Introduction Non-adherence to maintenance 5-ASA occurs in at least 30% of ulcerative colitis (UC) patients and is associated with adverse health outcomes and increased healthcare expenditure. Targeted strategies to convey information about benefits of medication to patients may improve adherence. We undertook to discover the preferred mode of information delivery among UC patients; the thresholds of benefit that would produce adherence were also assessed.

Methods Four methods of displaying information about the benefits of maintenance therapy in remission were explained to UC patients in remission, during face to face structured interviews. These were largely conventional numerical approaches: relative risk reduction [RR], absolute risk reduction [AR], number needed to treat [NNT]. The fourth was an optical representation via Cates plot [CP]). Patients understanding and preference for each approach were recorded. Patients were asked to state the minimum thresholds required to adhere to a hypothetical medication (with 5-ASA like properties) for the benefits of relapse and cancer reduction respectively. Thresholds were determined for each method of display.

Results Of 50 participants (mean age 50 years; 58% male) 48% preferred data presentation by RR over CP (28%), AR (20%) and NNT (4%). 94% found RR easy to understand, better than CP (74%), AR (88%) or NNT (48%). Thresholds required for adherence also differed between methods. For bowel cancer prevention, 94% indicated adherence for benefit levels of 61% RR or lower but only 57% would adhere when presented with the corresponding CP (p<0.001). For relapse prevention, 78% of patients chose a threshold of 40% or lower but only 43% chose the corresponding CP (p<0.001). When presented with RR, adherence minimum thresholds equivalent or lower to the actual 5-ASA benefits were applied by 98% of patients for cancer reduction and 78% for flare reduction.

Conclusion Ulcerative colitis patients prefer RR and CP as methods to display medication benefit. NNT is poorly understood and unpopular. Patients apply significantly higher thresholds for adherence when presented with CP in comparison to RR. Presented with information in this way, most patients would choose to adhere to 5-ASA medication when offered the actual benefit profile. Reduction of cancer risk may be a stronger motivator than maintenance of remission. Interventions to improve 5-ASA adherence should use RR and convey benefits for cancer and flare prevention.

Competing interests C Selinger: Grant/Research Support from: Shire, Ferring, Nycomed, Y Kinjo: None declared, J McLaughlin: None declared, A Robinson: None declared, R Leong: Grant/Research Support from: Shire, Ferring, Nycomed.