Saturday 26 May 2018

Causes of death in 2017 – the decline of coronary heart disease

To determine the causes of death of the population is not easy. The data from the UK Office of National Statistics (ONS) are comprehensive, complicated, and inevitably a few years after the event. An alternative is look out for samples, the causes of death of smaller groups rather than the population as a whole. This inevitably produces biases: if the sample is not random (a complex process) it might not be fully representative of the population. However it should give some idea of why people die.

There is an opportunity in a study of doctors. Doctors might not be entirely representative of society but the seminal study of the dangers of cigarette smoking, conducted by Sir Richard Doll in the 1950s, was a study of doctors. Its message was applied to the population as a whole.

Doctors are thought to have a longer than average life expectancy. This goes with affluence and a high level of education, both of which are related to longer life expectancy. There is a quoted study of doctors that states that underlying the longer life expectancy, the ways in which they differ from the general population include: only rarely visiting a doctor, taking very few medications, avoiding screening tests, and drinking rather a lot of alcohol. To this must be added in recent decades, that very few smoke cigarettes. However this alleged study is rather cynical and might be apocryphal. 

At present, average life expectancy in the UK is 79.4 years for males and 83.1 for females, overall 81.6. So let us look at doctors.


In the British Medical Journal each week we can usually find about six obituaries. In recent years most of these have given the cause of death in addition to personal and career details. So here we have up-to-date causes of death with age and sex, and this is an important source of information.

I have recorded these obituaries during 2016 and 2017. 

I have been particularly concerned to determine the current causes of death. We are still told that coronary heart disease (CHD) is the most important cause of death, but I have doubted this now that the epidemic has come to an end. What proportion of deaths are due to coronary heart disease at the present time?

I have recorded and analysed 568 consecutive obituaries. All give date of birth and death so that age of death profiles can be identified. 415 of these gave the cause of death. 

There are obituaries for 450 male doctors and 118 for females. This difference is inevitable as half of the total were born before the year 1930, a time when very few women entered medical training.

The overall age of death profile is as follows:

median age of death of UK doctors is currently 87
Figure 1. Age profile of deaths of doctors, 2016-17

Figure 1 shows the age at death (vertical axis) for each of the 568 individuals. Each vertical column on the horizontal axis represents one doctor, the height of the line indicating age at death. It allows us to see that the 50% median age at death is 87 years. It also shows the few early deaths (on the right).

Analysis of deaths

The causes of death, when given, must be grouped. This can be difficult as the important features are organ and disease type. For example "brain tumour" can be grouped under "brain" or "cancer". Similarly, "pneumonia" can be grouped under "lung" or "infection". I think that understanding is better if cancers are grouped together, and infections are grouped together. Otherwise the organ is the best grouping. I hope this is clear in the figures below.

No deaths are identified as being due to suicide, and the confidentiality of a suicide death is understandable. The two youngest men to die were age 34 and 35 years The cause of their deaths are not given and they might have taken their own lives.

For the purpose of analysis it is more informative to separate male deaths from female deaths.

Deaths of male doctors.

Of the 450 obituaries of male doctors, the cause of death is given in 316.

The average age at death of the 450 is 84 years, compared to 79.4 years for the current national average for men.

The median age at death of male doctors is 87 years, meaning that half died after that age.

Only 4 male doctors (0.9%) died before the age of 50 years. 2 of these died in cycling accidents. The cause of death in the other two is not given (as mentioned above, possibly suicide).

Only 37 (8.2%) died before the age of 65 years. 

Deaths of female doctors.

Of the 118 obituaries for female doctors, 89 give the cause of death.

The average (mean) age at death of the 118 is 77 years. The median age at death of females is 83 years, meaning that half of the deaths occurred after this age.

The average age at death is surprisingly early, the national average for females being 83 years. It reflects a number of early deaths of female doctors, many more than male deaths. This does not appear to be representative of society and was probably an unfortunate and atypical cluster of early deaths.

9 female deaths (7.6%) occurred below the age of 50 years. Of these 5 were due to cancer (not breast), one from brain haemorrhage (age 45) and one following aortic arch replacement (age 31), extremely hazardous cardio-vascular surgery. Another very young female died at the age of 33 due to brain tumour, and another aged 35 due to malignant sarcoma.

