Burden of Disease

OWID presents work for many different people and organizations. When citing this entry, please also cite the original data source. This entry can be cited as:

Max Roser and Hannah Ritchie (2018) - "Burden of Disease". Published online at OurWorldInData.org. Retrieved from: 'https://ourworldindata.org/burden-of-disease' [Online Resource]

The burden of disease can be measured in Disability Adjusted Life Years (DALYs) lost. DALYs are a standardized metric that allow for direct comparison of burdens of different diseases across countries and over time. Conceptually, one DALY lost is the equivalent of losing one year in good health because of either premature mortality or disability. Assessing health outcomes by both mortality and morbidity provides a more encompassing view on health outcomes than only looking at mortality or life expectancy alone.

This entry presents data on burden of health across the world, breakdown by age, types of disability and disease, and regional/country breakdowns. The visualisations which follow can be explored by any country or region using the "Change country" option in the charts below.

I. Empirical View

I.1 The global burden of disease

The global burden of disease is large. In the charts below we see total burden of disease rates, measured as the number of Disability Adjusted Life Years (DALYs) lost per 100,000 individuals. This is shown below as the sum of disease burden from all causes.

Overall we see that rates across the world vary from 40,000-70,000 DALYs per 100,000 individuals across high-burden countries, particularly in Sub-Saharan Africa. Disease burden across most of Europe, the Americas and the Middle East, North Africa & Central Asia tends towards the range of 10,000-30,000 DALYs per 100,000.

As shown in the charts below is the total global DALYs lost, disaggregated by region. At a global level, health burden in 2016 was over 2.3 billion DALYs lost. In absolute terms, the largest burden is in South Asia, with more than one-quarter of total disease burden. Burden in Sub-Saharan Africa is also high with more than 20 percent of global figures.

Disease burden by age

In the chart below we see the breakdown of total disease burden by age group from 1990 onwards. This is shown as the total disease burden as well as the rates of burden per 100,000 individuals within a given age group.

Overall we see a continued decline in health burden in children under 5 years old; both in absolute terms (falling as a share of the total by more than half, from 40 to below 20 percent since 1990), and in rates per 100,000 (falling more than 50 percent from over 160,000 to almost 70,000).

Nonetheless, rates of disease burden remain highest in the youngest and oldest in society. DALY loss rates in under-5s and those over 70 years old remain significantly higher than other age groups. They have, however, seen the most notable declines in recent decades.

At a global level, collective rates across all ages have been in steady decline.

Disease burden by cause

Disease burden can be broken down into three key categories of disability or disease: non-communicable diseases (NCDs); communicable, maternal, neonatal and nutritional diseases, and injuries. We provide a more detailed breakdown of what sub-categories fall within each of these three groupings in our Data Quality and Definitions section. We also look at a higher-resolution breakdown within each of these groupings in the sections which follow.

The chart below shows the total disease burden disaggregated across NCDs, communicable disease and injuries. At a global level, in 2016 more than 60 percent of health burden results from non-communicable diseases (NCDs), with 28 percent from communicable, maternal, neonatal and nutritional diseases, and just over 10 percent from injuries. This represents a notable shift since 1990, where communicable diseases held the highest share at more than 45 percent.

This shift in burden towards NCDs result from a significant reduction in communicable and preventable disease as incomes rise, overall health and living standards improve. In high-income nations, NCDs typically account for more than 80 percent of disease burden. In contrast, communicable diseases to be low, at less than 5 percent. The opposite is true in low-income nations; communicable disease still accounts for more than 60 percent across many countries.

In the charts below we see a higher-resolution breakdown of disease burden, in absolute numbers of DALYs lost, and each cause as a share of the total. Non-communicable diseases (NCDs) are shown in blue; communicable, maternal, neonatal and nutritional diseases shown in red; and injuries shown in grey. These figures and rankings of disease burden can be explored by country and region using the 'change country' option in the charts below.

At a global level the largest disease burden comes from cardiovascular diseases which account for 15-16 percent of the total. This is followed by diarrhea & common infectious disease (10 percent); cancers (9 percent); other NCDs (8 percent); newborn complications (7 percent); and mental and substance use disorders (7 percent).

This attribution varies significantly across the world, however. If we look at a typical lower-income country (e.g. Congo), we notice that the top four categories (accounting for more than half of total health burden) are all communicable and neonatal diseases, including diarrhea, HIV/AIDS and malaria. This is in stark contrast to a typical high-income nation (e.g. United States) where no communicable diseases fall within the top ten (which comprises of 8 NCD categories, followed by unintentional and road injuries). Cardiovascular disease, cancer and mental and substance use disorders form the top three health burdens across most upper-middle and high-income nations.

I.2 Non-communicable diseases (NCDs)

The visualisations below focus on the disease burden resultant from non-communicable diseases (NCDs), shown as the DALY rate per 100,000 from collective NCDs. Also shown is the total breakdown of NCD burden by specific sub-categories.

In the chart below we also see this breakdown of NCD burden by age category. Non-communicable diseases are typically low in children and adolescents; collectively less than 10 percent of the burden falls in those under 15 years old.

I.3 Communicable, neonatal, maternal and nutritional diseases

The visualisations below show the burden of disease from communicable, neonatal, maternal and nutritional diseases. First, this is shown in terms of DALY loss rates per 100,000 individuals. We see strong differentiation, with high burden across Sub-Saharan Africa and South Asia in particular. Most countries across these regions have DALY losses greater than 25,000 per 100,000 individuals, reaching over 50,000 in the Central African Republic. Rates in Europe and North America, in contrast, are typically greater than ten times lower, below 2500 per 100,000.

