Some form of direct observation is the only way to do meaningful food safety behavior research, and the phrase, consumer food safety education, should be banned.

Or at least try something new – the stuff that is out there just doesn’t work.

That’s what I take from a preliminary summary of research led by Christine Bruhn, director of the Center for Consumer Research at the University of California, Davis, and Ho Phang, prepared by Meatingplace.

Sure, those are a couple of my primary messages, so it’s easy to agree with someone who agrees with me, but nice to hear it confirmed.

Bruhn and colleagues videotaped 200 volunteers in their homes while they prepared burgers and salad. She observed their methods of defrosting the meat — frozen, preformed burgers — their refrigerators’ temperature, whether or not they put themselves at risk for cross-contamination and how they determined whether the meat was done.

Of those in the study:

* Twenty-five percent said they prefer their burgers pink.
* Eighty-three percent said they used visual clues, rather than a meat thermometer, to determine the doneness of their burgers.
* About half owned a meat thermometer, but almost all of those participants said they used it on larger cuts of meat, not burgers.
* Seventy-five percent said they were unlikely to use a meat thermometer on burgers.

Even though participants knew they were being videotaped, many did not follow recommended guidelines when preparing their burgers:

* Although 90 percent of consumers were observed washing their hands prior to food preparation, the average hand-washing time was just seven seconds, and only 31 percent dried their hands with a clean towel (either a paper towel or a cloth towel that had not been used previously).

* Potential cross-contamination — defined by the study as "an event in which pathogens could be transferred from one surface to another as a result of contact with a potential source of contamination" — occurred in 74 percent of the households.

* While a bar graph showing the temperature distribution of the finished burgers demonstrated that many were at or near the recommended 160 degrees F, a few of the burgers’ temperatures were recorded to be much lower — as low as 112 degrees F. (Study coordinators observing consumer behavior made sure all burgers were cooked to 160 F before volunteers consumed them.)

Even after the exercise, only 23 percent of participants said they would use a meat thermometer on burgers in the future.

Bruhn said,

"Consumer education is not sufficient. Take the extra step. It protects the public, and it protects you."