Thinking about getting a steroid shot to relieve the pain of knee arthritis? Think again. A new study by researchers at Tufts Medical Center suggests that you would do just as well if you injected a placebo. The New York Times reports on the research, which randomly assigned 140 men and women over 45 with painful knee osteoarthritis to injections of either a corticosteroid or a saline placebo. The subjects were injected every three months for two years, with neither the patients nor the people who gave the shots knowing who got the placebo. The Times reports that after two years, there was no overall difference between the two groups in pain or stiffness. The only observable difference was a clinically insignificant cartilage loss in the steroid group.