreatic and liver cancers contributed towards higher NCD mortality from 2010 to 2019 in most countries. In some countries, NCD mortality in working and older (≥65 years) ages changed in the same direction leading to large overall declines or increases; in others, it changed in opposite directions, diminishing the magnitude of the overall change. In 75 (41%) of 185 countries for females and in 73 (39%) countries for males, the change in NCD mortality from 2010 to 2019 was an improvement (ie, larger decline, smaller increase, or reversal of an increase) compared with the change from 2001 to 2010. These countries accounted for 29% and 63% of the world female and male population, respectively, and included both sexes in Russia and Egypt, and males in China, India, and Brazil. Decadal changes saw a deterioration (ie, smaller decline, larger increase, or reversal of a decline) in the remaining 110 (59%) countries for females and 112 (61%) countries for males, including in both sexes in the USA, Nigeria, and Papua New Guinea, and females in China, India, and Brazil. Change from 2010 to 2019 saw deterioration in direction or size compared with the preceding decade for both sexes in most high-income western countries, most countries in Latin America and the Caribbean, and in east and southeast Asia, and for females in south Asia. There was a decadal improvement in the direction or size of change for many countries in central and eastern Europe (eg, Russia) and central Asia, and in parts of the Middle East and north Africa. Improvements or deteriorations in the direction or size of change in NCD mortality between the two decades resulted from multiple NCD causes of death. Among causes of death, the decline in mortality from circulatory diseases was smaller from 2010 to 2019 than from 2001 to 2010 in most countries, except in countries in central and eastern Europe and some countries in central Asia, where these declines were larger from 2010 to 2019 than from 2001 to 2010. Change in lung cancer saw a decadal improvement in many countries, especially for males, and many other cancers saw a mix of improvement and deterioration.

Interpretation: From 2010 to 2019, NCD mortality declined in four of every five countries in the world. These improvements were not as large as the preceding decade for most countries, driven by smaller declines in mortality from multiple NCDs.

Funding: UK Medical Research Council, UK National Institute for Health and Care Research, and NCD Alliance.

Figure 1. Countries showing improvements in the direction or size of change from 2010 to 2019 compared with from 2001 to 2010 for females

Estimates are shown for all countries and territories in eight reporting regions. Countries are divided into three categories of changes observed across two timeframes (2001–10 and 2010–19). Within each of the three categories, countries are ordered by percentage point change from 2010 to 2019 (ie, the first country listed in each category had the largest decrease or smallest increase from 2010 to 2019 within its category). For the estimated probabilities and change in probabilities with uncertainty intervals see the appendix (pp 44–53). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 13–14). NCD=non-communicable disease.

Figure 2. Countries showing improvements in the direction or size of change from 2010 to 2019 compared with from 2001 to 2010 for males

Estimates are shown for all countries and territories in eight reporting regions. Countries are divided into three categories of changes observed across two timeframes (2001–10 and 2010–19). Within each of the three categories, countries are ordered by percentage point change from 2010 to 2019 (ie, the first country listed in each category had the largest decrease or smallest increase from 2010 to 2019 within its category). For the estimated probabilities and change in probabilities with uncertainty intervals see the appendix (pp 44–53). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 13–14). NCD=non-communicable disease.

Figure 3. Countries showing deteriorations in the direction or size of change from 2010 to 2019 compared with from 2001 to 2010 for females

Estimates are shown for all countries and territories in eight reporting regions. Countries are divided into three categories of changes observed across two timeframes (2001–10 and 2010–19). Within each of the three categories, countries are ordered by percentage point change from 2010 to 2019 (ie, the first country listed in each category had the largest decrease or smallest increase from 2010 to 2019 within its category). For the estimated probabilities and change in probabilities with uncertainty intervals see the appendix (pp 44–53). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 13–14). NCD=non-communicable disease.

Figure 3. Countries showing deteriorations in the direction or size of change from 2010 to 2019 compared with from 2001 to 2010 for females

Estimates are shown for all countries and territories in eight reporting regions. Countries are divided into three categories of changes observed across two timeframes (2001–10 and 2010–19). Within each of the three categories, countries are ordered by percentage point change from 2010 to 2019 (ie, the first country listed in each category had the largest decrease or smallest increase from 2010 to 2019 within its category). For the estimated probabilities and change in probabilities with uncertainty intervals see the appendix (pp 44–53). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 13–14). NCD=non-communicable disease.

