Mental health

Worried to death? Distress linked to early death

Those with an anxious disposition may want to look away now, as The Daily Telegraph is reporting that ‘even low levels of stress or anxiety can increase the risk of fatal heart attacks or stroke by up to a fifth’.

This news is based on a well-designed study which pooled data from over 68,000 adults in England and looked at how their levels of psychological distress affected their risk of death from any cause, or due to specific types of conditions such as heart attacks, stroke and cancer. The people were followed over the course of eight years.

The symptoms of psychological distress include:

  • anxiety
  • depression
  • social problems
  • loss of confidence

Previous studies have found links between moderate to severe psychological distress and serious conditions. However, the researchers were surprised to find that even mild feelings of psychological distress (so-called ‘sub-clinical symptoms’) also lead to an increased risk of heart attack or stroke; but interestingly, not cancer.

Only people with high levels of psychological distress were at increased risk of death from cancer.

The researchers did theorise that there may be a direct connection between psychological distress and physical disease. For example, it is known that acute feelings of stress can reduce the flow of blood to the heart and that depression can lead to increased levels of inflammation inside the body.

But whether these types of factors actually contribute towards early death is pure speculation at this moment in time.

Though it is rarely possible to be able to say conclusively from a single observational study, or pooling of such studies, that one factor definitely causes the other.

Further research is needed to determine whether reducing psychological distress could, in some way, potentially reduce the risk of earlier death.

Where did the story come from?

The study was carried out by researchers from The Scottish Dementia Clinical Research Network and other research centres in Scotland and England. The study received no specific funding.
The study was published in the peer-reviewed British Medical Journal.

Even though the headlines sound scary, they are a broadly accurate representation of the association between psychological distress and risk of early death that was identified in the study. However, it’s not possible to say for certain that ‘stress or anxiety’ directly cause the increased risk as some headlines may imply.

What kind of research was this?

This was a statistical pooling (meta analysis) of studies looking at the relationship between psychological distress and death. They say that some, but not all, studies have found a link between depression and anxiety, and risk of premature death, and these studies have been relatively small. In particular, the researchers were interested in psychological distress that would not meet the criteria for a mental health diagnosis.

By pooling the data from 10 large cohort studies, this gave the researchers a much larger sample, which can give more reliable results than smaller studies. The methods they used were based on getting data on each individual person and pooling this, as opposed to pooling the overall results data from each study. This individual patient method means the researchers can usually carry out a more detailed analysis of the data.

What did the research involve?

This study used data on psychological distress collected as part of the Health Survey for England which was carried out annually between 1994 and 2004. Only data for adults aged 35 or over was used. People who already had cancer or cardiovascular disease at the time of the survey were excluded. Individuals who died up to 2008 were identified using NHS mortality data.

Psychological distress was measured using a standard health questionnaire called the General Health Questionnaire (GHQ-12).
It covers symptoms of:

  • anxiety
  • depression
  • social dysfunction
  • loss of confidence

Scores on the GHQ-12 were used to group people as having no symptoms (asymptomatic), having a low level of symptoms (sub-clinically symptomatic), having a moderate level of symptoms (symptomatic), and having a high level of symptoms.

The causes of death were identified from death certificates, and the researchers were interested in deaths from cardiovascular causes, cancer, and external causes such as accidents, injury and intentional self-harm. The risk of death in all the groups with psychological symptoms was compared with the group with no symptoms. The analyses took into account:

  • age
  • gender
  • type of occupation
  • alcohol consumption
  • blood pressure
  • body mass index (BMI)
  • smoking
  • diabetes status

The researchers also carried out an analysis where they excluded people who died in the first five years of the study, to make sure they were not including people who were already sick when their psychological distress was measured.

What were the basic results?

The researchers analysed data from 68,222 people with an average age of 55.1 years. They were followed up for an average of 8.2 years. In this time there were 8,365 deaths (12% of participants). Of these, 40% were related to cardiovascular disease, 31% related to cancer and 5% to external causes.

Having symptoms of psychological distress was associated with an increased risk of death during follow-up. After taking into account other factors that could influence risk of death, compared to those with no psychological distress symptoms:

  • People with a low level of symptoms had a 16% higher risk of death (hazard ratio 1.16, 95% confidence interval 1.08 to 1.24).
  • People with a moderate level of symptoms had a 37% higher risk of death (hazard ratio 1.37, 95% confidence interval 1.23 to 1.51).
  • People with a high level of symptoms had a 67% higher risk of death (hazard ratio 1.67, 95% confidence interval 1.41 to 2.00).

The increasing level of risk with increasing levels of symptoms is interpreted as a sign that the link could be a real one, as this is what would be expected if distress was related to risk of death. Similar results were also found for death from cardiovascular causes. Excluding people who died in the first five years of the study did not have a large impact on these results.

When looking at death from external causes, the risk of death was not significantly higher in those with low levels of psychological symptoms, but was about twice as high in those with moderate levels of symptoms, and three times as high in those with high levels of symptoms compared to those with no symptoms.

For cancer deaths, the risk was only significantly higher in those with high levels of symptoms. This link was no longer significant if those who died in the first five years of the study were excluded. This suggests that there is the possibility that some people might have already had cancer at the start of the study, although this was not reported in the survey, and this may be influencing the results.

How did the researchers interpret the results?

The researchers concluded that psychological distress is associated with increased risk of death from several major causes, with higher levels of distress associated with higher levels of risk. They note that risk of death was raised, even at lower levels of distress.


This study was well designed and conducted. Its strengths include the large number of people it included, and the fact that it used individual data on each person, which allowed it to take into account factors other than psychological distress that could influence results. The fact that increasing levels of distress were associated with increasing levels of risk of death supports the possibility that this is a real association. The fact that the association with death from any cause or from cardiovascular causes remained even after taking into account factors that could influence the results, and removing people who may already have been ill at the start of the study also supports the results.

As with all studies, there are some limitations:

  • As the underlying studies were observational there is the possibility that unknown or unmeasured factors, other than the one of interest (in this case psychological distress), are influencing the results. The authors did try to minimise this risk by taking a range of factors into account in their analyses, such as smoking and occupational social class.
  • Cause of death was identified from death certificates, and these may not always be accurate. For example, a post-mortem will not always be carried out, and different doctors who write these certificates may vary in how they classify and record causes. However, the authors note that they used broad categories of causes of death, which means that they should be reasonably valid.
  • The authors note that the GHQ-12 cannot by itself be used to determine whether people have a clinical diagnosis of depression or anxiety, so we can’t say who in the study would definitely have such a diagnosis.
  • A relatively large number of participants were missing data on one or more of the factors being assessed. However, the authors carried out analyses that suggested that this was unlikely to have a large effect.

It is rarely possible to be able to say conclusively from a single observational study, or pooling of such studies, that one factor definitely causes the other. However, this study does suggest that symptoms of psychological distress may be associated with an increased risk of dying earlier. As the researchers themselves note, research is needed to determine whether reducing these symptoms in some way could potentially reduce this risk.

The research supports the importance of mental wellbeing – read more advice about improving mental wellbeing.

NHS Attribution