Besides containing articles on predictors of new-onset kidney disease and antibiotic use in relation to the risk of breast cancer, this month's Journal of the American Medical Association touched on a surprising subject--games.

　　In their article "Content and Ratings of Teen-Rated Video Games," researchers Kevin Haninger and Dr. Kimberly M. Thompson surveyed 396 commonly available video games rated "T" by the Entertainment Software Ratings Board. Of these games, 94 percent carried warning labels for violence, 15 percent for sexual themes and/or content, 14 percent for profanity, and 2 percent for tobacco, alcohol, and/or other intoxicating agents.

　　When Haninger and Thompson played a random sample of 81 games from the group, they found different results: 98 percent of games contained violence (42 percent featured blood), 27 percent presented sexual themes, 27 percent included profanity, and 15 percent featured the "use of substances."

　　However, of the games observed, only 73 percent that featured sexual themes, 64 percent that featured profanity, and 8 percent that featured substance use actually had front-cover descriptors regarding the potentially offensive content.

　　While she conceded the ESRB system "worked," Thompson told the Associated Press that it "is not providing complete information to parents... In many games, there's content we think parents would care about." As founder of Harvard University's Kids Risk Project, Thompson has made it her career to alert the public to dangers to children's health. (Haninger is a doctoral student at the Cambridge, Massachusetts, university.) The study's results and methodology can be seen on the Kids Risk Web site.

　　For its part, ESRB spokesman Matthew Kagan told Reuters that more independent research has yielded different results. He cited an October 2003 study conducted in which 400 American parents watched footage from 80 games. Of those parents surveyed, more than 75 percent agreed with the game's ESRB rating.