Abstract
Background
Despite various interventions to improve best-practice venous thromboembolism (VTE) prevention measures within hospitals, compliance remains poor. For health services utilising electronic medication management systems (eMMS), implementation of clinical decision support (CDS) tools could address this gap.
Aim
To evaluate whether local implementation of an integrated electronic alert system linked with a computerised physician order entry (CPOE)-based order set for VTE risk assessment within an eMMS improves the rates of timely VTE risk assessment and guideline-compliant VTE prophylaxis prescribing among hospitalised patients.
Method
A retrospective observational study conducted among hospitalised patients pre- and post-implementation of an electronic alert system combined with a CPOE-based order set to prompt VTE risk assessment documentation and VTE prophylaxis prescribing within a single tertiary hospital. Admissions were consecutively screened over 7-day periods before and after implementation for inclusion and assessed for compliance with a local VTE prevention protocol.
Results
Eight hundred and fifty patients (458 pre-intervention, 392 post-intervention) were included for evaluation. Rates of VTE risk assessment documentation within 24 h of admission increased by 29.9% (p < 0.001). Guideline-compliant VTE prophylaxis improved by 10.4% (p < 0.001). Patients with completed VTE risk assessments were significantly more likely to receive guideline-compliant VTE prophylaxis, compared to patients without documented VTE risk assessments (19.3% difference, p < 0.001). After adjusting for demographic differences, the odds of achieving positive outcomes significantly increased across all measures, with adjusted odds ratios ranging from 1.95 to 4.89 (p < 0.001).
Conclusion
Local implementation of CDS featuring CPOE within the eMMS improved rates of VTE risk assessment documentation and guideline-compliant VTE prophylaxis prescribing.
Supplementary Information
The online version contains supplementary material available at 10.1007/s11096-024-01857-0.
Keywords: Decision support systems, Clinical; Electronic health record; Electronic prescribing; Medical order entry systems; Risk assessment; Venous thromboembolism
Impact statements
We demonstrated the potential benefits of an electronic alert system linked to a computerised physician order entry-based order set for improving venous thromboembolism (VTE) prevention measures, providing a scalable and practical model for clinical decision support integration into electronic medication management system workflow.
We advocate for the widespread adoption of documenting VTE risk assessments through computerised physician order entry within electronic medication management systems to enhance visibility of VTE prevention measures prescribed against a documented risk.
Introduction
Deep vein thrombosis (DVT) and pulmonary embolism (PE), collectively known as venous thromboembolism (VTE), is estimated to be one of the leading preventable causes of disability and death worldwide [1]. VTE affects around 10 million people worldwide each year, contributing significantly to the global disease burden [2]. Hospitalisation is known to substantially increase the risk of developing VTE due to multiple risk factors including prolonged immobility, dehydration, critical illness and vascular injury from trauma or surgery [3].
International consensus guidelines endorse the use of prophylactic measures in preventing VTE, and appropriate VTE prevention measures have been shown to reduce the incidence of VTE by up to 70% in medical and surgical patients [4–6]. To support timely identification of at-risk individuals, formal VTE risk assessment tools have been universally adopted to provide standardised criteria for evaluating VTE risk. Introduction of these tools has been shown to correlate with significant reductions in VTE rates, shorter hospital stays and fewer deaths [7–9].
Various methods exist for documenting VTE risk assessments, including dedicated VTE risk assessment tools developed by national healthcare agencies or local health services, patient care plans and specific fields or forms within electronic medical records. However, despite the availability of these tools in addition to evidence-based guidelines, adherence to best practice remains poor, with large multinational studies continually demonstrating that significant proportions of at-risk patients do not receive appropriate thromboprophylaxis [10].
Various strategies have been proposed to improve uptake of VTE prevention measures, ranging from passive strategies such as guideline dissemination and audit feedback, to active interventions such as computer-generated alerts. Extensive literature evaluates the effectiveness of different interventions for improving uptake of thromboprophylaxis within hospital settings [11–14].
Among these strategies, clinical decision support (CDS) tools have shown great potential for enhancing VTE prevention practices [14–18]. A systematic review found that CDS implementation was associated with a significant increase in the proportion of surgical patients prescribed appropriate VTE prophylaxis and a significant decrease in the number of VTE events [14]. Another systematic review found that CDS intervention significantly improved appropriate VTE prophylaxis among medical patients [15]. Additionally, studies from multiple countries confirm the effectiveness of these tools for optimising the use of appropriate thromboprophylaxis following electronic medical record (EMR) integration [16–18].
Another tool often used alongside CDS is computerised physician order entry (CPOE), which provides predefined orders for streamlining medication prescribing. A study conducted in the United States found that introduction of a VTE CDS tool which included CPOE led to a higher proportion of patients being assessed for VTE risk within 24 hours of admission, increased the number of at-risk patients identified, and nearly doubled the proportion of patients receiving individualised VTE care plans [16]. Studies have also shown that implementation of CDS interventions featuring CPOE for VTE prophylaxis orders significantly improved rates of prescribed VTE prophylaxis and significantly reduced incidences of hospital-acquired VTE [19, 20].
