Objective: To compare depression-related treatment costs and total healthca
re costs for patients diagnosed with depression and treated with either ser
traline, paroxetine, or fluoxetine.
Patients and Methods: Claims records from a national database of patients d
iagnosed with depression who began treatment with an SSRI in 1995. followin
g an antidepressant medication-free period of at least 6 months, were inclu
ded. Treatment course and associated depression-related treatment and total
healthcare costs during the subsequent 12-month treatment period were exam
ined using univariate and multivariate methods.
Results: Nine hundred five (905) patients taking sertraline, 492 on paroxet
ine, and 945 on fluoxetine met inclusion criteria. The groups were similar
and representative with respect to gender and age. Mean dose over the 12-mo
nth treatment period increased 24%, indicating significant titration in all
cohorts. Patients treated with paroxetine had shorter treatment duration (
157.0 days) than did patients treated with fluoxetine (192.6 days) or sertr
aline (166.9 days, P < 0.001). Patients receiving index treatment with paro
xetine were most likely to switch to another SSRI (21.3%); those taking ser
traline were second most likely to switch (16.1%); and those on fluoxetine
were least likely (12.4%, P = 0.001). Mean costs for depression-related out
patient visits and hospitalizations were similar. Mean antidepressant presc
ription costs differed, being $586, $419, and $446 for fluoxetine, paroxeti
ne and sertraline cohorts, respectively (P < 0.001). In this sample, the fl
uoxetine cohort did not have lower nonpharmaceutical healthcare costs to of
fset higher pharmaceutical acquisition costs. Conclusions from median and m
ultivariate analyses were robust to these findings.
Conclusions: During this study period when fluoxetine, paroxetine, and sert
raline were all well-established agents, similar depression-related treatme
nt courses and cost characteristics among all 3 drugs were observed.