The Perils of Perception 2020: Global study explores how wrong people are about the causes of death

Ipsos’ latest Perils of Perception study highlights public¹ misperceptions across 32 countries about the proportion of people who die from diseases, violence, transport injuries and other causes. While patterns differ in different countries, overall on average people tend to underestimate how many deaths are caused by cancers and cardiovascular disease, and overestimate how many are caused by transport injuries, substance misuse and violence.

Visit the Perils of Perception Microsite

Causes of death around the world: Diseases and infections

• Cardiovascular diseases are the leading cause of death globally according to the IHME Global Burden of Disease Study 2017 (IHME). The second biggest cause are cancers.
• The majority of countries in the study underestimate the percentage of people who die from cardiovascular diseases each year. The average guess across all countries was 11% when the actual figure was almost three times that at 32%. In Romania, 56% of all deaths are due to cardiovascular disease and the average guess was just 12%.
• Most countries underestimate the proportion of people who die from cancer each year. On average people think 15% of deaths each year are a result of cancer when the actual figure is 24%. In France, people think cancer accounts for 16% of deaths each year when in reality it is double that at 32%.
• Deaths resulting from neurological disorders such as dementia or Parkinson’s disease are also underestimated in many countries. The average guess across all 32 countries is 5% when the actual figure is 9%. People are particularly likely to underestimate this in Japan with an average guess of 5% compared with the actual figure of 16%.
• Nearly every country overestimates the proportion of deaths resulting from HIV/AIDS or sexually transmitted infections (STIs). The average guess is 5% when in reality the figure is 1%. In South Africa, HIV/AIDs and STIs cause 29% of deaths each year. But the average guess is 11%.

Causes of death around the world: Conflict, terrorism and violence

• Every country overestimates the proportion of people who die through interpersonal violence each year. The average ‘actual’ figure across all countries is just 1% when the average guess is 8%. People in the Americas are particularly likely to overestimate the proportion of people dying from interpersonal violence. In Peru the average guess is 14% when the actual figures is just 0.8%.
• Nearly every country in the study overestimate the proportion of people who die from terrorism or conflict each year. The average across all countries is just 0.1% when the average guess is 5%. In Colombia and Turkey people are particularly likely to overestimate this with guesses of around 10% when the actual figures are all less than 1%.

Causes of death around the world: Substance use disorders

• The proportion of deaths resulting from drug or alcohol addiction is overestimated in all countries. On average people think this accounts for 8% of all deaths when in reality it is 0.7%. In Italy, the average guess is 10% when the actual percentage of deaths is 0.5%.

Causes of death around the world: Suicide

• In every country, people overestimate the number of people who die by suicide. The average proportion of deaths by suicide for countries included in the study is 1.6% compared with an average guess of 7.3%. In Japan, the average guess is 10.9% when in reality the figure is 2.1%.

Causes of death around the world: Tranport injuries

• Transport injuries are overestimated in almost every country. The average guess is 10% when the actual figure is much lower (2%). Spain, Poland and Hungary are particularly like to overestimate this. In Spain, the average guess is 13% when transport injuries account for 0.7% of total deaths.

How accurate is the British public…?

• Brits think 12% of people die from cardiovascular diseases each year when in reality the figure is 29%.
• Cancer is the biggest cause of death in the UK but we underestimate its scale. The average guess is that 19% of people die from cancer each year when the actual figure is 30%.
• We underestimate the number of people who die from neurological disorders each year in the UK. The average guess is 7% when in reality it is 13%.
• Brits overestimate the proportion of people who die from HIV/AIDS or STIs each year. The average guess is 3% compared with the actual figure of 0.1%.
• Interpersonal violence such as murder accounts for 0.1% of deaths in the UK each year. Brits are more accurate than the majority of countries on this question but still overestimate with an average guess of 4.8%.
• We overestimate the proportion of people who die as a result of terrorism or conflict each year. The average guess is 4.5% when the actual figure is less than 0.5% (2017)
• We overestimate the scale of deaths related to substance abuse. The average guess is 7% in Britain when the actual figure is 0.9%.
• Suicide accounts for 1% of deaths across all ages in the UK. The average guess is 8.7%.
• The UK has one of the lowest levels of deaths by transport injury across all countries in the study. The average guess for the proportion of deaths was 9% vs the actual figure of 1%.

Compared to other countries, Brits are less accurate than most – in fact, we are the 10th least accurate country in our “Misperceptions Index” – similar to Australia and Italy.

Looking across eleven questions where we get people to estimate factual realities, there are clear patterns in which countries have a more accurate view of their countries.  To capture this, we’ve calculated the Ipsos “Misperceptions Index”, as shown in the table below. 

