Despite the considerable clinical, radiologic, and functional benefits of biologic inhibitors in inflammatory arthritides, some concern exists regarding the occurrence of infections in patients treated with these agents. Clearly, comorbidities such as diabetes mellitus, heart disease, disability, and concurrent immunosuppressive medication all contribute to the risk of infection. Increased and closer observation may be in part responsible for some of the reported increases in the rates of mild infections with these drugs. The development of serious infections, particularly TB, in patients taking infliximab seems to be greater than would be expected in this population. Furthermore, experimental data from in vitro investigations and animal models demonstrate a link between decreased TNF alpha activity and increased susceptibility to TB. Why some patients, but not others, succumb to rapidly disseminated infection is unknown but may be related to the extent of TNF inhibition in different individuals. This difference in inhibition may also explain why the incidence of TB seems to be increased with infliximab in comparison with the other TNF blockers. Attribution analysis is the method used to assess the likelihood of a connection between two occurrences and includes such factors as temporal association, few alternative explanations, analogy with similar cases, and biologic plausibility. The putative relationship between anti-TNF treatment and infection is further strengthened by the presence of these factors . Continued vigilance is therefore required in the use of biologic agents in patients with RA, most of whom are already in some way immunocompromised. Everyone who is under consideration for such treatment should be carefully evaluated for the presence of infection, and prophylactic antituberculous therapy should be used if latent TB is discovered. Both patients and primary physicians need to be aware of the possibility that serious infection may develop; if such a problem is diagnosed, the biologic inhibitor should be discontinued until adequate treatment has been completed. Caution is advised in patients with recurring infections and in those with severe comorbidities, for example, poorly controlled diabetes mellitus or heart failure. Administration of live vaccines to patients taking these drugs is not recommended, but patients should be brought up-to-date with all immunizations relevant to their age group before commencement of therapy. Physicians prescribing biologic agents should be encouraged to report any suspected drug-related adverse event. Long-term observation will be required to determine the exact nature of the relationship between cytokine inhibition and infection.