Fig. 3.

Sensitivity analyses. For both the analysis on (a) diabetes and (b) prediabetes we performed several sensitivity analyses, using the fully adjusted model (model 3). In the first analysis, we excluded all participants with an eGFR below 60 ml min⁻¹ [1.73 m²]⁻¹. In the second analysis, we excluded all participants with hypomagnesaemia or hypermagnesaemia. For the analysis on prediabetes we performed an additional sensitivity analyses to study if misclassified prediabetes cases (i.e. those without a prediabetes diagnosis date) could have influenced our results