Exploring risk factors of non-adherence to immunosuppressive medication in kidney transplant recipients : improving methodology & reorienting research goals

摘要: 8.1. Background and aim of the research program Non-adherence to immunosuppressive therapy is an important issue in kidney transplant patients. About 20% kidney transplant patients are non-adherent to the regimen. Non-adherence contributes late acute rejection episodes 16% graft losses, results a decreased number of quality adjusted life years. A strategy increase long-term successful outcome after transplantation identify at risk for non-adherence target them for preventive adherence enhancing interventions. Comprehensive on risk factors addressing socio-economic, patient-, condition-, therapy-, and health care system/worker-related factors lacking. Especially care-related risk understudied. Moreover, existing studies hampered by of methodological shortcomings. An shortcoming that accurate measurement methods detecting non-adherence, such as electronic monitoring (EM) rarely adopted. EM, currently most sensitive assessment method, uses microchip technology register date time openings pill bottle. Although EM’s superior sensitivity detect makes it potential gold standard of adherence assessment, lack thorough validation well use of appropriate statistical methods multivariable and/or longitudinal data analysis EM data, hinder progress field. The main purpose this program was determine prevalence medication transplant patients. As additional purpose, we aimed improve validity EM measurement mapping assumptions underlying correct EM measurement. We tested these patients. 8.2. Methods 8.2.1. Prevalence non-adherence To study immunosuppressives, we conducted prospective study, which measured over 3- month period 250 adult renal sampled from two outpatient transplant centers Switzerland. We calculated prevalences and expressed percentage prescribed doses taken (taking adherence), the percentage days with dosing (dosing interdose intervals not exceeding 25% interval (timing and the drug holidays per 100 (>48h no intake if once; >24h twice daily intake). Selected were related (e.g. of transplantations, aids, symptom occurrence distress), patient related self-efficacy, beliefs, coping styles, busyness, behaviors), condition depression, substance use), system/team related regularity follow up). Testing occurred simple mixed logistic regression analysis, using sequence daily binary data. Factors significant after correction multiple testing entered into multiple mixed model. Because EM-study designed extensively investigate system or team-related factors, performed meta-analysis look whether differed between continents/countries. This metaanalysis on individual patient pooled three US (n=1563), Netherlands (n=85), Belgium (n=187) and Switzerland (n=342). Adherence Siegal scale, self-report instrument measuring immunosuppressives. Patients were categorized they reported have missed dose of immunosuppression last 4 weeks. Data analysed mixed logistic center random effect continent/country fixed effects, while controlling several demographical clinical characteristics the included samples. 8.2.2. Validation assessment To measurement, summarized evidence on processes may bias assessment. Unbiased requires fulfillment four assumptions, being (1) correctly functioning equipment, (2) correspondence EM-bottle actual dose, (3) absence influence patient’s normal behavior, (4) sample representativeness. We examined above mentioned sample of 250 whose EM. More specifically, determined non-functioning systems, (2) examined impact patient-reported discrepancies cap and actual intakes prevalence, (3) explored whether non-adherence increased started (4) screened differences between participating who refused participate or dropped out study. 8.3. Results Mean taking, dosing, timing 98%, 96%, 93%, 1.1 days, respectively. Variables associated nonadherence were: higher self-reported (OR= 3.08; 95%CI: 1.69-5.61), no usage pillbox 0.31; 0.16-0.61), male gender 0.46; 95%CI: 0.26-0.81), lower self-efficacy 0.49; 0.22-1.07). Furthermore, a gradually declining could be observed Monday Sunday (OR= 1.04; 1.02-1.07). The examining continents/countries showed to immunosuppressives U.S. Europe 19.3% 13.2.%, respectively The higher confirmed multiple logistic (OR=1.78; 1.10-2.89). non-adherence differed (16%) (14.1%) (OR=0.27; 95% CI: 0.09-0.80) (11.4%) (OR=0.17; CI: 0.0- 0.42). The all fulfilled: one malfunctioned, (2) mismatches bottle openings 62% (n=155), (3) non-adherence increased during initial monitoring, primarily during the first 5 weeks, indicating had intervention effect. The caused 5- week minimal. bottle openings larger, but minimized correcting downloaded selfreports (i.e., guidelines notes made the patient ingestions). 8.4. Conclusions This Its contribution literature lies fact comprehensive number of including neglected system factors, been explored, improvements methodological approach proposed. The profile factor appearing final suggest forgetfulness a major driver non-adherence. system might also on individual suggested found differences of non-adherence European among patients. These findings change focus transplant population. Methodological put forward throughout primarily concern measurement behavior Novel techniques are proposed allow multivariate inclusion time-varying variables models. Besides, that, although assumptions valid violated, can certain extent prevented incorrect omitting analysis.