The widespread use of cellphones makes it more difficult to conduct national health surveys, experts say. During the first half of 2008, the National Center for Health Statistics reported that 16 percent of American adults resided in households that had cellphones only and no landline. Just three years earlier, that figure was 7 percent.

The federal government has always relied on landline telephones for its Behavioral Risk Factor Surveillance System (BRFSS), which is its key tool for gauging Americans' health practices. But using data from landline-only users may be problematic. Youth, men, and Hispanics are more likely to only use cellphones, so they are missed during traditional surveys.

Also, scientists cite the "mode effect" as the reason people answer questions differently depending on how they are reached. Because of this, when a group of people with the same age, racial and educational traits were called on a landline, 25 percent said they smoked. But when reached by cellphone, 31 percent said they smoked. Reached by landline, 38 percent reported having been tested for HIV; but when called on a cellphone, 54 percent reported testing.

Ali H. Mokdad, a former CDC epidemiologist who recently conducted the BRFSS, said it is possible that landline users "are less likely to say something bad about their own behavior," he said. "It's like, 'This is my house.'"

BRFSS surveyors will include cellphone users in 10 percent of this year's respondents, but it will not be easy. It is illegal to use usual automatic dialers to cellphones, which slows down surveyors. Many cellphone users are children and thus not eligible be interviewed. Cellphone users may refuse to take a survey because they have to pay for incoming calls.

Nevertheless, the "cooperation rate" for the BRFSS was 72 percent in 2007, which was the same as 1994. The smaller, in-person National Health Interview Survey had a response rate of 87 percent in 2006, compared to 92 percent a decade earlier.

Back to other news for January 2009

Search the Newsroom archive