Marco Boonstra

195 Figure 6.1 Markov model structure MODEL PARAMETERS The main parameters used in the model regarded the transition probabilities to subsequent CKD stages, mortality, quality of life, and costs. Transition probabilities and mortality CKD transition probabilities for the control group were retrieved from the literature (Table 6.1). For the intervention group, these probabilities were multiplied by the effect of GoYK on CKD progression. We estimated this effect by synthesizing the effect of GoYK on hypertension (from the GoYK study) and the effect of hypertension on CKD progression (from medical literature). More details on this estimation can be found in Supplementary File 1. We assumed the effect of the intervention would decrease by 12.5% per year,[20] but return to its full effect every five years, when the intervention would be repeated. The effect of GoYK on mortality was derived from published mortality risks. We used Dutch age-dependent all-cause mortality in the general population for CKD 1/2 [21] and multiplied it by the relative mortality risk to obtain the mortality risk for CKD 3/4[21-23]. The same was done to obtain mortality risks for ESRD, but the ESRD relative mortality risk was adjusted based on age, given that the differences in mortality risk between ESRD and the general population vary with age[24]. The impact of these estimations was investigated in sensitivity analyses. All input data included in the model can be found in Table 6.1. CKD stage 1 or 2 CKD stage 3 or 4 Death ESRD CKD: chronic kidney disease, ESRD: end-stage renal disease

RkJQdWJsaXNoZXIy MTk4NDMw