Gastrointestinal bleeding (GIB) is a clinical challenge in kidney failure. The INSPIRE group assessed if machine learning could assist with determining a hemodialysis (HD) patient's 180-day GIB hospitalization risk. Model was developed using adult HD patient data from United States (2017-2020). Patient data was randomly split (50% training, 30% validation, and 20% testing). HD treatments <=180 days before GIB hospitalization were classified as positive observations, and others were negative observations. Datasets were randomly sampled to build an XGBoost model that considered 386 exposures initially and was refined to the top 50 exposures. Unseen testing dataset was used to determine final model performance. Incidence of 180-day GIB hospitalization was 1.18% in the HD population (n=451,579), and 1.16% among patients in the testing dataset (n=27,991). Model showed an area under the curve=0.69, sensitivity=57.9%, specificity=68.9%, accuracy=68.8% and balanced accuracy=63.4%. Exposures with largest effect size per Shapley values were older age (group mean GIB event=68.2 years vs no GIB event=63.4 years), shorter days since last all-cause hospital admission (group mean GIB event=203.2 days vs no GIB event=253.2 days), and higher serum 25-hydroxy (OH) vitamin D levels from most recent lab (group mean GIB event=33.4 ng/mL vs no GIB event=30.5 ng/mL). Other important predictors included lower hemoglobin and iron indices, longer dialysis vintage, and proton pump inhibitor use. Model appears suitable for early detection of GIB event risk in HD, yet prospective testing is needed. The association between higher 25OH vitamin D and GIB events was unexpected and warrants investigation.

J.W.L., S.L., S.C., J.W., A.C.W., Y.J., M.S.G., L.A.U., J.L.H., F.W.M. report being an employee of Fresenius Medical Care. J.W.L., S.C., M.S.G., L.A.U., J.L.H., F.W.M. report having share options/ownership in Fresenius Medical Care. J.W.L., S.C., L.A.U., J.L.H., F.W.M. report being an inventor on patent(s) in the field of dialysis. J.W.L. reports receipt of honorarium from The Lancet and being on the Editorial Board of Frontiers in Physiology and Frontiers in Medicine, Nephrology. L.A.U. reports being an advisory board member for Privacy Analytics Inc. F.W.M. reports directorships in Fresenius Medical Care Management Board and Vifor Fresenius Medical Care Renal Pharma. D.C.W. reports consultancy or honoraria from Amgen, AstraZeneca, Boehringer Ingelheim, Bayer, GlaxoSmithKline, Janssen, Mundipharma, Napp, Mitsubishi, Ono Pharma, and Vifor Fresenius Medical Care Renal Pharma. P.S. reports serving on scientific advisory boards or speaker honoraria for AstraZeneca, REATA, Vifor, Baxter Healthcare, GSK, Pfizer, Invizius, Astellas, Novo Nordisk, and Fresenius Medical Care. J.F. reports consultancy or speaker honoraria from Astellas, AstraZeneca, Bayer, Boehringer, Fresenius Medical Care, and Vifor.

There was no external funding for the work presented in this manuscript. Fresenius Medical Care internally supported the conduct of the analysis and manuscript composition, which included employee salaries and company infrastructure.

I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.

Yes

The details of the IRB/oversight body that provided approval or exemption for the research described are given below:

This project was reviewed and approved by New England Independent Review Board (Needham Heights, MA, United States; Work Oder# 1-1502098-1). It was determined by the Independent Review Board that this analysis was exempt due to deidentification of data and consent was not required per title 45 Code of Federal Regulations part 46.104(d)(4) in the United States. The analysis adhered to the Declaration of Helsinki.

I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.

Yes

I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).

Yes

I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable.

Yes

The dataset used in the presented analyses is not publicly available. The dataset was captured from a private electronic medical record system that is restricted to use by only authorized employees of Fresenius Medical Care. The dataset can be made available upon reasonable request to access the dataset, which would require an agreement to be established between Fresenius Medical Care and an external institution of any applicable requestor.