CoChat
Sign Up
Cited by 2Machine Learning in HealthcareArtificial Intelligence in HealthcareEmergency and Acute Care Studies

Development and evaluation of a machine learning prediction model for short-term mortality in patients with diabetes or hyperglycemia at emergency department admission

Open access
Wändell, Per, Wierzbicka, Marcelina, Sigurdsson, Karolina, Olofsson, Anna, Wachtler, Caroline, Wessman, Torgny, Melander, Olle, Ekelund, Ulf +3 more·2025·Cardiovascular Diabetology
DOI 10.1186/s12933-025-02954-8PMC PMC12492943
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: Patients with diabetes admitted to emergency care face a higher risk of complications, including prolonged hospital stays, admissions to the intensive care unit and mortality. Aim: To develop a machine learning (ML) model to predict 30-day mortality in patients with diabetes admitted to the emergency department (ED). Design and setting: A cohort study utilizing data from all nine ED’s in Region Skåne 2017 to 2018. Totally 74,611 patient visits, representing 34,280 unique patients aged > 18 years with diabetes or hyperglycemia (glucose were > 11 mmol/L). The analysis focused on four groups, men and women aged 40–69 and ≥ 70 years. Methods: Stochastic gradient boosting was employed to develop a model predicting 30-day mortality. Variable importance was assessed using normalized relative influence (NRI) scores. Variables in certain hospitals were used to train the models, and the models were tested in other hospitals. Results: Key predictors included laboratory values (pH, base excess, pCO 2, standard bicarbonate, oxygen saturation, lactate, CRP, and leukocytes), as well as age, triage category, and time to doctor consultation. The sensitivity of the models ranged from 86–97%, the specificity from 86–94%, and accuracy between 86% and 94%. The area under the curve (AUC) ranged from 0.84 to 0.93 and Cohen’s kappa ranged from 0.34 to 0.45. Positive predictive values accurately identified mortality in 23% to 37% of cases across the four groups. Conclusions: A machine learning model based on routinely collected data in the ED accurately predicted 30-day mortality with high specificity and sensitivity. This approach shows promise in identifying high-risk patients requiring close monitoring and timely interventions.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/s12933-025-02954-8

Cites

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Machine learning and artificial intelligence approaches have demonstrated growing promise in diabetes risk prediction, complication detection and healthcare utilization forecasting.

    “More recently, machine learning and artificial intelligence approaches have demonstrated growing promise in diabetes risk prediction, complication detection and healthcare utilization forecasting [21–28].”

    From Machine learning-driven risk prediction for post-hospitalization diabetes case management: Integrating clinical and social determinants of health (2026) · cites this paper as [27]

    The Introduction says ML has shown promise in emergency-department outcome prediction, including complications and length of stay. The Abstract Conclusions report accurate prediction of 30-day mortality among patients with diabetes. The paper does not demonstrate prediction of diabetes risk or healthcare utilization in this population; it discusses length-of-stay prediction as prior work in broader emergency settings. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