32 of the deaths (27.1%) occurred before the age of 65 years compared to 8.2% in males. There were 5 deaths from breast cancer between the ages of 50 and 65. 

Further details of deaths of male and female doctors are shown below.

Details - male deaths

This is the analysis of 316 recorded causes of death in male doctors.

Figure 2. Deaths of male doctors

34% of all recorded male deaths are from cancer (Figure 3). The second major category is diseases of the brain and nervous system, 18.3%.  Heart disease was responsible for only 10% of deaths, the same as old age.

Figure 3. Fatal cancers in male doctors
(percentage is of total deaths)

The most frequent fatal male cancer is cancer of the prostate, but it is the cause of death in only 6% of deaths in which the cause is given. Colon cancer and brain tumour are in second and third position. 11 of the cancer deaths did not name the cancer site.

The second major category is Brain and Nervous System , and the details can be seen in Figure 4.

Figure 4. Death from diseases of the brain and nervous system in male doctors
(percentage is of total deaths)

Dementia is the major cause of neurological death. The cause of dementia is not specified in 9 of the 22 deaths. Alzheimer's disease is specified in 7, vascular dementia in 4, and Pick's disease in 2. Death from motor neurone disease is not as rare as might be thought, 5 deaths, the same as brain haemorrhage.

We have always been told that heart attacks (myocardial infarction, MI) and coronary heart disease are responsible for more than 25% of male deaths. However this is far from true in doctors in the early 21st century century.
 Death from diseases of the heart in male doctors  (percentage is of total deaths)
Figure 5. Death from diseases of the heart in male doctors 
(percentage is of total deaths)
Of the 316 recorded deaths, 34 (10.7%) are the result of heart disease. Heart failure is the most common within this group. Myocardial infarction and coronary heart disease were responsible for only 11 deaths, 3.5% of the total. 

The ages of the 11 male doctors who died from myocardial infarction and coronary heart disease are 52, 72, 72, 82, 86, 88, 88, 90, 90, 92, 94 (mean 82, median 87). Most are therefore very elderly and only one died from myocardial infarction before normal retirement age.

This is confirmation of my previous Blog Posts that the epidemic of coronary heart disease is now at an end.

There are 5 sudden deaths with the precise cause not recorded. It is likely that some of these would be the result of myocardial infarction, but it would make little difference to the remarkably low number of coronary heart disease deaths. Suicide could be be included under sudden death.

Death from trauma is not common but it is perhaps disturbing that 3 deaths were from cycling accidents. The ages of these individuals were 37, 43, and 60 years. The ski-ing death was at the age of 66 years.

Figure 6. Death from trauma in male doctors 
(total deaths 316)

Old age is now an important cause of death, in this study between the ages of 86 and 106 years.

Figure 7. Death from old age in male doctors 
(percentage is of total deaths)

More than 10% of men dying as the result of old age can be seen as a triumph. It might be the ambition of us all to die from old age rather than a specific disease, but the down-side is that death from old age is a slow process. 

The main lung disease causing death is pulmonary fibrosis (7 out of 16). The main infection causing death is pneumonia (20 out of 25).

Details - female deaths

The analysis is of 89 female doctors in which the cause of death is specified.

Figure 9. Deaths of female doctors

As with men (34.1%) cancers are the most common causes of death in women (47.2%).

The cancers causing female deaths are as follows.

cancer deaths in female doctors
Figure 10. Fatal cancers in female doctors
(percentage is of total deaths)
The most common cancer is breast cancer (10.1% of all female deaths). Cancer of the pancreas was responsible for 6.7% of deaths, higher than in men (2.5%). Death from brain tumours is the same in both males (3.2%) and females (3.4%).

The disease of the brain and nervous system most commonly causing death is dementia (including Alzheimer's disease), causing 7.8% of female deaths (6.9% in males).

Figure 11. Fatal diseases of brain and nervous system in female doctors
(percentage is of total deaths)

Death from stroke is less common in women (3.4%) than in men (6.3%). Death from motor neurone disease is again uncommon but significant.

