Electronic Health Record Risk-Stratification Tool Reduces Venous Thromboembolism Events in Surgical Patients

Safety of postdischarge extended venous thromboembolism prophylaxis after hepatopancreatobiliary surgery

BACKGROUND

The risk of venous thromboembolism (VTE) after hepatopancreatobiliary (HPB) surgery is high. Extended postdischarge prophylaxis in this patient population has been controversial. This study aimed to examine the safety of postdischarge extended VTE prophylaxis in patients at high risk of VTE events after HPB surgery.

METHODS

Adult patients risk stratified as very high risk of VTE who underwent HPB operations between 2014 and 2020 at a quaternary care center were included. Patients were matched 1:2 extended VTE prophylaxis to the control group (patients who did not receive extended prophylaxis). Analyses compared the proportions of adverse bleeding events between groups.

RESULTS

A total of 307 patients were included: 103 in the extended prophylaxis group and 204 in the matched control group. Demographics were similar between groups. More patients in the extended VTE prophylaxis group had a history of VTE (9% vs 3%; P = .045). There was no difference in bleeding events between the extended VTE prophylaxis and the control group (6% vs 2%; P = .091). Of the 6 patients with bleeding events in the VTE prophylaxis group, 5 had gastrointestinal (GI) bleeding, and 1 had hemarthrosis. Of the 4 patients with bleeding events in the control group, 1 had intra-abdominal bleeding, 2 had GI bleeding, and 1 had intra-abdominal and GI bleeding.

CONCLUSION

Patients discharged with extended VTE prophylaxis after HPB surgery did not experience more adverse bleeding events compared with a matched control group. Routine postdischarge extended VTE prophylaxis is safe in patients at high risk of postoperative VTE after HPB surgery.

Prediction of Venous Thromboembolism in Diverse Populations Using Machine Learning and Structured Electronic Health Records

BACKGROUND

Venous thromboembolism (VTE) is a major cause of morbidity and mortality worldwide. Current risk assessment tools, such as the Caprini and Padua scores and Wells criteria, have limitations in their applicability and accuracy. This study aimed to develop machine learning models using structured electronic health record data to predict diagnosis and 1-year risk of VTE.

METHODS

We trained and validated models on data from 159 001 participants in the Mount Sinai Data Warehouse. We then externally tested them on 401 723 participants in the UK Biobank and 123 039 participants in All of Us. All data sets contain populations of diverse ancestries and clinical histories. We used these data sets to develop small, medium, and large models with increasing features on a range of optimizing portability to maximizing performance. We make trained models publicly available in click-and-run format at https://doi.org/10.17632/tkwzysr4y6.6.

RESULTS

In the holdout and external test sets, respectively, models achieved areas under the receiver operating characteristic curve of 0.80 to 0.83 and 0.72 to 0.82 for VTE diagnosis prediction and 0.76 to 0.78 and 0.64 to 0.69 for 1-year risk prediction, significantly outperforming the Padua score. Models also demonstrated robust performance across different VTE types and patient subsets, including ethnicity, age, and surgical and hospitalization status. Models identified both established and novel clinical features contributing to VTE risk, offering valuable insights into its underlying pathophysiology.

CONCLUSIONS

Machine learning models using structured electronic health record data can significantly improve VTE diagnosis and 1-year risk prediction in diverse populations. Model probability scores exist on a continuum, affecting mortality risk in both healthy individuals and VTE cases. Integrating these models into electronic health record systems to generate real-time predictions may enhance VTE risk assessment, early detection, and preventative measures, ultimately reducing the morbidity and mortality associated with VTE.

Standardized trauma intake form with clinical decision support prompts improves care and reduces mortality for seriously injured patients in non-tertiary hospitals in Ghana: stepped-wedge cluster randomized trial

BACKGROUND

The WHO Trauma Care Checklist improved key performance indicators (KPIs) of trauma care at tertiary hospitals. A standardized trauma intake form (TIF) with real-time clinical decision support prompts was developed by adapting the WHO Trauma Care Checklist for use in smaller low- and middle-income country hospitals, where care is delivered by non-specialized providers and without trauma teams. This study aimed to determine the effectiveness of the TIF for improving KPIs in initial trauma care and reducing mortality at non-tertiary hospitals in Ghana.

METHODS

A stepped-wedge cluster randomized trial was conducted by stationing research assistants at emergency units of eight non-tertiary hospitals for 17.5 months to observe management of injured patients before and after introduction of the TIF. Differences in performance of KPIs in trauma care (primary outcomes) and mortality (secondary outcome) were estimated using generalized linear mixed regression models.

RESULTS

Management of 4077 injured patients was observed (2067 before TIF introduction, 2010 after). There was improvement in 14 of 16 primary survey and initial care KPIs after TIF introduction. Airway assessment increased from 72.9 to 98.4 per cent (adjusted OR 25.27, 95 per cent c.i. 2.47 to 258.94; P = 0.006) and breathing assessment from 62.1 to 96.8 per cent (adjusted OR 38.38, 4.84 to 304.69; P = 0.001). Documentation of important clinical data improved from 52.4 to 76.7 per cent (adjusted OR 2.14, 1.17 to 3.89; P = 0.013). The mortality rate decreased from 17.7 to 12.1 per cent among 302 patients (186 before, 116 after) with impaired physiology on arrival (hypotension or decreased level of consciousness) (adjusted OR 0.10, 0.02 to 0.56; P = 0.009).

CONCLUSION

The TIF improved overall initial trauma care and reduced mortality for more seriously injured patients.

REGISTRATION NUMBER

NCT04547192 (http://www.clinicaltrials.gov).