Stress, and its role in illness - I

The studies we have been able to locate are limited to two types of stress: stress at work and marital stress.

We note that all the studies we report on correct for all know factors of the illnesses, so that the effect of stress as reported, is expected to be as bias-free as possible.

The first study involved 812 employees of a Finnish factory who were followed for an average of 26 years. Workers with initial cardiovascular diseases were excluded. The study used a self assessment questionnaire to measure stress and effort-reward imbalance. The questions included issues such as work demands and the degree of responsibility, task difficulty, mental load, decision authority, mental load, income, fairness of supervision, job security, and promotion prospects. Of the 812 employees, 73 deaths from CVD occurred during followup. They found that there was a risk ratio of 2.90 (1.25 to 6.71 Margin of Error) for cardiovascular mortality with high job strain. There was a similar ratio for reward-effort imbalance: workers who were poorly rewarded for their effort had an increased risk ratio of 2.59 (1.18 to 5.68 Margin of Error). They did not find any enhanced risk for high job demands or high efforts.

Reference 2 involves workers at 27 workplaces in Scotland. 5606 men were tracked for 5 years. Stress was determined using a 4-question test:

1. Am I generally tense or nervous?
2. Is there a great amount of nervous strain connected with my daily activities?
3. Am I completely exhausted mentally and physically at the end of the day?
4. Are my daily activities extremely trying and stressful?

The researchers find that for the workers classified with stress in the top third (high stress versus low and medium stress), there is an increased risk ratio for angina (perceived, self-reported severe chest pain that occurs when the heart muscle doesn't receive a supply of enough oxygen) - 2.66 ( 1.61 to 4.41 Margin of Error interval). However, for Ischemia (actual decreased flow of oxygenated blood to the heart due to obstruction in an artery), there is an indication of reduced risk - 0.67 ( 0.36 to 1.26 Margin of Error). They conclude that "higher stress was not associated with hospital admissions... Nor was stress associated with mortality". They state that "results showed a reporting bias - that is, people who viewed their lives as more stressful were also more likely to report more symptoms". In lay language, this says that people who are more stressed feel much worse, but this is not due to actual illness. One may wonder whether this method of stress determination is too simplistic, or else indeed the effect of stress has been oversold (based on this one study).