The citation lists haven’t been fetched yet

References 36 · Cited by 2

Research works printed but not linked: 38

  1. [1]Worldwide trends in diabetes. Since 1980: a pooled analysis of 751 population-based studies with 4.4 million participants. Lancet. 2016;387(10027):1513–30. PMC5081106 27061677
  2. [2]Menke A, Casagrande S, Geiss L, Cowie CC. Prevalence of and trends in diabetes among adults in the United States, 1988–2012. JAMA. 2015;314(10):1021–9. 26348752 · DOI · Paper page
  3. [3]Charfen MA, Ipp E, Kaji AH, Saleh T, Qazi MF, Lewis RJ. Detection of undiagnosed diabetes and prediabetic states in high-risk emergency department patients. Acad Emerg Med. 2009;16(5):394–402. 19302369 · DOI
  4. [4]Egi M, Furushima N, Makino S, Mizobuchi S. Glycemic control in acute illness. Korean J Anesthesiol. 2017;70(6):591–5. 29225740 PMC5716815 · DOI
  5. [5]Finney SJ, Zekveld C, Elia A, Evans TW. Glucose control and mortality in critically ill patients. JAMA. 2003;290(15):2041–7. 14559958 · DOI
  6. [6]Investigators N-SS, Finfer S, Chittock DR, Su SY, Blair D, Foster D, Dhingra V, Bellomo R, Cook D, Dodek P, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med. 2009;360(13):1283–97. 19318384 · DOI · Paper page
  7. [7]Chang CH, Wang JL, Wu LC, Chuang LM, Lin HH. Diabetes, glycemic control, and risk of infection morbidity and mortality: a cohort study. Open Forum Infect Dis. 2019;6(10):ofz358. 31660337 PMC6765350 · DOI
  8. [8]Wandell P, Carlsson AC, Larsson A, Melander O, Wessman T, Arnlov J, Ruge T. The association between BMI and 90-day mortality in patients with and without diabetes seeking care at the emergency department. Ups J Med Sci. 2021;126. PMC8494247 34650780
  9. [9]Jansson SP, Fall K, Brus O, Magnuson A, Wandell P, Ostgren CJ, Rolandsson O. Prevalence and incidence of diabetes mellitus: a nationwide population-based pharmaco-epidemiological study in Sweden. Diabet Med. 2015;32(10):1319–28. 25662570 · DOI
  10. [10]Wirehn AB, Ostgren CJ, Carstensen JM. Age and gender differences in the impact of diabetes on the prevalence of ischemic heart disease: a population-based register study. Diabetes Res Clin Pract. 2008;79(3):497–502. 18006174 · DOI
  11. [11]Rosengren A, Edqvist J, Rawshani A, Sattar N, Franzen S, Adiels M, Svensson AM, Lind M, Gudbjornsdottir S. Excess risk of hospitalisation for heart failure among people with type 2 diabetes. Diabetologia. 2018;61(11):2300–9. 30094466 PMC6182656 · DOI
  12. [12]Tancredi M, Rosengren A, Svensson AM, Kosiborod M, Pivodic A, Gudbjornsdottir S, Wedel H, Clements M, Dahlqvist S, Lind M. Excess mortality among persons with type 2 diabetes. N Engl J Med. 2015;373(18):1720–32. 26510021 · DOI
  13. [13]Uppal TS, Chehal PK, Fernandes G, Haw JS, Shah M, Turbow S, Rajpathak S, Narayan KMV, Ali MK. Trends and variations in emergency department use associated with diabetes in the US by sociodemographic factors, 2008–2017. JAMA Netw Open. 2022;5(5):e2213867. 35612855 PMC9133946 · DOI
  14. [14]Wang J, Li J, Wen C, Liu Y, Ma H. Predictors of poor glycemic control among type 2 diabetes mellitus patients treated with antidiabetic medications: a cross-sectional study in China. Med (Baltim). 2021;100(43):e27677. PMC8556014 34713865 · DOI
  15. [15]Dhatariya K, Corsino L, Umpierrez GE. Management of diabetes and hyperglycemia in hospitalized patients. In: Feingold KR, Anawalt B, Blackman MR, Boyce A, Chrousos G, Corpas E, de Herder WW, Dhatariya K, Dungan K, Hofland J South Dartmouth (MA) (eds) Endotext. edn. 2000. 25905318
  16. [16]Raita Y, Goto T, Faridi MK, Brown DFM, Camargo CA Jr., Hasegawa K. Emergency department triage prediction of clinical outcomes using machine learning models. Crit Care. 2019;23(1):64. 30795786 PMC6387562 · DOI
  17. [17]Joseph JW, Leventhal EL, Grossestreuer AV, Wong ML, Joseph LJ, Nathanson LA, Donnino MW, Elhadad N, Sanchez LD. Deep-learning approaches to identify critically ill patients at emergency department triage using limited information. J Am Coll Emerg Physicians Open. 2020;1(5):773–81. 33145518 PMC7593422 · DOI · Paper page
  18. [18]Sanchez-Salmeron R, Gomez-Urquiza JL, Albendin-Garcia L, Correa-Rodriguez M, Martos-Cabrera MB, Velando-Soriano A, Suleiman-Martos N. Machine learning methods applied to triage in emergency services: a systematic review. Int Emerg Nurs. 2022;60:101109. 34952482 · DOI
