Tracking cases of infections occurring in vaccinated pop. vs. non-vaccinated pop. would be ideal to evaluate vaccine efficacy. We don't have that kind of epidemiologic data for lupus patients. Surrogate markers of vaccine response (e.g: hemagglutination inhibition assay, seroconversion/seroprotection rates) are all we have. Lupus patients seroconvert and exhibit seroprotection rates after flu vaccination, albeit these numbers are lower compared to controls. It is assumed when the seroprotection rate reaches a certain threshold, a degree of protection against that particular microorganism exists; in studies of patients with various comorbidities, the rates for infx, hospitalizations and deaths were reduced when the seroprotection threshold was met. Also remember, vaccines are only effective against the microorganism that it is designed to target (last year's flu vaccine was a bust because of a poor match). Pneumonia vaccine efficacy data in SLE are scant, but many published reports cited pneumonia as a preventable cause of death. Given that vaccinations are one of the few tools we have to prevent infection with proven safety and efficacy, why wouldn't you vaccinate your patient?