How long do you think you’ll live for? Considering physical and mental health, and other external factors, there’s a good chance your estimate is still going to be biased. What about if a doctor predicted your risk? Even this estimate will probably be off. A new study, has found that of all people to most accurately estimate risk of death, it’s the people trained to administer health surveys.

Face, Movements, and Speech Tell It All

People trained to survey participants about their health more accurately assessed their overall health and risk of dying, even though they weren’t health care professionals. Their predictions were more accurate than both physicians and the participants themselves. This was because they were able to analyze both the detailed responses as well as the respondents’ movements, appearance, and responsiveness.

“Your face and body reveal a lot about your life,” Noreen Goldman, professor of demography and public affairs at Princeton University’s Woodrow Wilson School of Public & International Affairs, said in a statement. “We speculate that a lot of information about a person’s health is reflected in their face, movements, speech, and functioning, as well as information explicitly collected during interviews.”

Goldman’s results come from data obtained during the 2006 wave of her study, the Social Environment and Biomarkers of Aging Study (SEBAS), and death registries of participants through 2011. SEBAS used in-depth home interviews, biological specimens, and medical examinations to determine the links between social environments, stress, and health among middle-aged and older Taiwanese adults. It began with a self-reported health assessment. At the end of each assessment, participants, physicians, and interviewers each answered the question, “Regarding your/the respondent’s current state of health, do you feel it is excellent (5), good (4), average (3), not so good (2), or poor (1)?”

The participants were given physical exams and abdominal ultrasounds by doctors who were not their primary health care providers. These doctors rated the participants’ health on the same scale based on the exams and on a review of the participants’ medical histories. Lastly, the participants were interviewed and given a set of tasks to complete, which included walking short distances and climbing up and down from a chair.

How Physicians and Participants Assessed Health

Goldman and her team were surprised to learn that doctors’ ratings were relatively inaccurate predictors of participants’ survival. “Given access to such information (medical histories and ultrasounds), we anticipated stronger, more accurate predictions of death,” Goldman said in the statement. “These results call into question previous studies’ assumptions that physicians’ ‘objective health’ ratings are superior to ‘subjective’ ratings provided by survey participants themselves.”

But the survey participants were just as off-base. Although it’s used widely by health professionals, and has been shown to have some credibility in predicting mortality, “it suffers from many biases,” Goldman said. “People use it because it’s simple to use … the problem with self-rated health is that we have no idea what reference group the respondent is using when evaluating his or her own health. Different ethnic and racial groups respond differently as do varying socioeconomic groups.” She says that adding interviewer ratings could be an easy way “to improve our measurement of health in any population.”

Other studies have also looked into the risk of death. One study, funded by the National Institute on Aging, measured life expectancy from death backward. It found that certain disease prevalence, such as heart disease and cancer, were rising, while neurodegenerative disease prevalence was dropping. By looking at how long before death these diseases emerged, the researchers were able to determine life expectancy. According to the Centers for Disease Control and Prevention, average U.S. life expectancy is 78.7 years.

Source: Tood M, Goldman N. Do Interviewer and Physician Health Ratings Predict Mortality?: A Comparison with Self-Rated Health. Epidemiology. 2013.