Objective: Mesalamine non-adherence is common among patients with ulcerative colitis (UC), and can be difficult to identify in practice. We sought to determine whether a random urine test for salicylates could be used as a marker of 5-aminosalicylic acid (5-ASA) ingestion and identify patients at risk of non-adherence. Our aim is to determine whether measurement of salicylates in a random urine sample correlates with 5-ASA levels, and predicts an individual's risk of mesalamine non-adherence.

Methods: Prospective observational study. Urinary salicylates (by colorimetry) and 5-ASA (by liquid chromatography and tandem-mass spectrometry) were measured in a random urine sample at baseline in patients and controls. Mesalamine adherence was quantified by patient self-reports at enrollment and pharmacy refills of mesalamine over 6 months.

Results: A total of 93 patients with UC taking mesalamine maintenance therapy were prospectively enrolled from the clinic. Random urine salicylate levels (by colorimetry) were highly correlated with urine 5-ASA metabolite levels (by mass spectrometry; R2=0.9). A random urine salicylate level above 15 mg/dl distinguished patients who had recently taken mesalamine from controls (area under the curve value 0.9, sensitivity 95%, specificity 77%). A significant proportion of patients (27%) who self-identified as "high adherers" by an adherence questionnaire (Morisky Medication Adherence Scale-8) had random levels of urine salicylate below this threshold. These patients were at higher risk of objectively measured non-adherence to mesalamine over the subsequent 6 months (RR: 2.7, 95% CI: 1.1-7.0).

Conclusions: A random urine salicylate level measured in the clinic can identify patients who have not recently taken mesalamine, and who are at higher risk of longitudinal non-adherence. This test could be used to screen patients who may warrant interventions to improve adherence and prevent disease relapse.