Since the beginning of both USDA and FDA, these agencies have only talked about the bimetallic coil thermometer—a highly unreliable thermometer that must be inserted in food 3 inches in order to get the coil sensor into the center of the food. It only gives an average temperature of the food over the 3-inch immersed stem. There is no way one can accurately cook thin food (e.g., steak, chops, fish, shrimp, etc.) and know, + or –2F, what the real temperature of thin food is, using the bimetallic coil thermometer.
Now that the USDA has acknowledged that a tip-sensitive, digital thermometer is the correct one to use, it has one more problem to solve. The thermometer pictured on all of the labels in grocery stores is the bimetallic coil thermometer. The USDA knows this is wrong—we have discussed this. However, three years ago, I was told that it would cost too much to change the picture. As a result, if U.S. consumers use the type of thermometer as pictured, they will most likely overcook their food. I hope that the USDA will consider showing a picture of a tip-sensitive thermometer on the labels of raw meat, fish, and poultry in the market so that consumers will be able to accurately judge the doneness of these foods without overcooking them.
The USDA still has a long way to go, because we do not need to cook to 160F. The reference it uses for 100,000-to-1 Salmonella kill also points out that 15 seconds at 155F or 52 seconds at 150F or 2.7 minutes at 145F will all give the same kill [Goodfellow, S.J. and Brown, W.L. 1978. Fate of Salmonella inoculated into beef for cooking. J. Food Protect. 41(8):598-605]. Of course, the hamburger is even more red at these lower temperatures.
In summary, the USDA has made a small step forward in giving the public accurate information on cooking food. We have a lot more to do.