Background/Purpose: The National Quality Forum recently endorsed a controversial quality measure that assesses liver toxicity monitoring for patients receiving oral methotrexate (MTX).  Using national data from the Veterans Health Administration (VHA), we assessed the frequency of liver function testing (LFT) and rates of MTX discontinuation after minimally elevated LFTs among patients taking oral MTX.

Methods: We created a national cohort of incident MTX users ≥ age 65 using linked pharmacy and laboratory data from the VHA during fiscal years 2007-2008. Patients were included if they had ≥ 28-day supply of MTX dispensed and use of VHA services for a minimum of 180 days prior and 90 days after the index MTX prescription.  Patients were excluded if they had any diagnosis of inflammatory bowel disease (ICD-9 codes 555.x, 556.x) or if they had evidence of having obtained care outside of the VHA (patients with any medical encounter billed to Medicare during the study period were removed from the sample). We defined performance on the MTX liver toxicity monitoring quality measure through 2 methods (Table). Using the value of the first abnormal LFT after the index MTX prescription, we assessed rates of MTX discontinuation (defined as lack of MTX dispensed for ≥ 90 days after the anticipated refill date).

Results: 899 new users of MTX met inclusion and exclusion criteria for the study. 97% were male, mean age was 71 years (SD 6.3), and mean follow-up period was 267 days (SD 133). Mean MTX dose received was 11.4 mg weekly (SD 5.1). Performance on the quality measure is described in the Table.  148 (16.5%) patients did not have any LFT testing after MTX was initiated. 136/899 (15%) patients had any abnormality of AST or ALT during the follow-up period; 49/899 (5%) had an elevation of ≥ 1.5 x upper limit of normal (ULN). MTX was discontinued in 28/87 (32%) subjects with LFT elevations < 1.5 x ULN and 27/49 (55%) subjects with elevations ≥ 1.5 x ULN. Compared to patients with no LFT elevations during the follow-up period, patients with LFT elevations < 1.5 x ULN had 1.4 greater odds (95% CI 1.1, 1.8) of stopping MTX.

Conclusion:  Despite many MTX users not receiving liver toxicity monitoring with the frequency intended by the National Quality Forum quality measure, one third of patients with mild LFT elevations may be having their MTX stopped unnecessarily.  These findings suggest that there may be unintended negative consequences to a policy that encourages more frequent liver toxicity monitoring.

Table. Performance on methotrexate liver toxicity monitoring quality measure in a national cohort of veterans