Figure 4. Countries showing deteriorations in the direction or size of change from 2010 to 2019 compared with from 2001 to 2010 for males

Estimates are shown for all countries and territories in eight reporting regions. Countries are divided into three categories of changes observed across two timeframes (2001–10 and 2010–19). Within each of the three categories, countries are ordered by percentage point change from 2010 to 2019 (ie, the first country listed in each category had the largest decrease or smallest increase from 2010 to 2019 within its category). For the estimated probabilities and change in probabilities with uncertainty intervals see the appendix (pp 44–53). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 13–14). NCD=non-communicable disease.

Figure 4. Countries showing deteriorations in the direction or size of change from 2010 to 2019 compared with from 2001 to 2010 for males

Estimates are shown for all countries and territories in eight reporting regions. Countries are divided into three categories of changes observed across two timeframes (2001–10 and 2010–19). Within each of the three categories, countries are ordered by percentage point change from 2010 to 2019 (ie, the first country listed in each category had the largest decrease or smallest increase from 2010 to 2019 within its category). For the estimated probabilities and change in probabilities with uncertainty intervals see the appendix (pp 44–53). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 13–14). NCD=non-communicable disease.

Figure 5. Probability of dying from an NCD between birth and age 80 years in 2019 and change in probability from 2010 to 2019

For change from 2010 to 2019, green indicates a decline in NCD mortality and red indicates an increase. The density plot alongside each map shows the smoothed distribution of estimates across countries. Countries and territories with no mortality estimates are shown in grey. For the estimated probabilities and change in probabilities with uncertainty intervals see the appendix (pp 44–53). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 54–55). NCD=non-communicable disease.

