We know very little about the relative cost effectiveness of commonly
used drug treatments for rheumatoid arthritis (RA), largely due to dif
ficulty obtaining appropriate data and the complexity of the required
analysis. There is growing interest in combination therapy for RA. How
ever, it is unclear whether combination therapy will be associated wit
h increased benefit compared to monotherapy, and whether the degree of
benefit will offset the anticipated increased costs. We show 3 specif
ic examples derived from a longitudinal database of adults with RA in
which we compare the cost effectiveness of longterm drug regimens, Fun
ctional status serves as our measure of effectiveness, and we use a co
nservative estimate of the total annual cost of drug therapy. We also
consider the analytic complexity of differences in baseline disease ac
tivity. These examples illustrate that in addition to variability in b
aseline disease activity, there is also substantial variability over t
ime in both effectiveness and costs associated with drug treatment. Th
is variability underscores the importance of considering both the rela
tive effectiveness and costs when making treatment decisions. To make
progress in the area of cost effectiveness of longterm treatments for
RA and other chronic conditions, it is essential that studies of treat
ment continue for sufficiently long periods and that relevant cost inf
ormation be collected in conjunction with measures of effectiveness.