#3028 Predictive validity of the Basel Assessment of Adherence to Immunosuppressive Medication Scale (BAASIS©) after renal transplantation

Abstract

Abstract Background and Aims Adherence towards immunosuppressive medication is crucial for allograft survival after renal transplantation. However, its assessment in daily clinical practice remains challenging. The BAASIS© questionnaire, is a validated self-report instrument to detect non-adherence. However, its predictive validity for clinical relevant consequences of non-adherence needs to be established. Method The prospective AdTorque-trial (DRKS00026674) includes a consecutive cohort of 226 adult kidney graft recipients transplanted at the Medical University of Vienna between 01/2018 and 12/2019. Medication adherence was monitored up to two years by a multi-component assessment including the BAASIS©, electronic drug monitoring, pharmacy refill records, tacrolimus trough level, records by medical personal, individual evaluation by a transplant psychologist and concurrent immune monitoring by the quantification of Torque Teno virus load. The BAASIS© was applied at three-month intervals during the first year and at 24 months during routine follow-up visits. The adherence phases implementation and persistence were assessed by three items (i.e. taking, timing and dose-reduction) and one item of the BAASIS©. Non-adherence was defined by YES on any of these items. The primary clinical outcome was the occurrence of biopsy-proven allograft rejection during a maximum clinical follow-up of up to 4 years after transplantation. Herein we present the preliminary analysis of the multimodal monitoring focusing on the BAASIS© questionnaire. Results Of all 226 recipients (33% female, median age 57 years), 153 completed the adherence monitoring for 2 years, resulting in a total of 973 BAASIS© assessments. Non-adherence was reported at least once by 124 (55%) recipients and 67 recipients (30%) revealed non-adherence multiple times. The proportion of non-adherent recipients increased from 11% to 31% within the first three months and was between 27% to 32% at subsequent visits from month 6 to 24 post-transplant. Issues in the punctual timing of dose taking was the most frequent cause for non-adherence. During the clinical follow-up of 44 months (median, IQR 12-48), non-adherent recipients had a significantly higher rate of biopsy-proven rejection than adherent recipients (25%, n = 30 vs. 7%, n = 7; p < 0.001). A time-dependent cox regression model showed that non-adherence assessed by the BAASIS© predicted an increased risk for allograft rejection (HR 2.87, 95% CI 1.47-5.6). Conclusion Medication non-adherence was substantially reported already in early phases post-transplant and was associated with an increased risk for allograft rejection. Our analysis demonstrated the predictive validity of the BAASIS©, a self-report instrument which can easily be integrated in daily transplant practice. Regular adherence assessments might identify patients at risk for poor clinical outcome due to medication non-adherence and provide rationale for intervention.