Purpose: Interferon alfa treatment in multiple myeloma marginally impr
oves relapse-free and overall survival, Often it does so of the expens
e of toxicity and financial cost, If patients are unwilling or enable
to participate in the decision of whether to initiate such treatment,
known patient preferences can serve as guidelines for the physician, W
e interviewed myeloma patients in the United States to obtain informat
ion that might facilitate medical decision-making. Patients and Method
s: Three hundred fifty-five myeloma patients throughout the United Sta
tes were interviewed by telephone. Without identifying interferon alfa
as the treatment agent, interviewers described potential adverse effe
cts, financial cost, and self-injection procedures. The potential bene
fits of four treatment choices, derived from a meta-analysis of publis
hed data, were presented as gains in remission rate (+10%), remission
duration (an additional 4 and 7 months, respectively, for induction an
d maintenance treatment), and overall survival (an additional 3 and 6
months, respectively, for induction and maintenance treatment), Patien
ts' choices for or against use of the unidentified substance were reco
rded, and interferon was subsequently disclosed as the treatment. The
profiles of patients making different choices were determined using mu
ltivariate regression techniques. Results: Approximately half of the p
atients accepted the unidentified treatment if remission and/or surviv
al improved by at least 6 months. Accepters were younger and more like
ly to have used interferon. Of patients who rejected the unidentified
treatment, 25% to 50% would have been willing to accept it if the bene
fits were greater than or equal to 12 months. Test/retest reliability
of all choices, determined in 36 cancer patients, was 0.896. Conclusio
n: In multiple myeloma, interferon therapy and, by inference, other tr
eatments with comparable features are acceptable to approximately half
of the patients if a 6-month gain in relapse-free or overall survival
can be expected. (C) 1997 by American Society of Clinical Oncology.