In a finding that will surprise few parents, UC San Francisco researchers have shown that what adults say about the probability of harm from a risky behavior can have different - even opposite - meanings to teenagers.
"Those differing interpretations can have serious effects if a teen makes the wrong choice about a behavior, whether it's neglecting to take a needed medicine or experimenting with unprotected sex," said UCSF adolescent medicine expert Bonnie Halpern-Felsher, PhD. She and her colleague Michael Biehl, MA, presented data from an ongoing study at the Pediatric Academic Societies' annual meeting in San Francisco on Monday, May 3.
"Physicians and other health practitioners typically use probability terms like 'likely' to communicate risk," said Halpern-Felsher and Biehl in their presentation. "Often there's a very good reason: we don't know the precise percentage chance that a risky action will lead to trouble. But our data show that children, adolescents and even adults don't understand and interpret these terms as we would expect - not even terms like 'rarely' or 'never.' "
"If you give the same risk message to two people of different ages, they may walk away with different interpretations and may make different decisions," Halpern-Felsher said. That could mean, for example, that after a conversation with a physician, a parent assumes that her daughter has learned that the chance of getting a sexually transmitted disease from unprotected sex is high - after all, the doctor described it as "very likely." But the teenager may interpret the risk as much lower.
Alternatively, when parent and child listen to a doctor describe a treatment with a "small chance" of side-effects, the parent may interpret that probability as 1 out of 10 while the child may envision it much more fearfully - as perhaps 7 out of 10.
The researchers studied 5th graders, 7th graders, 9th graders and adults in their 20s, using questionnaires that measured the subjects' interpretation of probability expressions in several different ways. They found that adults, adolescents and children interpreted the same terms very differently. Adults interpreted the term "will definitely happen" as meaning there was a 97 percent chance the event would occur. Ninth graders and seventh graders thought the chance was not so high - 86 and 81 percent, respectively. Fifth graders rated "will definitely happen" as having only a 69 percent likelihood of occurring. There also was a large range of interpretations among the judgements within each age group, with children and adolescents having a range of answers that varied more than 25 points in many cases.
For health practitioners who want to make sure that their messages about probability are getting across, Halpern-Felsher said that a numeric scale may be more effective than vague terms like "probably" or "possibly." To make sure that both the physician and the patient are visualizing the same numbers, it may help to ask the patient to repeat the concept in his or her own words. Adults and adolescents may understand percentages or phrases like "15 out of 100." For younger children, it may be more appropriate to use a thermometer or to have the child color in 15 figures in a drawing of 100 stick figures, to represent 15 percent.
It is often useful to ask the patient to repeat information about risk in his or her own words, so all parties know how the patient interprets the chance of danger.
"These other methods also need to be tested - we don't know for sure how well they work to convey risk information," said Halpern-Felsher. She also said that the sample size for this pilot study was relatively small. While the results about children and adolescents reflect the confusion about probability expressions that have shown up in studies of adults, a larger study is needed for definitive data about the differences in understanding among age groups.
Co-authors for the study "Adolescents' and Adults' Understanding of Probability Expressions" are Bonnie L. Halpern-Felsher, PhD, assistant professor of pediatrics at UCSF and Michael Biehl, MA, of the division of adolescent medicine, department of pediatrics, UC San Francisco. The research was supported by a grant from the National Institute for Child Health and Human Development to Halpern-Felsher and principal investigator Susan Millstein, MD, professor of pediatrics at UCSF.