Identifying subtypes of type 2 diabetes mellitus with machine learning: development, internal validation, prognostic validation and medication burden in linked electronic health records in 420 448 individuals

Does the introduction clearly articulate the research question and provide adequate context? Is the study's significance within the field convincingly established?

Does the study design test the authors' hypothesis? Are the methods described in enough detail for independent replication?

Were the study's findings analyzed and interpreted reasonably? Is the resulting data open and auditable? Are the figures of high resolution, describe data appropriately and adequately labelled? Please provide comments - positive or negative on each figure including supplementary ones.

Do the results support the authors’ conclusions? Are there any alternative interpretations of the study's findings that the authors should have considered?