Stress and anxiety

Research on stress uses several kinds of measure. Self-report questionnaires ask about symptoms or feelings: the 12-item General Health Questionnaire measures psychological distress on a scale from 0 to 12 [2], and perceived stress is measured with validated scales or simple self-report questions about how stressed people feel [4]. Physiological measures record the body's response, such as cortisol, heart rate and blood pressure. Allostatic load combines several such markers into one score, and studies define it in different ways, which limits comparisons between them [1].

Anxiety disorders are diagnosed clinically and recorded in health registers. They often occur together with depression, and this overlap complicates the study of either condition alone [7]. Mortality in depression and other mental disorders is described under mental health conditions and lifespan.

The pooled analysis of 10 Health Survey for England studies measured distress with the General Health Questionnaire in adults aged 35 and over who were free of cardiovascular disease and cancer at the start. Compared with a score of 0, hazard ratios (the risk in one group relative to another; 1.20 means 20% higher) for death were 1.20 for scores of 1 to 3, 1.43 for 4 to 6 and 1.94 for 7 to 12. Similar graded associations were found for deaths from cardiovascular disease and from external causes, while cancer deaths were associated only with high distress [2].

Stress measured in the body shows a similar pattern. A 2022 meta-analysis of 17 studies found that high allostatic load was associated with higher all-cause mortality in every individual study, with a pooled hazard ratio of 1.22 (95% confidence interval, the range within which the true value probably lies, 1.14 to 1.30) for all-cause mortality and 1.31 for cardiovascular mortality, although results varied widely between studies [1].

Beliefs about stress may also play a part. In 28,753 US adults, those who reported a lot of stress and believed it affected their health a lot had 43% higher mortality over about eight years (hazard ratio 1.43, 1.2 to 1.7) [8]. In a Danish national register, people with diagnosed anxiety disorders had 1.39 times the rate of death from natural causes and 2.46 times the rate from unnatural causes compared with the general population, and co-occurring depression accounted for an important part of the excess [7].

Most research on stress and physical disease concerns cardiovascular disease. A review of pooled data sets concluded that severe stress in childhood, such as physical abuse and household substance misuse, increases the risk of several chronic conditions in adulthood, while the effects of stress in adult life are generally smaller than those of smoking, high blood pressure or high cholesterol. Adult stress appears to act mainly as a trigger of heart attacks and strokes in people who already have extensive atherosclerosis (fatty plaque in the arteries), and as a factor in outcomes after cardiovascular disease has developed [9].

The proposed mechanisms include a lowered threshold for abnormal heart rhythms, increased activity of the sympathetic nervous system with rises in blood pressure, and inflammatory and clotting responses [9]. The role of inflammation in chronic disease is described under inflammation. In a meta-analysis of 6 prospective studies with 118,696 participants, high perceived stress was associated with a risk ratio of 1.27 (1.12 to 1.45) for new coronary heart disease [4]. Stress and worry also delay and fragment sleep, a link described under insomnia and sleep quality.

Long working hours are one of the most precisely counted sources of occupational stress. The World Health Organization and the International Labour Organization estimated that 488 million people, 8.9% of the global population, worked 55 hours a week or more in 2016. They attributed 745,194 deaths from ischaemic heart disease and stroke that year to this exposure, 3.7% of all ischaemic heart disease deaths and 6.9% of stroke deaths, compared with working a standard 35 to 40 hours [5]. These are modelled estimates that combine survey data on working hours with relative risks from earlier systematic reviews.

Positive states are associated with lower mortality independently of negative ones. A 2008 meta-analysis of prospective studies found that positive psychological well-being, including positive mood and traits such as optimism and life satisfaction, was associated with lower mortality in initially healthy populations (hazard ratio 0.82, 0.76 to 0.89) and, more weakly, in populations with existing disease (0.98). The association held in studies that controlled for negative mood, and there were indications of publication bias [3].

A sense of purpose in life has been studied as a separate trait. In 6,163 US adults in the Midlife in the United States study, each standard deviation higher in purpose was associated with 15% lower mortality over 14 years (hazard ratio 0.85, 0.78 to 0.93), regardless of age or retirement [10]. Among 6,985 US adults over 50, those with the lowest purpose scores had 2.43 times the mortality of those with the highest over four years [11].

Studies of very old people have examined personality. In the Tokyo Centenarian Study, 70 people aged 100 to 106 scored higher on openness than expected from younger controls, and centenarian women also scored higher on conscientiousness and extraversion [12]. The study was small and cross-sectional (measured at one time point), so it cannot show whether these traits contributed to survival.

Randomised trials of stress management have mostly measured symptoms rather than lifespan. In one trial, 151 outpatients with coronary heart disease were assigned to 12 weeks of cardiac rehabilitation alone or combined with stress management training. Over a median of 3.2 years, 18% of the combined group had a clinical event compared with 33% of the rehabilitation-only group (hazard ratio 0.49, 0.25 to 0.95) [6].

A meta-analysis of 47 trials with 3,515 participants found that mindfulness meditation programmes produced moderate improvements in anxiety (effect size 0.38 at 8 weeks) and depression (0.30 at 8 weeks) compared with active control conditions, and low-grade evidence of improved stress or distress. Meditation was not found to be better than other active treatments such as drugs, exercise or behavioural therapies [13]. Exercise, including yoga and tai chi, reduced depressive symptoms in a network meta-analysis of 218 trials [14].

Time in green space has also been studied. A meta-analysis of 103 observational and 40 intervention studies found that greater green space exposure was associated with lower salivary cortisol, lower heart rate and lower diastolic blood pressure, although many of the pooled results were limited by poor study quality and large differences between studies [15]. Regular walking is a common way such exposure occurs.

Full deep dive: Stress, calm and lifespan.

The mortality evidence on stress, distress and well-being is observational. Illness can cause distress, so reverse causation is likely to account for part of each association, even in studies that excluded people with heart disease or cancer at the start [2]. Measures of stress differ widely between studies, and allostatic load has no standard definition [1]. Estimates of deaths attributable to long working hours are modelled rather than directly observed [5].

The trials are small or measure short-term symptoms. The stress management trial enrolled 151 patients [6], and the meditation trials measured symptoms over weeks to months rather than disease or death [13]. No trial has shown that reducing everyday stress extends life.