Do people who feel more tired die sooner?
Middle-aged and older adults who reported the most fatigue were more likely to die over the next 16 years, mainly from heart and circulatory disease, even after accounting for many other health factors.
Source
Fatigue is associated with excess mortality in the general population: results from the EPIC-Norfolk study
Study at a glance
- Design
- Cohort — Prospective population cohort: fatigue measured once with the four-item SF-36 vitality scale, then deaths tracked by national death-certificate linkage; Cox models compared highest vs lowest fatigue quartile with stepwise adjustment for confounders and candidate mechanisms.
- N
- N=18101 · 18,101 of the 25,639 baseline attendees who had complete SF-36 vitality data; 4397 deaths occurred during follow-up. Models excluding baseline cardiovascular disease or cancer used a smaller subset.
- Population
- Middle-aged and older adults (40-79 at baseline) from general-practice registers in Norfolk, UK, recruited 1993-1997.
- Outcome
- All-cause mortality, plus cardiovascular and cancer mortality, by quartile of self-reported fatigue.
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The researchers used the EPIC-Norfolk cohort, which recruited adults aged 40 to 79 from GP lists in eastern England between 1993 and 1997. Around 18 months after baseline, 18,101 participants answered four questions about energy and tiredness, and they were split into quarters from least to most fatigued. Deaths were followed through national records until 2013, and the team added confounders (such as age, smoking, depression and physical activity) and possible biological mechanisms (thyroid hormone, C-reactive protein, haemoglobin) to the models one at a time.
What they found
There were 4397 deaths over an average of 16.6 years. After excluding people with prior heart attack, stroke or cancer and adjusting for confounders, the most fatigued quarter had about a 40% higher risk of death than the least fatigued (hazard ratio 1.40); adding thyroid function reduced this to 1.26, while CRP barely changed it. The extra risk was concentrated in cardiovascular deaths (adjusted hazard ratio 1.45), with no clear increase in cancer deaths, and an excess all-cause risk remained even in the fully adjusted model (hazard ratio 1.24).
The limits
What it doesn't show
This is observational, so it shows fatigue marks higher risk, not that fatigue itself causes death or that treating tiredness would extend life. Fatigue and every covariate were measured only once, and only 40-45% of invited people joined the cohort, with about 30% of them missing the fatigue questions. 'High fatigue' here simply means the most tired quarter of a general population over one month, so the results cannot be applied to chronic fatigue syndrome, where symptom duration was not measured. The labelling of confounders versus mechanisms was decided by a simple statistical rule, and after full adjustment the cardiovascular-death estimate was no longer statistically clear.
Key terms
- Prospective cohort study
- A study that measures an exposure in a group of people first and then follows them forward in time to see who develops the outcome.
- Hazard ratio
- The ratio of the rate of an event (here death) in one group compared with another at any point in follow-up; 1.40 means a 40% higher rate.
- Confounder
- A factor linked to both the exposure and the outcome that can create or distort an apparent association if not accounted for.
- Mediator (mechanism)
- A factor on the causal pathway between exposure and outcome; adjusting for it should shrink the association if it explains part of it.
- Reverse causality
- When the outcome, or an early hidden stage of it such as undiagnosed disease, produces the exposure rather than the other way round.
- SF-36 vitality scale
- Four questions from a widely used health questionnaire asking how full of life, energetic, worn out or tired someone felt over the past four weeks.
Flashcards
0 of 11 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
What type of study design was used?
Common questions
Couldn't undiagnosed illness make people tired and also kill them?
That is the main alternative explanation. The authors tried to reduce it by excluding people with known heart disease, stroke or cancer, adjusting for aspirin use as a marker of cardiovascular risk, and dropping deaths in the first two years; none of these changed the results much, though subtle hidden disease cannot be ruled out completely.
Why does thyroid function matter here?
An underactive thyroid slows metabolism, causes tiredness and is linked to hardening of the arteries. Adding thyroid-stimulating hormone to the model shrank the fatigue-death link more than any other factor, suggesting thyroid status explains part of it.
Is being very tired as dangerous as smoking?
In this cohort the unadjusted risk for the most fatigued quarter was similar in size to that of current smokers, but the authors caution that the two hazard ratios are measured on different scales and cannot be compared directly.
More on Cardiovascular