For 2020, Turkey receive the unwelcome prize of ‘least accurate’ in their perceptions after coming 3rd in 2018. They are closely followed by Romania and Spain.

Brazilians are the most accurate, followed by South Korea and the Netherlands.

As part of the study, participants were also asked questions about things that could influence their guesses including what they saw most in the news, what they felt had least control over and what they thought would be the most unpleasant way to die.

• People say they see transport injuries (38%), interpersonal violence (37%) and terrorism/conflict (35%) most frequently in the news.
• On average, people are mot likely to have been personally affected by cancer (70%), cardiovascular diseases (60%) and diabetes or kidney diseases (58%).
• When asked what would be the most unpleasant way to die, people are most likely to say cancer (40%), followed by an accident (27%), terrorism or transport injury (26%).
• People think they have least control over being a victim of a terrorist attack (32%), getting cancer (31%) or suffering a transport injury (30%).
• Across all countries, on average, people think they’re most likely to get cancer (31%), a transport injury (25%) or cardiovascular disease (24%).

Gideon Skinner, head of politics research at Ipsos, said:

Our latest Ipsos Perils of Perception study highlights many countries are very wrong when estimating the main causes of death in their country.
Across all 32 countries in the study people underestimate how many people die each year from the biggest killers such as cancers and cardiovascular diseases.  But the public in many countries also overestimate the scale of other causes of death such as those resulting from murder, transport injuries, suicide or substance abuse.
We know there are lots of factors which can influence perceptions so this year we also explored issues such as what people see most in the news, what they have been personally affected by and what they felt they had least control over. In some cases, we can see these appearing to have an influence - for example, in some countries seeing more stories about them in the news does seem to be related to higher guesses around issues like transport injuries and interpersonal violence, even while people may have a better idea of what is actually likely to happen to them. The picture at an individual level is more complex and highlights how challenging it is to reduce to a single factor the influences on our perceptions of the world around us.
Death might not be a topic many of us want to talk about, but our misperceptions around it have clear public health and policy implications – and a more sophisticated understanding of these perceptions will enable a better conversation with the public about it.

Note 1: Interviews carried out online with adults aged under 65.  In countries with a low level of internet penetration, the sample will reflect a more urban, educated, and higher-income profile than the general population

Technical note:

These are the findings of the Ipsos Perils of Perception Survey 2020. 16,000 interviews were conducted between 22 November and 6 December 2019.

The survey is conducted in 32 countries around the world, via the Ipsos Online Panel system in Argentina, Australia, Belgium, Brazil, Canada, Chile, SAR China, Colombia, France, Germany, Great Britain, Hong Kong, SAR, China, Hungary, India, Italy, Japan, Malaysia, Mexico, Netherlands, Peru, Poland, Romania, Russia, Saudi Arabia, Singapore, South Africa, South Korea, Spain, Sweden, Turkey and the USA.

Approximately 1000 individuals aged 16-74 were surveyed Argentina, Australia, Belgium, Brazil, Chile, China, France, Germany, Great Britain, Italy, Japan, Mexico and Spain. Approximately 1000 individuals aged 18-74 were surveyed in the USA and Canada. Approximately 500 individuals aged 16-74 were surveyed in Argentina, Belgium, Colombia, Hong Kong, SAR, China, Hungary, India, Malaysia, Netherlands, Peru, Poland, Romania, Russia, Saudi Arabia, Singapore, South Africa, South Korea, Sweden and Turkey.

18 of the 32 countries surveyed online generate nationally representative samples in their countries (Argentina, Australia, Belgium, Canada, France, Germany, Great Britain, Hong Kong, SAR, China, Hungary, Italy, Japan, Netherlands, Poland, Singapore, South Korea, Spain, Sweden and United States).
Brazil, Chile, Colombia, SAR China, India, Malaysia, Mexico, Peru, Philippines, Romania, Russia, Saudi Arabia, South Africa, and Turkey produce a national sample that is more urban & educated, and with higher incomes than their fellow citizens.  We refer to these respondents as “Upper Deck Consumer Citizens”. They are not nationally representative of their country.

The “actual” data for each question is taken from the Institute for Health Metrics and Evaluation Global Burden of Disease Study (2017). More information can be found at ghdx.healthdata.org/gbd-2017. The source for Hong Kong, SAR, China data is https://www.healthyhk.gov.hk/. Where results do not sum to 100 or the ‘difference’ appears to be +-1 more/less than the actual, this may be due to rounding, multiple responses or the exclusion of don't knows or not stated responses.

Data are weighted to match the profile of the population.