  19. [19]Naemi A, Schmidt T, Mansourvar M, Naghavi-Behzad M, Ebrahimi A, Wiil UK. Machine learning techniques for mortality prediction in emergency departments: a systematic review. BMJ Open. 2021;11(11):e052663. 34728454 PMC8565537 · DOI
  20. [20]Fernandes M, Vieira SM, Leite F, Palos C, Finkelstein S, Sousa JMC. Clinical decision support systems for triage in the emergency department using intelligent systems: a review. Artif Intell Med. 2020;102:101762. 31980099 · DOI
  21. [21]Zwawi A, Wessman T, Wandell P, Melander O, Carlsson AC, Ruge T. A simplified clinical frailty scale predicts mortality in emergency department patients with acute dyspnea. Geroscience 2025. PMC13356002 40931291
  22. [23]Lee S, Kisiel MA, Lindberg P, Wheelock AM, Olofsson A, Eriksson J, Bruchfeld J, Runold M, Wahlstrom L, Malinovschi A, et al. Using machine learning involving diagnoses and medications as a risk prediction tool for post-acute sequelae of COVID-19 (PASC) in primary care. BMC Med. 2025;23(1):251. 40307834 PMC12044741 · DOI
  23. [24]Wallensten J, Wachtler C, Bogdanovic N, Olofsson A, Kivipelto M, Jonsson L, Petrovic P, Carlsson AC. Machine learning to detect alzheimer’s disease with data on drugs and diagnoses. J Prev Alzheimers Dis. 2025;12(5):100115. 40057461 PMC12184014 · DOI
  24. [25]Friedman JH. Greedy function approximation: a gradient boosting machine. Ann Statist. 2001;29(5):1189–232. · DOI · Paper page
  25. [26]Brekke IJ, Puntervoll LH, Pedersen PB, Kellett J, Brabrand M. The value of vital sign trends in predicting and monitoring clinical deterioration: a systematic review. PLoS ONE. 2019;14(1):e0210875. 30645637 PMC6333367 · DOI
  26. [27]Chen P, Wang B, Zhao L, Ma S, Wang Y, Zhu Y, Zeng X, Bai Z, Shi B. Machine learning for predicting intrahospital mortality in ST-elevation myocardial infarction patients with type 2 diabetes mellitus. BMC Cardiovasc Disord. 2023;23(1):585. 38012550 PMC10683359 · DOI
  27. [28]Li C, Zhang Z, Ren Y, Nie H, Lei Y, Qiu H, Xu Z, Pu X. Machine learning based early mortality prediction in the emergency department. Int J Med Inf. 2021;155:104570. 34547624 · DOI
  28. [29]Lauque D, Khalemsky A, Boudi Z, Ostlundh L, Xu C, Alsabri M, Onyeji C, Cellini J, Intas G, Soni KD et al. Length-of-Stay in the emergency department and In-Hospital mortality: a systematic review and meta-analysis. J Clin Med. 2022;12(1). PMC9821325 36614835
  29. [30]Jones S, Moulton C, Swift S, Molyneux P, Black S, Mason N, Oakley R, Mann C. Association between delays to patient admission from the emergency department and all-cause 30-day mortality. Emerg Med J. 2022;39(3):168–73. 35042695 · DOI
  30. [31]Tazmini K, Nymo SH, Louch WE, Ranhoff AH, Oie E. Electrolyte imbalances in an unselected population in an emergency department: a retrospective cohort study. PLoS ONE. 2019;14(4):e0215673. 31022222 PMC6483356 · DOI
  31. [32]Ford W, Self WH, Slovis C, McNaughton CD. Diabetes in the emergency department and hospital: acute care of diabetes patients. Curr Emerg Hosp Med Rep. 2013;1(1):1–9. 23997994 PMC3756609 · DOI
  32. [33]Berbudi A, Rahmadika N, Tjahjadi AI, Ruslami R. Type 2 diabetes and its impact on the immune system. Curr Diabetes Rev. 2020;16(5):442–9. 31657690 PMC7475801
  33. [34]Erdur A, Guven R, Can D, Gurkan TT, Ak E, Avci A. Prognostic importance of lactate and blood gas parameters in predicting mortality in patients with critical malignancies. Ethiop J Health Sci. 2023;33(2):255–62. 37484184 PMC10358382 · DOI
  34. [35]Schork A, Moll K, Haap M, Riessen R, Wagner R. Course of lactate, pH and base excess for prediction of mortality in medical intensive care patients. PLoS ONE. 2021;16(12):e0261564. 34929006 PMC8687550 · DOI
  35. [36]Walther LH, Lassen AT, Mogensen CB, Christensen EF, Mikkelsen S. Prehospital blood gas analyses in acute patients treated by a ground-based physician-manned emergency unit: a cohort study. Scand J Trauma Resusc Emerg Med. 2023;31(1):102. 38115069 PMC10729417 · DOI
  36. [37]Hökenek NM, Seyhan AU, Erdogan MÖ, Tekyol D, Yilmaz E, Korkut S. Evaluation of blood gas analysis as a mortality predictor. South Clin Ist Euras. 2019;30(3):228–31.
  37. [38]Pongmanee W, Vattanavanit V. Can base excess and anion gap predict lactate level in diagnosis of septic shock? Open Access Emerg Med. 2018;10:1–7. 29302195 PMC5741987 · DOI
  38. [39]Wandell P, Ljunggren G, Carlsson AC. The most common diagnoses in primary care, and changes over time, in the total population of Stockholm, Sweden. BMC Prim Care. 2025;26(1):235. 40751125 PMC12315263 · DOI

