April 29, 2019

expert reaction to sense of smell and death

Research published in the Annals of Internal Medicine shows that poor olfaction is associated with higher long-term mortality among older adults.

Dr Sara Imarisio, Head of Research at Alzheimer’s Research UK, said:

“This study highlights a link between a poor sense of smell and an increased risk of death, and the researchers found a particularly strong link with deaths from dementia or Parkinson’s disease.

“This is well-conducted research, but there are many reasons why someone’s sense of smell could change and it’s not necessarily a sign of a disease.

“Previous research has suggested that changes in sense of smell could be an early indicator that diseases like Alzheimer’s are underway in the brain.

“While scientists are exploring the potential of smell tests to help detect diseases like Alzheimer’s at an early stage, these need to be refined and carefully evaluated in clinical trials before they could be used to support a diagnosis.”

Dr Latha Velayudhan, Senior Clinical Lecturer in Old Age Psychiatry at the Institute of Psychiatry, Psychology & Neuroscience, King’s College London, said:

“The press release accurately reflects the science. It is a good study backed up by good data but with some limitations as mentioned below, limiting its generalizability. It is in keeping with previous studies and adds on some further information such as associations by sex, race and contributions of some known common factors.

“Sense of smell declines with age and is also found to occur at early stages of neurodegenerative disorders such as Parkinson’s disease (PD) and dementia. Several studies have examined the association between smell ability and risk for death in older adults, however little is known about the factors contributing to it. Also, few studies have reported follow up beyond 5 years or examined the associations by sex or race. This new research examined factors to explain the relationship between diminished smell and increased risk for death in in 2289 older adults in United States aged 71 to 82 years. Participants were reassessed at 4 time points across a 13-year period, and 1211 participants died by the end of the study. The smell ability was tested using Brief Smell Identification Test at the baseline and those scoring ≤8 out of 12 were classified as those with poor sense of smell. The study reports poor olfaction had a 46% higher cumulative risk for death at year 10 and a 30% higher risk at year 13 and the associations were similar for both sexes and races (both white and black participants). The findings were driven by participants who self-rated their health as good or excellent at the baseline. Dementia or PD and weight loss together explained only about 30% association of higher long-term mortality with poor olfaction.

“The findings are interesting; however, the authors admit that there could be unobserved confounders affecting the results. Since the participants included in the study were all well-functioning and between 71-82 years of age, it limits the generalizability of the results to younger and less functioning older adults. The smell ability was tested only once at the baseline, so effects of any longitudinal changes over years is not available. This is important as diminished sense of smell in otherwise cognitively intact people predates onset of dementia or Parkinson’s disease by 4 to 8 years. Also, the olfactory impairment tested by BSIT could also have been caused by conditions unrelated to aging such as nasal surgery, nasal polyps, chronic rhinosinusitis. Hence headlines such as ‘poor sense of smell? You’re 50% more likely to die in the next 10 years’ would be inaccurate. 

“People should report if notice developing poor sense of smell, as it could be an indicator of early stages of neurogenerative disorder, and as it can affect feeding behaviour and body weight. However, not all older adults become aware of decline in their olfactory ability, and this is not routinely tested in clinical practice. Further research is necessary for better understanding of the health implications of the olfactory impairment with ageing, the associated mechanisms, and the usefulness of incorporating smell assessment into routine clinical practice.”

Prof Kevin McConway, Emeritus Professor of Applied Statistics at The Open University, said:

“Previous research has found an association between a poor sense of smell in older adults and increased death rates. This new research extends those findings, by following up the participants over a longer period of time (up to 13 years), generally looking at more details of the association, and investigating some possible ways in which sense of smell might be linked to mortality. This is a real advance, in my view.

“The trouble with any observational research, like this, is that it’s impossible to be sure about what’s causing what. That’s because, here, there are lots of differences between the people who could smell well and those who couldn’t, apart from their sense of smell, and any differences in death rates could be due to these other differences and not to the sense of smell at all. The researchers made statistical adjustments to account for other difference that they had data on. However, they acknowledge clearly that there may be other influences that they couldn’t take into account because they had no data on them (so-called ‘unmeasured confounders’), and they did some statistical calculations that indicate that such influences would not have to be particularly strong to change some of their conclusions. That’s got to be taken into account when examining their findings.

“The causation here could be very complicated. One possibility is that poor sense of smell is a sign, perhaps an early sign, of some underlying illness (and that this illness is what leads to increased mortality). Another is that poor sense of smell might itself lead to illness, perhaps because poor smell affects how food tastes, and therefore might contribute to poor nutrition which could lead to bad health. Most likely, both of these possibilities, and others, could be in play. The researchers looked at two specific possible links between poor smell and increased mortality risk – via degenerative conditions of the brain and nervous system (dementia and Parkinson’s disease), and via weight loss. They did find evidence that there were associations between poor sense of smell and increased mortality via both these routes, but also that a considerable amount of the association could not be explained by either of these routes. So there’s clearly a lot more to be found out in future research – and again we must bear in mind that all of these associations might be affected by confounding that couldn’t be measured in this study.

“So should you be concerned about these findings if you feel you have a poor sense of smell? Well, not necessarily. As always with such studies, the results apply to populations of people, not immediately to specific individuals like you. If you’re outside the age range that was studied – participants were aged 71-82 when they entered the study – the results don’t directly relate to you at all. The association between sense of smell and mortality was found only in some groups of participants. Also, it’s plausible that changes in sense of smell, perhaps a worsening over time, might be important, but this research couldn’t consider that because people’s sense of smell was measured only once. And you should bear in mind that (as the accompanying editorial points out) there can be sizeable differences between people’s own assessment of how well they can smell and measurement of their sense of smell by an objective test (as was used in this research).”

Prof Robert Howard, Professor of Old Age Psychiatry at UCL, said:

“This carefully conducted study adds to our understanding of how loss of the sense of smell in people who are in their 70s and 80s contributes to their increased risk of dying over the next 10 years. 

“We have known for many years that loss of sense of smell in later life may precede other symptoms of neurodegenerative disorders such as Parkinson’s or Alzheimer’s by many years. This is because the earliest location of neuropathology in these conditions is in the olfactory system. We also know that loss of sense of smell is associated with weight loss. This is unsurprising, given the contribution of smell to our sense of taste and flavour and enjoyment of food.  

“The study showed that the risk of dying in the next 10 years was increased by about a half in people with impaired sense of smell and that only some of this risk could be explained by the development of Parkinson’s or Alzheimer’s or by weight loss. Most of the increased mortality risk could not be explained by associations with specific illnesses such as cancer or cardiovascular disorders. This raises the interesting possibility that loss of smell may be a marker of generalised ageing and should be taken seriously by older people and their doctors.”

‘Relationship Between Poor Olfaction and Mortality Among Community-Dwelling Older Adults’ by Bojing Liu et al. was published in Annals of Internal Medicine at 10pm UK time on Monday 29 April 2019.

Declared interests

Prof McConway is a member of the SMC Advisory Committee, but his quote above is in his capacity as a professional statistician.

Dr Velayudhan: nothing to declare

No others received.