Alice Reid

Her Majesty Queen Elizabeth II died on 8th September 2022. Aged 96, her death certificate gave her cause of death as simply ‘Old Age’. It’s undeniable that she was old when she died, but how common is old age as a cause of death now and in the past, and what can the history of death from old age tell us? 

The Queen’s entry in the register of deaths. Photograph: National Records of Scotland.

The decline of old age as a cause of death 

Causes of death first started to be routinely recorded in England and Wales in 1837, and in Scotland in 1855, and this allows us to examine the causes of death among people of particular age groups over time.

In the 19th century over half of deaths to people over the age of 75 were attributed to old age but, as the graph below shows, this fell fairly constantly (with a brief hiatus in the late 19th and early 20th centuries) until the 1970s. Differences between England and Wales and Scotland are likely to be due to their different registration systems, cause of death categorisations, and levels of medical provision. These factors also influenced the decline in deaths from old age.  

Lower access to a doctor was associated with more deaths from old age 

One of the reasons that more deaths among old people were attributed to old age in the early years of cause of death registration may be different levels of medical certification of death. Every death was supposed to have been certified by a doctor who identified the cause of death, but if a doctor was not available, then the cause of death was given by the informant (usually a family member or neighbour) who reported the death.

Although Scotland had similar numbers of doctors per 1,000 people as England, much of the country was sparsely populated and the distance to a doctor was likely to have been a disincentive to calling medical help for either care or certification, particularly where the patient was reaching their ‘natural end’.

For example, on the Isle of Skye (off the West coast of Scotland) in the 1860s, only 12.5% of deaths to people aged 55 and over were certified by a doctor. Even if a doctor did arrive, he (virtually all doctors were men at this time) was very unlikely to have performed any further investigation such as an autopsy or post-mortem, relying instead on the family’s reports of the course of their last illness. 

The decline of old age as a cause of death, at least in rural places such as the Isle of Skye, is partly attributable to improving medical provision – by the 1890s on Skye the percentage of deaths to people aged 55+ that were certified by a doctor had risen to nearly 50 percent. 

Doctors were instructed to stop using vague terms (such as ‘old age’) as causes of death 

The main reason for the decline in old age as a cause of death, however, was linked to rapid improvements in medical knowledge. The ways that diseases and medical conditions developed were becoming better understood, and this fed through into attitudes about what terms were acceptable as causes on death certificates. The Registrars in charge of death registration developed the view that old age was a vague and uninformative term. For instance, the Registrar General for Scotland wrote in his 1872 report: ‘It is quite plain that the term Old Age, when used as denoting a cause of death, is little better than a confession of ignorance, and greatly vitiates our statistics’.  

Elizabeth Shaw, aged 117. Coloured etching by B. Howlett, 1800, after R. Sheardown, 1780. Wellcome Collection.

It is likely that doctors were instructed to choose a more informative cause for death certificates, citing the specific diseases, conditions, and pathology which brought about the death. As old age as a cause of death declined, it was mainly replaced on death certificates with circulatory diseases and cancers. It is likely that older people had been dying of circulatory diseases and cancers all along, but their deaths had been classed as old age.  

In the Scottish town of Kilmarnock, doctors who commonly used old age as a cause of death before 1872 heeded the call to shift to alternative causes of death, but it is interesting that some of those practicing across this time period never used old age as a cause of death (even in the 1860s). Choosing a cause of death was influenced by personal preference as well as disease prevalence and medical knowledge, and some doctors chose vague terms on purpose to shield the family from association with particular causes of death (such as sexually transmitted infections, alcoholism, and suicide), although such causes were more prevalent among young and middle-aged adults than older people.  

Changing death certification and theories of change 

Changes in causes of death over time can therefore represent changes in medical provision, medical knowledge, and ideas about what constitutes an acceptable cause of death as much as, or more than, real changes in what people were dying from. This has broader implications, because theories about why mortality in general declined so much over the late 19th and early 20th centuries have relied on uncritical readings of published cause of death statistics.  

One such theory is that of ‘epidemiologic transition’, proposed by Abdul Omran in 1971. Omran noted that mortality decline was accompanied by changes in the distribution of causes of death, writing: ‘pandemics of infection are gradually displaced by degenerative and man-made diseases as the chief form of morbidity and primary cause of death’. Omran was primarily talking about the balance between different causes of death, and when overall mortality is declining, risks from the diseases that he classed as ‘degenerative and man-made’ (which would today be called ‘non-communicable’) do not have to actually increase in order to become a larger fraction of all deaths. However, the fact that deaths from old age predominantly transferred into deaths from cardiovascular diseases and cancers (non-communicable) means the official statistics give an erroneous impression of increasing risks from the latter. 