Figure 6. Change in NCD mortality from 2001 to 2019

Each line connects the probability of dying between birth and age 80 years from an NCD for 2001, 2010, and 2019 for one country. For each country, the difference in level between consecutive pairs of years respectively represents change over the intervals from 2001 to 2010 and from 2010 to 2019. Data are shown for 185 countries and territories, divided into eight reporting regions. Lines are coloured by region and labelled with ISO3 codes for each country. The bold black line in each panel connects the mean levels (across countries in that panel, unweighted for population) in 2001, 2010, and 2019. Regions are ordered by increasing mean probability of dying of the countries in each region for females in 2001. For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 56–57). NCD=non-communicable disease. High-income western countries: AUS=Australia, AUT=Austria, BEL=Belgium, CAN=Canada, CHE=Switzerland, CYP=Cyprus, DEU=Germany, DNK=Denmark, ESP=Spain, FIN=Finland, FRA=France, GBR=United Kingdom, GRC=Greece, IRL=Ireland, ISL=Iceland, ISR=Israel, ITA=Italy, LUX=Luxembourg, MLT=Malta, NLD=Netherlands, NOR=Norway, NZL=New Zealand, PRT=Portugal, SWE=Sweden, USA=United States of America. Latin America and the Caribbean: ARG=Argentina, ATG=Antigua and Barbuda, BHS=The Bahamas, BLZ=Belize, BOL=Bolivia, BRA=Brazil, BRB=Barbados, CHL=Chile, COL=Colombia, CRI=Costa Rica, CUB=Cuba, DOM=Dominican Republic, ECU=Ecuador, GRD=Grenada, GTM=Guatemala, GUY=Guyana, HND=Honduras, HTI=Haiti, JAM=Jamaica, LCA=Saint Lucia, MEX=Mexico, NIC=Nicaragua, PAN=Panama, PER=Peru, PRI=Puerto Rico, PRY=Paraguay, SLV=El Salvador, SUR=Suriname, TTO=Trinidad and Tobago, URY=Uruguay, VCT=Saint Vincent and the Grenadines, VEN=Venezuela. East and southeast Asia: BRN=Brunei, CHN=China, IDN=Indonesia, JPN=Japan, KHM=Cambodia, KOR=South Korea, LAO=Laos, MMR=Myanmar, MYS=Malaysia, PHL=Philippines, PRK=North Korea, SGP=Singapore, THA=Thailand, TLS=Timor-Leste, TWN=Taiwan, VNM=Viet Nam. Central and eastern Europe: ALB=Albania, BGR=Bulgaria, BIH=Bosnia and Herzegovina, BLR=Belarus, CZE=Czechia, EST=Estonia, HRV=Croatia, HUN=Hungary, LTU=Lithuania, LVA=Latvia, MDA=Moldova, MKD=North Macedonia, MNE=Montenegro, POL=Poland, ROU=Romania, RUS=Russia, SRB=Serbia, SVK=Slovakia, SVN=Slovenia, UKR=Ukraine. South Asia: AFG=Afghanistan, BGD=Bangladesh, BTN=Bhutan, IND=India, LKA=Sri Lanka, NPL=Nepal, PAK=Pakistan. Central Asia, Middle East and north Africa: ARE=United Arab Emirates, ARM=Armenia, AZE=Azerbaijan, BHR=Bahrain, DZA=Algeria, EGY=Egypt, GEO=Georgia, IRN=Iran, IRQ=Iraq, JOR=Jordan, KAZ=Kazakhstan, KGZ=Kyrgyzstan, KWT=Kuwait, LBN=Lebanon, LBY=Libya, MAR=Morocco, MNG=Mongolia, OMN=Oman, PSE=Palestine, QAT=Qatar, SAU=Saudi Arabia, SYR=Syria, TJK=Tajikistan, TKM=Turkmenistan, TUN=Tunisia, TUR=Türkiye, UZB=Uzbekistan, YEM=Yemen. Sub-Saharan Africa: AGO=Angola, BDI=Burundi, BEN=Benin, BFA=Burkina Faso, BWA=Botswana, CAF=Central African Republic, CIV=Côte d’Ivoire, CMR=Cameroon, COD=DR Congo, COG=Congo (Brazzaville), COM=Comoros, CPV=Cabo Verde, DJI=Djibouti, ERI=Eritrea, ETH=Ethiopia, GAB=Gabon, GHA=Ghana, GIN=Guinea, GMB=The Gambia, GNB=Guinea-Bissau, GNQ=Equatorial Guinea, KEN=Kenya, LBR=Liberia, LSO=Lesotho, MDG=Madagascar, MLI=Mali, MOZ=Mozambique, MRT=Mauritania, MUS=Mauritius, MWI=Malawi, NAM=Namibia, NER=Niger, NGA=Nigeria, RWA=Rwanda, SDN=Sudan, SEN=Senegal, SLE=Sierra Leone, SOM=Somalia, SSD=South Sudan, STP=São Tomé and Príncipe, SWZ=Eswatini, SYC=Seychelles, TCD=Chad, TGO=Togo, TZA=Tanzania, UGA=Uganda, ZAF=South Africa, ZMB=Zambia, ZWE=Zimbabwe. Pacific Island nations: FJI=Fiji, FSM=Federated States of Micronesia, KIR=Kiribati, PNG=Papua New Guinea, SLB=Solomon Islands, TON=Tonga, VUT=Vanuatu, WSM=Samoa.

Figure 7. Contributions of mortality from different NCD causes of death and in different age groups to overall change in NCD mortality from 2010 to 2019

(A) The contribution of 20 mutually exclusive, collectively exhaustive NCD causes of death to the change in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019. Each column represents a cause of death, as stated in the appendix (pp 11–12), with causes arranged by disease category. Results are presented for aggregated cause groups in the appendix (pp 35–36). (B) The contribution of 5-year age groups to the change in this probability over the same period, with each column representing a 5-year age group. In both panels, each row represents a country. Results are shown for 63 countries, of which 51 were identified as having high-quality data and 12 were selected based on population size. Countries are grouped and coloured by region and ordered from the largest decrease to the smallest decrease or largest increase in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019. Each tile shows the absolute contribution of an NCD cause of death or age group to the total change in this probability for one country. Two colour palettes are used: one for the overall change in NCD mortality from 2010 to 2019, and one for contributions of individual NCD causes of death or age groups. For overall change, black indicates a decrease, purple an increase, and white no change. For contributions to change, green indicates a contribution to lowering NCD mortality, red a contribution to increasing it, and white a contribution of zero. For numerical results see the appendix (pp 26–28). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 58–60). NCD=non-communicable disease. *12 countries selected based on population size.

