The prominent researcher has been put on administrative leave pending an investigation into unspecified allegations.
Study finds that a handful of biological indicators can predict a healthy person’s risk of dying within five years.
February 27, 2014|
FLICKR, ROSEMARY VOEGTLIFour blood-based biomarkers can predict a healthy person’s chance of dying within five years, according to a study published this week (February 25) in PLOS Medicine. A team led by investigators at the University of Tartu screened blood samples from more than 17,000 volunteers in Finland and Estonia. The researchers examined more than 100 different biomarkers and found that abnormal levels of just four—alpha-1-acid glycoprotein, albumin, very-low-density lipoprotein particle size, and citrate—were predictive of death risk. Of these, only albumin had previously been tied to mortality.
They then developed a biomarker score that encompassed all four indicators, and found that people with the highest scores were more likely to die within a five-year window. The link persisted even after the research team accounted for factors like age, sex, smoking, drinking, obesity, high blood pressure, and cholesterol levels.
“These biomarkers appeared to actually predict mortality independent of disease,” study coauthor Marcus Perola of the Finland Institute for Health and Welfare told The Telegraph. “These were all apparently healthy people but, to our surprise, it appears these biomarkers show an undetected frailty which people did not know they had.”
The researchers said they next plan to explore what might connect these four biomarkers and account for this hidden frailty, and suggested that their methodology could be adapted for clinical use down the line.
March 14, 2014
Ironic that you should publish this in the same issue as you publish a discussion and link to the information quoted below. I tend to believe David Colquhoun:
Because most clinical tests produce at least some false positives, “screening healthy people is generally a bad idea,” pharmacologist David Colquhoun from University College London wrote at his blog this week (March 10). Colquhoun was responding to recent research touting screening tests for Alzheimer’s disease, including a study published in Nature Medicine this week (March 9) that showed blood-based biomarkers could predict future cognitive impairment in asymptomatic older adults, touting greater than 90 percent accuracy. He noted that this study was affected by the “hazards” of statistics in that up to two-thirds of “positive” tests could be false.
Show me the population you are screening. It is an unfortunate truth that some of us will die within 5 years.