Only 3 females died from heart disease, 1 from heart failure (aged 93 years) and 1 each from myocardial infarction and coronary heart disease. Just 3.4 % of female deaths from heart disease is very low,  lower than in males (10.7%) as expected. The ages of the two females who died from myocardial infarction and coronary heart disease are quite young at 51 and 70.

Deaths from infections are 6 from pneumonia and 1 from meningitis.

Deaths from trauma include 2 road traffic accidents (age 56 and 93), 1 fall (age 90), and 1 ski-ing accident (age 58).

Figure 12. Death from old age in female doctors
(percentage is of total deaths)

As with males (10.7%), old age is now a major cause of death in females, 10.1% (age range 78 to 103).

Old age

Perhaps deaths from "old age" are even greater, in both male and female doctors. Let us look at deaths from "pneumonia" in both males and females, shown in Figure 13.
Figure 13. Deaths from pneumonia, showing age at death.
We can see the ages at which the deaths occurred. Of the 22 deaths, 17 died at the age of 90 or beyond. Were these deaths in reality due to old age, with "pneumonia" simply accompanying the inevitable? If we ascribe the pneumonia deaths to old age, then the total deaths from old age become 16.5% for male and 14.6% for females doctors.


This is an up-to-date view of how people in the UK die, now. The example is doctors, not completely representative of the population. Although the age at death might be greater than average, there is no reason to expect that the causes of death are not representative.

It would have been interesting to repeat the present study of doctors comparing with obituaries in the British Medical Journal in previous decades. This has not been possible because it is only in recent years that causes of death have been given.

We have seen that cancers are now the main cause of death, of both males (about one third) and females (about one half). During the latter half of the 20th century the major cause would have been coronary heart disease. The pattern of cancers is not surprising – breast in female and prostate in males. In this study death from colon cancer is much more common in males than in females. 

The rapid decline of deaths from coronary heart disease is responsible for cancers taking first place. This is clear from the overall greater life expectancy. If there were an absolute increase in cancer deaths overtaking coronary heart disease deaths, then we would see people dying younger.

The second major category of death is diseases of the brain and nervous system, in both sexes. Dementia is the major category of fatal brain disease. 

Heart disease is the third group of causes of death of male doctors. Heart failure is most common in this group, more common than coronary heart disease, which accounts for only 3.5% of all deaths in males.

Old age is the fourth category in males and the third category in females, more than 10% in both. In males death from old age is three times more likely than death from coronary heart disease, and in females five times more likely. This is remarkable.

It is interesting that "old age" is now an accepted cause of death. In fact a major purpose of medicine is to enable people to die from old age, to "conk out" at some time around 90 to 100 years of age. Another purpose of medicine should be to enable this to be a peaceful rather than a medicalised process. Coroners in the UK have been reluctant to accept "old age" as the cause of death, but this is obviously changing.

Infections are also high in the cause of death in males and females, pneumonia being the most common fatal infection. These are potentially preventable causes of death.

It is sad to see premature deaths from trauma, 3 fatal cycling accidents and 2 fatal ski-ing accidents in particular.

The main message from this study is the major decline from death from coronary heart disease, which is no longer a major cause of death. 

11 male doctors died as a result of myocardial infarction or coronary heart disease, just 3.5%. Death from old age is three times more likely.

Only 2 females out of 89 died from myocardial infarction or coronary heart disease  This is only 2.25%, and means that death from old age is now five times more likely. 

The 20th century pandemic of coronary heart disease is now clearly at an end.


  1. Interesting data about doctor's lifestyles and personal health initiatives - particularly drinking (which I long suspected).I am told by purveyors of 'common knowledge' that I should worry about my own liver in view of my thirty odd units alcohol per week. I am rather relieved to see the low level of deaths due to liver disease.

    1. Hello Roger – I agree that there is some reassuring news in this information. Things are generally far better than we would expect from the media, but the media prefer bad news to good news. The sad thing is premature deaths due to cancer in young women, and not preventable cancers. The early preventable deaths would be those from cycling and ski-ing, but I doubt if these lifestyles would be changed. I cannot imagine a public health announcement that cycling should be discouraged / banned / taxed etc.