Figure 7. Contributions of mortality from different NCD causes of death and in different age groups to overall change in NCD mortality from 2010 to 2019

(A) The contribution of 20 mutually exclusive, collectively exhaustive NCD causes of death to the change in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019. Each column represents a cause of death, as stated in the appendix (pp 11–12), with causes arranged by disease category. Results are presented for aggregated cause groups in the appendix (pp 35–36). (B) The contribution of 5-year age groups to the change in this probability over the same period, with each column representing a 5-year age group. In both panels, each row represents a country. Results are shown for 63 countries, of which 51 were identified as having high-quality data and 12 were selected based on population size. Countries are grouped and coloured by region and ordered from the largest decrease to the smallest decrease or largest increase in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019. Each tile shows the absolute contribution of an NCD cause of death or age group to the total change in this probability for one country. Two colour palettes are used: one for the overall change in NCD mortality from 2010 to 2019, and one for contributions of individual NCD causes of death or age groups. For overall change, black indicates a decrease, purple an increase, and white no change. For contributions to change, green indicates a contribution to lowering NCD mortality, red a contribution to increasing it, and white a contribution of zero. For numerical results see the appendix (pp 26–28). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 58–60). NCD=non-communicable disease. *12 countries selected based on population size.

Figure 8. Contributions of mortality from different NCD causes of death and in different age groups to slowdown or acceleration of change in NCD mortality from 2010 to 2019 compared to change from 2001 to 2010

(A) The contribution of 20 mutually exclusive, collectively exhaustive NCD causes of death to the difference in change in the probability of dying from an NCD between birth and age 80 years between two decades (2010–19 and 2001–10). Each column represents a cause of death, as stated in the appendix (pp 11–12), with causes arranged by disease category. Results are presented for aggregated cause groups in the appendix (pp 37–38). (B) The contribution of 5-year age groups to this decadal difference in change, with each column representing a 5-year age group. In both panels, each row represents a country. Results are shown for 63 countries, of which 51 were identified as having high-quality data and 12 were selected based on population size. Countries are grouped and coloured by region and ordered from the largest improvement in NCD mortality from 2010 to 2019 compared with the preceding decade to the largest deterioration. Each tile shows the absolute contribution of a specific NCD cause of death or age group to the decadal difference in change in NCD mortality (ie, difference in change between the two decades) for one country. Two colour palettes are used: one for the overall decadal difference in change, and one for contributions of individual NCD causes of death or age groups. For overall decadal difference in change, yellow indicates improvement (a larger decline, smaller increase, or reversal of an increase), pink indicates deterioration (a smaller decline, reversal of a decline, or a larger increase), and white indicates no difference in the magnitude of change. For contributions to decadal difference in change, green indicates a contribution to improvement of change (a larger decline, smaller increase, or reversal of an increase), brown indicates a contribution to deterioration of change (a smaller decline, reversal of a decline, or a larger increase), and white indicates a contribution of zero. For numerical results see the appendix (pp 29–31). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 61–63). NCD=non-communicable disease. *12 countries selected based on population size.

Figure 8. Contributions of mortality from different NCD causes of death and in different age groups to slowdown or acceleration of change in NCD mortality from 2010 to 2019 compared to change from 2001 to 2010

