The purpose of this study is to explore patient factors associated with differences in methotrexate (MTX) dosing and to compare patient factors and MTX-dosing patterns between those who remained on MTX monotherapy and those who were switched or had additional therapy. A retrospective cohort of 7,017 patients with newly diagnosed rheumatoid arthritis (RA) was identified in the United States Department of Veterans Affairs administrative databases between 1 October 1999 and 30 September 2009. Regression analyses were used to study the association of MTX start and maximum dose attained with various patient characteristics and compare differences between groups who had therapeutic change (having switched to or added another anti-rheumatic agent or having steroids increased by 2.5 mg of prednisone or equivalent) with those remaining on MTX monotherapy. Abnormal serum creatinine (>1.5 mg/dL) was associated lower start and peak MTX doses (p < 0.01). Older RA patients were less likely to attain peak MTX dose of 15 mg or more (p < 0.01). Males and patients 75 and older (compared with <45) had lower risk of therapeutic change (hazard ratio, [HR] 0.80, 95 % confidence interval [CI] 0.72-0.90, and HR 0.42, 95% CI 0.42-0.36-0.50, respectively). Patients who attained higher peak MTX dose had lower risk of therapeutic change compared with those dosed at less than 15 mg/week (HR 0.85, 95% CI 0.77-0.92 for 15 to <20 and HR 0.79, 95% CI 0.72-0.86 for 20 or more). Injectable MTX use conferred lower risk of therapeutic change (HR 0.64, 95% CI 0.52-0.78). Two thirds did not attain a maximum MTX dose of 20 mg/week or more before therapeutic change occurred. Older age and renal insufficiency were barriers to the use of higher MTX maximum dosages. Use of injectable MTX and higher maximum MTX dose were independently associated with higher likelihood to remain on MTX monotherapy. Further studies are needed to explore targeted interventions that may optimize MTX dosing to improve success rates of MTX monotherapy.