For example, in 1855 a Scot aged 55 or over had about a 3 in 100 chance of dying from a non-communicable cause, if old age is included in such a category with cardio-vascular disease and cancer. By 1949 this risk had scarcely altered, although the chance that such deaths were ascribed to circulatory disease rather than old age had quadrupled. 

From old age, senility, and senile dementia to Alzheimer’s 

The death series in the graph above are derived from published statistics which gathered the vast array of different causes offered by doctors and classified them into a far smaller number of broad groupings. In the 19th century there were a relatively small number of categories, of which either ‘old age’ or ‘senility’ was one (the different terms were used in different years).

We don’t have the precise instructions to coding clerks about how they were supposed to categorise potentially ambiguous causes on death certificates, but places for which we have both original death certificates and published numbers in different categories allow insight. This exercise indicates that in the 19th century causes such as ‘senile dementia’ were placed in the ‘old age’ or ‘senility’ group, rather than being categorised with mental and behavioural disorders (as would be the case today). 

By the late 20th century, there were far more reported categories. Senility was now split into ‘senility without mention of psychosis’ and ‘senility with mention of psychosis’. There were also separate categories for senile dementia, pre-senile dementia of various forms, and Alzheimer’s disease. The series shown in the graph only include ‘senility without mention of psychosis’, but if these other categories were included the lines would start to rise again from the 1980s, reaching around 10 percent of deaths to people over age 75 in 2010 and nearly 18 percent of deaths in those over age 80 in 2022.  

A doctor visiting a senile old man and discussing his verdict with the patient’s wife. Wood engraving by C.E. Brock, 1902. Wellcome Collection.

The fact that dementia (including senile dementia) was included in the old age category in the 19th and early 20th centuries means that deaths from old age are probably a bit inflated compared to the late 20th and early 21st centuries. However, there is evidence that dementia and Alzheimer’s are rising as people are living longer and postponing many forms of ill-health to older ages, so such deaths are likely to have been lower in the past and the distortion of the earlier series may only be modest.  

How should we interpret the Queen’s death from old age? 

The long-term move away from vague causes such as old age, towards causes of death which describe a precise medical process, makes the fact that the Queen’s death was attributed to old age particularly interesting. Although most doctors feel old age as a diagnosis is unhelpful, there is also a recognition that advanced age is associated with complex health conditions, and a feeling among some medics that it might be good to recognise this on health records and death certificates.  

When the WHO’s International Classification of Disease was updated from ICD10 to ICD11 on 1st January 2022, there was a proposal to replace the cause ‘senility’ (used in ICD10), which was increasingly seen as a negative term, with old age (which had been used in the 19th century). In the end, worries about ageism and the possibility that this would lead to age being seen as a disease meant that this change did not happen, with the term ‘ageing associated decline in intrinsic capacity’ chosen to replace ‘senility’ instead. 

Nevertheless, in England and Wales ‘old age’, ‘frailty of old age’, ‘debility of old age’ and ‘senility’ are still listed as acceptable causes of death among people over the age of 80. The authorities state that they should only be used in very limited circumstances of gradual decline and the absence of any identifiable disease that contributed to the death. Such usage captures the spirit of the ICD11 term ‘ageing associated decline in intrinsic capacity’, and might well have reflected the Queen’s condition. But it is likely that the choice of old age as a cause of death was also inspired by a desire to protect the Queen’s privacy (and that of her surviving family), harking back to the use of vague terms by 19th century doctors. This does not imply that the real cause was embarrassing in any way (as sometimes in the 19th century), but in this case any specific cause was likely to have alerted considerably more speculation than old age. 

A very old man, suffering from senility. Colour stipple engraving by W. Bromley, 1799, after T. Stothard. Wellcome Collection.

Further reading 

• Adhiyaman, V., and Chattopadhyay, I., ‘Is it appropriate to link ‘old age’ to certain causes of death on the medical certificate of cause of death?’ Future Healthcare Journal 8:3 (2021), e686-e688.  
• Hawley, C.L., ‘Is it ever enough to die of old age?’ Age Ageing 32 (2003), 484–6.  
• Omran, A.R., ‘The epidemiologic transition: a theory of the epidemiology of population change’, Milbank Memorial Fund Quarterly 49:4 (1971), 509-538. 
• Rabheru, K., et al, ‘How “old age” was withdrawn as a diagnosis from ICD-11’, The Lancet Healthy Longevity 3:7 (2022), e457-e459. 
• Reid, A., Garrett, E., Dibben, C., and Williamson, L., ‘‘A confession of ignorance’: deaths from old age and deciphering cause-of-death statistics in Scotland, 1855–1949’, The History of the Family 20:3 (2015), 320–344.  

Tags: death, demography, doctors, mortality, old age

| This entry was posted on Thursday, November 14th, 2024 at 12:02 am and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.