Authors
Wändell, P., Wierzbicka, M., Sigurdsson, K., Olofsson, A., Wachtler, C., Wessman, T., Melander, O., Ekelund, U., Björkelund, A., Carlsson, A. C., & Ruge, T.
Affiliations
  • Lund University (SE)
  • Karolinska Institutet (SE)
  • Region Stockholm (SE)
  • Dalarna University (SE)
Corresponding author
Wändell, Per
ORCID
Venue
Cardiovascular Diabetology
Publisher
Springer Science and Business Media LLC
Published
2025-10-03
Volume
24
Issue
1
Pages
383-383
ISSN
1475-2840 (ISSN-L)
DOI
10.1186/s12933-025-02954-8
PMC
PMC12492943
OpenAlex
W4414777234
Cited by
2
References
36
Citations by year
2026: 2
FWCI
0.80
Type
journal article
Language
en
Open access
Licence
Locations
  • Cardiovascular Diabetology (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Lund University Publications (Lund University) (submittedVersion)
Funders
Karolinska Institute
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Health Sciences › Medicine › Emergency Medicine › Emergency and Acute Care Studies
Keywords
diabetes mellitus, hyperglycemia, 30-day mortality prediction, emergency department admission, stochastic gradient boosting, routinely collected clinical data, lactate, C-reactive protein, bicarbonate, oxygen saturation, triage categories, high-risk patient identification
MeSH terms
Machine Learning, Adult, Aged, Blood Glucose, Decision Support Techniques, Diabetes Mellitus, Emergency Service, Hospital, Female, Humans, Hyperglycemia, Male, Middle Aged, Patient Admission, Predictive Value of Tests, Prognosis, Retrospective Studies, Risk Factors, Time Factors, Reproducibility of Results, Biomarkers, Hospital Mortality, Risk Assessment

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review