Physiological effects of steroids can be estimated reasonably well because it can reasonably be supposed that few if any potential users are going to have significant pre-existing medical problems. But when trying to evaluate mental effects, that supposition has no basis. As Darkes (see farther below) and many others have pointed out, one of the chief failings of many studies of steroids and psychiatry is the failure to design the studies so that the cause-and-effect relationship is not tangled. While there are, in some reports, evidences of some possible correlation of steroid use and mental problems, what few if any of those studies address is which is cause and which effect .
Many abusers who inject anabolic steroids may use nonsterile injection techniques or share contaminated needles with other abusers. In addition, some steroid preparations are manufactured illegally under nonsterile conditions. These factors put abusers at risk for acquiring lifethreatening viral infections, such as HIV and hepatitis B and C. Abusers also can develop endocarditis, a bacterial infection that causes a potentially fatal inflammation of the inner lining of the heart. Bacterial infections also can cause pain and abscess formation at injection sites.