(A) The contribution of 20 mutually exclusive, collectively exhaustive NCD causes of death to the difference in change in the probability of dying from an NCD between birth and age 80 years between two decades (2010–19 and 2001–10). Each column represents a cause of death, as stated in the appendix (pp 11–12), with causes arranged by disease category. Results are presented for aggregated cause groups in the appendix (pp 37–38). (B) The contribution of 5-year age groups to this decadal difference in change, with each column representing a 5-year age group. In both panels, each row represents a country. Results are shown for 63 countries, of which 51 were identified as having high-quality data and 12 were selected based on population size. Countries are grouped and coloured by region and ordered from the largest improvement in NCD mortality from 2010 to 2019 compared with the preceding decade to the largest deterioration. Each tile shows the absolute contribution of a specific NCD cause of death or age group to the decadal difference in change in NCD mortality (ie, difference in change between the two decades) for one country. Two colour palettes are used: one for the overall decadal difference in change, and one for contributions of individual NCD causes of death or age groups. For overall decadal difference in change, yellow indicates improvement (a larger decline, smaller increase, or reversal of an increase), pink indicates deterioration (a smaller decline, reversal of a decline, or a larger increase), and white indicates no difference in the magnitude of change. For contributions to decadal difference in change, green indicates a contribution to improvement of change (a larger decline, smaller increase, or reversal of an increase), brown indicates a contribution to deterioration of change (a smaller decline, reversal of a decline, or a larger increase), and white indicates a contribution of zero. For numerical results see the appendix (pp 29–31). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 61–63). NCD=non-communicable disease. *12 countries selected based on population size.

Figure 9. Contributions of mortality from different NCD causes of death and in different age groups to how much NCD mortality in each country lags its regional benchmark

(A) The contribution of 20 mutually exclusive, collectively exhaustive NCD causes of death to the difference in the change in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019, compared with a country benchmark within each region. Benchmarks are identified as the country in each region with the largest reduction in NCD mortality over this period. Each column represents a cause of death, as stated in the appendix (pp 11–12), with causes arranged by disease category. Results are presented for aggregated cause groups in the appendix (pp 39–40). (B) The contribution of 5-year age groups to this difference, with each column representing a 5-year age group. In both panels, each row represents a country. Results are shown for 51 countries identified as having high-quality data. Countries are grouped and coloured by region and ordered from the largest decrease to the smallest decrease or largest increase in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019. The benchmark for each region is the country in the first row of its region grouping and is shown in bold font. Each tile shows the absolute contribution of a specific NCD cause of death or age group to the difference in change compared with the benchmark country for one country. Two colour palettes are used: one for the overall difference in change compared with the benchmark, and one for contributions of individual NCD causes of death or age groups compared with those of the benchmark. For overall difference in change, black indicates a decrease compared with the benchmark, purple indicates an increase compared with the benchmark, and white indicates no difference in change. For contributions to difference in change, green indicates a contribution towards a larger decline or smaller increase compared with the benchmark, red indicates a contribution towards a smaller decline or larger increase compared with the benchmark, and white indicates a contribution of zero. For numerical results see the appendix (pp 32–34). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 64–66). NCD=non-communicable disease.

Figure 9. Contributions of mortality from different NCD causes of death and in different age groups to how much NCD mortality in each country lags its regional benchmark

(A) The contribution of 20 mutually exclusive, collectively exhaustive NCD causes of death to the difference in the change in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019, compared with a country benchmark within each region. Benchmarks are identified as the country in each region with the largest reduction in NCD mortality over this period. Each column represents a cause of death, as stated in the appendix (pp 11–12), with causes arranged by disease category. Results are presented for aggregated cause groups in the appendix (pp 39–40). (B) The contribution of 5-year age groups to this difference, with each column representing a 5-year age group. In both panels, each row represents a country. Results are shown for 51 countries identified as having high-quality data. Countries are grouped and coloured by region and ordered from the largest decrease to the smallest decrease or largest increase in the probability of dying from an NCD between birth and age 80 years from 2010 to 2019. The benchmark for each region is the country in the first row of its region grouping and is shown in bold font. Each tile shows the absolute contribution of a specific NCD cause of death or age group to the difference in change compared with the benchmark country for one country. Two colour palettes are used: one for the overall difference in change compared with the benchmark, and one for contributions of individual NCD causes of death or age groups compared with those of the benchmark. For overall difference in change, black indicates a decrease compared with the benchmark, purple indicates an increase compared with the benchmark, and white indicates no difference in change. For contributions to difference in change, green indicates a contribution towards a larger decline or smaller increase compared with the benchmark, red indicates a contribution towards a smaller decline or larger increase compared with the benchmark, and white indicates a contribution of zero. For numerical results see the appendix (pp 32–34). For results based on cancers, cardiovascular diseases, chronic respiratory diseases, and diabetes in ages 30–70 years see the appendix (pp 64–66). NCD=non-communicable disease.