In the second chart below we see this burden broken down by specific causes. There has been a significant reduction in global burden from communicable diseases in recent decades, falling from over 1.1 billion in 1990 to below 670,000 in 2016 (around a 40 percent reduction). This category of health conditions are dominated by diarrheal and other infectious diseases, and neonatal disorders. Globally, these two causes alone account for more than 60% of communicable disease DALY losses.

The chart below shows the breakdown of communicable, neonatal, maternal and nutritional disease burden by age group. The burden in under-5s represents over half of losses (although this share continues to decline, falling from almost 75 percent in 1990).

I.4 Injuries (including violence and self-harm)

The category of injuries is broad and encompasses not only accidents (unintentional injuries such as falls, fire and drowning, as well as transport injuries), but also natural disasters and violence including interpersonal violence, conflict, terrorism and self-harm. See Data Quality and Definitions for a breakdown of these categories.

The charts below provide an overview of health burden from injuries. This is first shown in terms of DALY loss rates per 100,000, then total burden broken down by these various sub-categories.

Road accidents are particularly dominant within this category. However, interpersonal violence and self-harm also constitute a high share of health burden. You will notice that burden attributed to both conflict & terrorism and natural disasters are highly volatile (creating dramatic spikes from year-to-year. We discuss the impact of this volatility on overall trends in the context of death in our blog post here.

II. Correlates, Determinants, & Consequences

II.1 Income and disease burden

The visualisation below shows the relationship between average income – measured by GNI per capita – and the Burden of Disease. The Burden of Disease is disaggregated into the health burden due to communicable diseases and non-communicable diseases.

The chart shows that communicable diseases in particular are closely correlated to average income levels. The relationship that was estimated by Sterck et al. 20171 is shown in the legend. GNI per capita has a strong negative correlation with log DALYs lost due to communicable diseases with an elasticity of -0·88. On the other hand, the non-communicable disease burden is much less strongly associated with average income (the elasticity is estimated to be -0·13). Another conclusion we can draw from this chart is that the relationship between GNI per capita and DALYs lost due to the disease burden of communicable diseases is best captured by a log-log function.

II.2 Income and disease burden to communicable diseases

The health burden due to communicable diseases vs GDP per capita is shown in the following visualisations. The correlation between both measures is apparent: both DALY loss rates and the total share from communicable diseases tend to decline with increasing incomes. But despite this correlation, Sterck et al. 20172 find that GNI is not a significant predictor of health outcomes once other factors are controlled for. The first of these other factors is individual poverty – relative to a health poverty line of 10.89 international-$ per day. The second factor is the epidemiological surrounding of a country which captures the health status of neighbouring countries. And the third important factor is institutional capacity.

II.3 Income and disease burden to non-communicable diseases (NCDs)

The charts below highlight two important relationships between non-communicable disease burden and income. The first suggests that rates of burden from NCDs is highest at lower-incomes and tends to decline with development. DALYs lost from NCDs are typically lower at higher incomes.

However, it is also true (as shown in the second chart below) that NCDs constitute a dominant share of disease burden at higher incomes (often over 80 percent). The fact that both of these relationships are true: that NCD burden tends to decline with development, but increases its share of overall disease burden further highlights that total health burden declines significantly with improving living standards and healthcare.

The fact that NCD DALY losses at low-income are high, but still only constitute a small share of overall health burden emphasises the scale of DALY losses from communicable and preventable diseases which remain.

II.4 Disease burden and health expenditure

The visualisation below shows the relationship between total health burden, given as rates of DALY losses per 100,000 individuals (from all causes) versus average per capita health expenditure (in US dollars). At low levels of health expenditure we see a steep decline in health burden as per capita expenditure increases. However, towards mid-range health expenditure levels we begin to see a significant tailing off of burden reduction.

This diminishing rate of return stagnates at around 20,000 DALYs per 100,000 individuals. Nonetheless, per capita health expenditure at this level of health burden varies by several multiples. Countries such as the United States, Norway and Switzerland have a per capita expenditure over $9,000 per year, but have achieved little or no burden reduction relative to other high-income nations with a per capita expenditure less than half of these figures. Some countries - such as South Korea - have achieved one of the lowest rates of health burden with an expenditure of around $2,000 per capita.

III. Measurement, Data Quality & Definitions

Disability Adjusted Life Years (DALYs) lost is a standardized metric allowing for direct comparison and summing of burdens of different diseases. Conceptually, one DALY is the equivalent of one year in good health lost because of premature mortality or disability (see Murray et al. 20153). Assessing health outcomes by both mortality and morbidity provides a more encompassing view on health outcomes than only looking at mortality or life expectancy alone.

Three categories of health conditions and burdens are distinguished:

    1. Communicable, maternal, perinatal and nutritional diseases;
    2. Non-communicable diseases (NCDs);
    3. Injuries (which include violence and conflict).

The sub-categories of disease or health burden, as differentiated in the data provided in this entry from the Institute of Health Metrics and Evaluation (IHME) are detailed in the table below.

Communicable, maternal, neonatal and nutritional diseasesNon-communicable diseases (NCDs)Injuries
Diarrhea, lower respiratory & other common infectious diseasesCardiovascular diseases (inc. stroke, heart disease and heart failure)Road injuries
Neonatal disordersCancersOther transport injuries
Maternal disordersRespiratory diseaseFalls
Malaria & neglected tropical diseasesDiabetes, blood and endocrine diseasesDrowning
Nutritional deficienciesMental and substance use disordersFire, heat and hot substances
HIV/AIDSLiver diseasesPoisonings
TuberculosisDigestive diseasesSelf-harm
Other communicable diseasesMusculoskeletal disordersInterpersonal violence
Neurological disorders (including dementia)Conflict & terrorism
Other NCDsNatural disasters

IV. Data Sources

Institute for Health Metrics and Evaluation (IHME) – Burden of Disease study