P335. Thiopurine nonadherence is more common in adolescents than adults with IBD, is associated with psychological distress and impacts on the management and prognosis
J. Goodhand1, N. Kamperidis1, B. Sirwan1, L. Macken1, N. Tshuma1, Y. Koodun1, F. Chowdhury1, N. Croft1, N. Direkze1, L. Langmead1, P. Irving2, D. Rampton1, J. Lindsay1, 1Barts and The London SMD, Gastroenterology, London, United Kingdom, 2Guy's and St Thoma's School of medicine, Gastroenterology, London, United Kingdom
Medication non-adherence is common in patients with IBD. The reasons for this are poorly elucidated and the impact of thiopurine nonadherence on clinical outcome has not been reported.
We hypothesised that nonadherence to thiopurine therapy is more common in adolescents than adults with IBD and that psychological distress would be a contributing factor. We also sought to validate the Morisky 8-item Adherence Scale (MMAS-8), and assess the impact of nonadherence on clinical outcome. We conducted a prospective cross-sectional study of 145 consecutive outpatients attending adolescent (median [range] age 19 [16–21], n = 56) and adult (33 [22–78], n = 89) IBD clinics. Non-adherence was defined as a serum 6-TGN level <100pmol/RBC ×108, when the patient was appropriately dosed according to TPMT levels and weight. Psychological distress was assessed using the Hospital Anxiety Depression Scale (HADS) and self-reported adherence using the MMAS-8 (higher scores indicate better adherence).
In total 12% patients were non-adherent. Multivariate analysis [OR, (95% CI), p value] confirmed that being an adolescent [6.2 (1.5–26.5), 0.01], of lower socioeconomic status [1.1 (1.0–1.1), 0.01], of Afro-Caribbean descent [6.2 (1.1–36.2), 0.04] and reporting higher HADS-D scores [1.2 (1.0–1.4], 0.01] were associated with nonadherence. MMAS-8 scores were significantly lower in non-adherent patients (4 [1–6] vs. 6.5 [0–8], p < 0.001 respectively) and negatively correlated with HADS-A (r = −0.21, p < 0.001) and HADS-D (r = −0.17, p < 0.001) scores. Receiver operator curve analysis of MMAS-8 scores gave an area under the curve [95% CI] of 0.85 [0.77–0.92], (p < 0.0001): using a cut off of <6, the MMAS-8 score has a sensitivity of 94% and a specificity of 64% to predict thiopurine non-adherence. Non-adherent patients were significantly more likely than adherent patients to require changes to medication (44% vs 21%; p = 0.02), hospital admission (22% vs 3%; p < 0.01) and surgery (17% vs 3%; p = 0.01) within the subsequent 6 months of follow up.
Nonadherence to thiopurines in IBD patients is more common in adolescents than adults, is associated with psychological distress and impacts significantly on the prognosis and management. MMAS-8 score is an inexpensive and efficient way to monitor non adherence.