Impact of an intervention program on drug adherence in patients with ulcerative colitis: Randomized clinical trial

Evaluate the impact of an intervention program in non-adherent patients with ulcerative colitis.

Parallel controlled randomized clinical trial (1:1), approved by the ethics committee (No. 3.068.511/2018) and registered at The Brazilian Clinical Trials Registry (No. RBR-79dn4k). Non-adherent ulcerative colitis patients according to the Morisky-Green-Levine-test were included. Recruitment began in August 2019 until August 2020, with 6-month follow-up. All participants received standard usual care, and additionally the intervention group received educational (video, educational leaflet, verbal guidance) and behavioral interventions (therapeutic scheme, motivational and reminder type short message services). Researchers were blinded for allocation prior to data collection at Visits 1 and 2 (0 and 6 months). Primary outcome: 180-day adherence rate, with relative risk 95%CI. Secondary outcome: 180-day quality of life according to SF-36 domains, using Student's t test. Variables with p<0.20 were selected for regression. Analysis included data from August/2019 to May/2021.

Forty-six and 49 participants were allocated in control and intervention groups, respectively. Two were excluded due to intervention refusal, and 4 and 6 were lost to follow-up in control and intervention groups. There was no post-intervention adherence rate difference, even after adjustment for type of non-adherence (unintentional/both/intentional) as confounder, or if considered as adherent the intervention group participants lost in follow-up. Interventions promoted better quality of life scores even after multivariate analysis for "Pain", when adjusted for ulcerative colitis severity, sex, and marital status (β = 18.352, p = 0.004), "Vitality", when adjusted for ulcerative colitis severity (β = 10.568, p = 0.015) and "Emotional Aspects", when adjusted for disease severity, income, and education (β = 24.907, p = 0.041).

The intervention program was not able to produce a significant medication adherence rate difference between comparative groups, however, there was a significant improvement in quality of life. Study limitations may include: sample size calculated to identify differences of 30%, leading to a possible insufficient power; non blinded participants, exposing the results to the risk of performance bias; outcomes based on self-reported data.