This scale is a self-report instrument that consists of 4 questions on the taking and timing of medication, drug holidays, reduction of the dose, and persistence over the past month (Table 2). An affirmative answer to any of the first dilution calculator 4 questions results in assignment to the nonadherent group. This scoring is intentionally strict due to an assumption of the underreporting of nonadherence. Patients also rated their own adherence using a visual analogue scale from 0% (medication never taken as prescribed) to 100% (medication always taken as prescribed). The BAASIS measure was selected as it is short, reliable, valid, and sensitive to both timing and taking which is of particular importance for the immunosuppressive regime after kidney transplantation [20].

A number of studies have demonstrated support for the validity of both parts of the instrument [22, 23]. Specificity and sensitivity of the visual analogue scale have been shown to be high [22]. Table 2 Adherence 6 weeks after transplantation as measured with the BAASIS?-interviewa (n = 113). Secondly, we calculated patient intraindividual variability in the pharmacokinetics of immunosuppressive medication, in this case tacrolimus (Prograft) [24]. Whole blood tacrolimus concentrations in different measurements over time within patients were used to calculate intraindividual variability. Patients with a high intrapatient variability have tacrolimus concentrations that are often outside the therapeutic window. Underexposure may lead to immune activation, and overexposure can result in CNI-induced nephrotoxicity.

Both could affect long-term outcome. Borra et al. [24] showed that high intraindividual variability in the pharmacokinetics of tacrolimus leads to reduced graft survival. One of the most likely causes for intrapatient variability is medication nonadherence. To calculate the intrapatient variability in tacrolimus concentrations we used the method previously described by Borra et al. [24]. For the clinical endpoints, we collected information about rejection (yes/no) and graft failure (yes/no) two years after transplantation. 2.3. Statistical Analysis Independent t-tests and chi-squared analyses were conducted to test differences between responders and nonresponders. As the BAASIS overall rating scale was negatively skewed, a Mann-Whitney test was used to test the difference on this scale between adherent and nonadherent patients.

One-way ANOVA and chi-squared tests were used to test the association between attitudes and adherence. When cell values were small, Fisher’s exact tests were used to test 2 �� 2 associations. Survival analyses were calculated with Kaplan-Meier and life table. Analyses were carried out using the Statistical Package for GSK-3 Social Sciences, version 20.0. Q-methodological data were analysed using PQMethod 2.11 developed by Schmolck and Atkinson 2002. 3. Results 3.1.