The Boston Globe reports that researchers at Harvard and Stanford found workplace stress is as dangerous to physical and mental health as is second-hand cigarette smoke.
A meta-analysis of 228 studies published in Behavioral Science & Policy Association examined 10 workplace stressors impacting physical and mental health. Among others, these stressors included: long working hours, shift work, low social support, and lack of employer-provided healthcare. The researchers then measured how those stressors impacted 4 health outcomes: self-rated poor physical health, self-rated poor mental health, physician-diagnosed health problems, and death.
- Work-family conflict increased the likelihood that employees will experience self-reported mental health problems by 200%+ and increased the risk of physical health problems by 90%+.
- Job insecurity correlated with an increase in self-reported physical problems.
- A sense of low organizational justice ("a lack of perceived fairness in the organization"), increased the likelihood of a physician-diagnosed condition by ~50%.
In sum, workplace stressors may predict negative health consequence almost as well as exposure to second-hand smoke.
Omigosh! Is it time for the federal goverment--through the Department of Health and Human Services and the Occupational Safety and Health Administration--to regulate the amount of stress that's tolerable in workplaces?
- The Centers for Disease Control (CDC) firmly asserts that "[s]econdhand smoke harms children and adults, and the only way to fully protect nonsmokers is to eliminate smoking in all homes, worksites, and public places." [NOTE: That’s pretty strong language implying not just a correlation but causation!]
- A study published the Journal of the National Cancer Institute detailed 76k women over 10+ years. There was no statistically significant relationship between lung cancer and exposure to passive smoke. How can this be, given the CDC’s supposed correlation if not causation?
- That same study found that only among women who had lived with a smoker for 30+ years was there a relationship that exhibited "borderline statistical significance."
Might not the identical problem exhibit itself with the finding that workplace stressors may predict negative health consequence almost as well as exposure to second-hand smoke ? That is, correlation and causation are implied when, in fact, neither has been demonstrated.
RULE #1 for reading research findings: Science doesn't "prove" anything. Research attests that the data gathered indicate a relationship that's not due to chance alone and at a predetermined probability of error (e.g., 1/10, 1/100, 1/1k, 1/10k). One can set the level of probability of error so low that just about any relationship can be demonstrated. But, never "proven."
This fact used to be taught in junior high science. Perhaps the folks at the CDC need a refresher course in statistics.
Let the discussion begin…
To read the Boston Globe article, click on the following link:
To read the CDC report, click on the following link:
To read the Insights article, click on the following link: