Implementation of a clinical decision support tool in the electronic health record potentially increased the percentage of adequately prescribed LMWH thromboprophylaxis in non-surgical high-risk patients from 78% to 91%, though this effect diminished after adjustment for pre-intervention trends and autocorrelation.
Key Findings
Results
The percentage of adequately prescribed LMWH thromboprophylaxis increased from 78% pre-intervention to 91% post-intervention following CDS tool implementation.
The crude increase was 13% (95% CI: 6%–20%)
400 total patients were included: 200 pre-intervention and 200 post-intervention
Patients were non-surgical hospitalized adults (≥18 years) at high risk for VTE based on the Padua Prediction Score (PPS)
An interrupted time series analysis was used to assess both crude and adjusted pre- and post-intervention differences
Results
After adjustment for the pre-intervention slope and autocorrelation, the effect of the CDS tool on adequate LMWH prescription diminished and became statistically non-significant.
The adjusted difference was 8% (95% CI: -4%–19%), which includes zero, indicating uncertainty about the true effect
The interrupted time series analysis accounted for potential time trends and autocorrelation
The attenuation from 13% to 8% suggests a pre-existing positive trend in adequate prescribing before the intervention
Background
Adherence to thromboprophylaxis guidelines is generally low in non-surgical hospitalized patients, who are at increased risk of venous thromboembolism.
Non-surgical hospitalized patients have an increased risk of developing VTE
The Padua Prediction Score (PPS) was used to identify high-risk patients
Low-molecular-weight heparins (LMWHs) can reduce VTE risk, but guideline adherence is noted to be generally low in this population
Methods
A validated clinical decision support tool integrated into the electronic health record was used as the intervention to improve thromboprophylaxis prescribing.
The CDS tool was described as 'validated'
Data on PPS and thromboprophylaxis were collected at different time points before and after implementation
The study used a stepwise approach to describe the percentage of adequately prescribed LMWH pre- and post-intervention
Conclusions
The authors concluded that CDS tools may represent an efficient and sustainable intervention to improve healthcare quality as more data becomes available in electronic health records.
The baseline rate of adequate prescribing was already high at 78% pre-intervention
The authors noted that 'with increasing amounts of data available in electronic health records, CDS tools might be an efficient and sustainable intervention to improve healthcare quality'
The study suggests CDS tools could further increase an already high rate of adequate prescribing
What This Means
This research suggests that adding a computer-based decision support tool to hospital electronic health records may help doctors more consistently prescribe blood clot prevention medication (low-molecular-weight heparins) to hospitalized non-surgical patients who are at high risk for dangerous blood clots (venous thromboembolism). Before the tool was introduced, 78% of high-risk patients were receiving appropriate preventive treatment; afterward, this rose to 91% — a raw increase of 13 percentage points. However, when the researchers accounted for the fact that prescribing rates were already gradually improving before the tool was introduced, the estimated benefit of the tool itself dropped to 8 percentage points and became statistically uncertain (meaning the result could plausibly reflect chance or the pre-existing trend rather than the tool's effect).
The study involved 400 hospitalized adult patients (200 before and 200 after the intervention) whose clot risk was measured using a standardized scoring system called the Padua Prediction Score. The decision support tool was embedded directly in the electronic health record system already used by clinicians, making it a relatively low-effort intervention to implement.
This research suggests that digital decision support tools integrated into hospital computer systems could be a practical and scalable way to close gaps between clinical guidelines and actual prescribing behavior. The findings also highlight the importance of accounting for existing trends when evaluating whether a new tool is truly responsible for observed improvements, as hospitals may already be on an upward trajectory in care quality before an intervention is introduced.
van Gosliga F, Pals D, van Ojik A, van Roon E. (2026). Implementation of a clinical decision support tool to improve the adequate prescription of low-molecular-weight heparins in non-surgical patients.. PloS one. https://doi.org/10.1371/journal.pone.0350017