Over one year, the expected total societal costs were $38,358 with placebo, $37,565 with pregabalin 150mg and $36,418 with pregabalin 225mg. The corresponding responder days were 59.58, 119.16 and 121.78. Both dosages of pregabalin were dominant over placebo, which was simultaneously less effective and more expensive.
The costs and responder days of the other treatments were: $38,001 and 90.27 with duloxetine 60mg; $38,476 and 92.51 with duloxetine 120mg; $37,788 and 90.03 with gabapentin; $37,041 and 109.96 with tramadol; $36,464 and 129.76 with amitriptyline; $38,380 and 72.09 with milnacipran 100mg; and $38,197 and 66.38 with milnacipran 200mg. Pregabalin 150mg was dominant over both duloxetine dosages, gabapentin and both milnacipran dosages, had an incremental cost per responder day of $57 compared to tramadol, but was dominated by amitriptyline. Pregabalin 225mg dominated all treatment except amitriptyline, which was the preferred treatment with an incremental cost per responder day of $6.
Greater cost savings were associated with pregabalin in subgroups of patients with severe fibromyalgia, use of sleep or anxiety medications and sleep problems at baseline.
Sensitivity analysis showed the robustness of base case findings to variations in key inputs. Compared to amitriptyline, pregabalin was cost effective when studies of amitriptyline that reported heterogeneous efficacy results and a small sample size were excluded from the meta-analysis. The probability of response after 12 weeks of treatment was a key input of the model.