Authors: Juraj Secnik, Hong Xu, Emilia Schwertner, Niklas Hammar, Michael Alvarsson, Bengt Winblad, Maria Eriksdotter, Sara Garcia‐Ptacek, Dorota Religa
Published: 2020-12-07
DOI: 10.1002/alz.039772
Source: Full article
AbstractBackgroundCurrently, there is no evidence‐based approach to treatment of diabetes mellitus (DM) when dementia is present. To reflect current clinical practice, we aimed to determine the differences in antidiabetic drug usage among DM patients with and without dementia.MethodIn this open‐cohort study we used five national Swedish registers to extract patients with DM & dementia and compared them to patients with DM who were dementia‐free. Overall, 10,470 DM & dementia patients registered to the Swedish Dementia Registry (until October 2018) were propensity‐score matched to 10,470 DM & dementia‐free controls originating from the Total Population Register. Information on comorbidities, antidiabetic and other medication and mortality originated from the Patient Register, Prescribed Drug Register and Cause of Death Register, respectively. Matching characteristics included index date (date of dementia diagnosis), age, sex, comorbidity score, education, DM type (excluding type 1 DM), DM duration and cardiovascular, psychotropic and dementia medication. Competing‐risk regression models (death as competing risk) were used to evaluate whether dementia status may alter the probability of antidiabetic drug usage defined using the date of first dispensation after dementia diagnosis (or index date) in patients without history of usage. Medications of interest were insulin, metformin, sulfonylurea derivates (SU), thiazolidinediones and more modern medication like dipeptidyl‐peptidase‐4 inhibitors (DPP‐4i), glucagon‐like peptid‐1 agonists (GLP‐1a) and sodium‐glucose cotransporter‐2 inhibitors (SGLT‐2i). We also assessed the likelihood of treatment with other antidiabetic drugs in addition to metformin as a proxy for treatment intensification.ResultThe dementia and DM group was significantly more likely to be treated with insulin (sHR 1.31 [95% CI 1.21‐1.43]) and less likely to receive DPP‐4i (0.79 [0.72‐0.88]), GLP‐1a (0.50 [0.40‐0.63]) and SGLT‐2i (0.42 [0.34‐0.52]). Dementia patients were also more likely to receive further medication in addition to metformin (1.08 [1.01‐1.15]).ConclusionsIn comparison to dementia‐free controls, dementia patients were less likely to receive modern antidiabetic drugs, with simultaneous higher likelihood of being prescribed insulin.