This body of evidence proves that combined use of CPOE and CDS can significantly increase appropriate VTE prophylaxis prescribing and reduce VTE-related harm when implemented with EMR systems, however there is a significant literature gap concerning the incorporation of VTE CDS tools into electronic medication management systems (eMMS), despite their widespread use. While the EMR is designed for comprehensive clinical documentation, eMMS are primarily used for medication management, making CDS integration into eMMS more complex. Tailoring CDS to align with medication workflows poses a unique technical challenge for health services looking to adapt these tools for use within their local eMMS. A recent study found that introduction of a CPOE-based CDS tool within the local eMMS led to a significant increase in VTE risk assessment documentation and appropriate VTE prophylaxis prescribing, but more research is needed to explore viable integration methods and enable more widespread adoption of effective VTE prevention measures [21].
Aim
This study aimed to evaluate whether implementation of an integrated electronic alert system linked with a CPOE-based VTE risk assessment order set within an eMMS improves the rates of VTE risk assessment and guideline-compliant VTE prophylaxis prescribing among hospitalised patients within 24 hours of admission.
Ethics approval
This minimal risk study was reviewed and approved by the Cabrini Health Research Governance Office (reference 09–07-04–22, date of approval 23/03/2022).
Method
A retrospective observational study was conducted within a single 508-bed tertiary metropolitan hospital located in Victoria, Australia. The service utilises a hybrid health record system, featuring an eMMS (Medchart; Dedalus, Queensland, Australia) with paper-based medical record including progress notes. The service utilises stratified VTE risk assessment scores (low, intermediate and high) for classifying and documenting VTE risk. Prior to May 2022, VTE risk assessments were documented within paper progress notes or paper medical record. An indiscriminate pop-up alert would appear for the first 48 hours of admission upon every attempt to access the eMMS, prompting prescribers to document a VTE risk assessment and prescribe VTE prophylaxis if indicated.
Alert system intervention
In May 2022, the health service implemented an integrated electronic alert system linked to a CPOE-based order set for VTE risk assessment within the eMMS. The alert system is rule-based and triggers on the order confirmation screen of the prescriber workflow if a VTE risk assessment has not already been prescribed in the eMMS or is not being co-prescribed with the current prescription orders. It is linked to the local VTE prevention protocol and prompts the prescriber to select one of three pre-defined protocolised orders (‘High VTE risk’, ‘Intermediate VTE risk’, or ‘Low VTE risk’) to be prescribed within the eMMS or to provide a free-text override reason. If indicated, VTE prophylaxis (pharmacological and/or mechanical) can be simultaneously prescribed on the same order screen. Once prescribed, the alert stops triggering to avoid redundant prompting. The VTE risk assessment orders are duration-specific with set order durations of 24 hours, 3 days or 7 days, selectable at the prescriber’s discretion within different clinical settings. The alert reappears to prompt re-documentation of a VTE risk assessment 24 hours after the chosen duration has lapsed.
Data collection
A review of paper and electronic medical records was conducted at 24 hours of admission. Using consecutive sampling, all patients admitted to the service over 7-day periods from 06/04/2022 to 13/04/2022 (pre-implementation) and 27/03/2023 to 02/04/2023 (post-implementation) were screened for inclusion. Patients were excluded from analysis if they were admitted for less than 24 hours, under the age of 18, or admitted under the Maternity and Rehabilitation services. To establish the effects of continued compliance to the intervention, the post-intervention audit was conducted 10 months post-implementation.
Auditors systematically reviewed patient medical records and collected information on baseline demographics (age, gender, body weight and height), medical history and laboratory findings. Medication charts were reviewed for details of pharmacological and mechanical VTE prophylaxis ordered and administered.
The study aimed to determine the rate of VTE risk assessment documentation at 24 hours of admission anywhere within the medical record (pre-intervention) and within the eMMS (post-intervention). VTE prophylaxis prescribing compliance was assessed according to a local VTE prevention protocol. Patients were categorised as either surgical (admitted for surgical procedures) or medical (admitted for acute medical conditions). For surgical patients, guideline compliance was assessed based on VTE risk associated with specific surgical procedures in addition to individual medical risk factors for VTE. Guideline compliance was further subdivided into pharmacological and mechanical prophylaxis.
Prior to the intervention, any documented form of VTE risk was accepted as a complete risk assessment including risk stratification or risk identification. Post-intervention, only electronic VTE risk assessment prescription orders were accepted as per local protocol. Due to the variability in pre-intervention VTE risk assessment documentation, only post-intervention data was used to analyse whether an accurately documented VTE risk assessment impacted the rate of guideline-compliant VTE prophylaxis prescribing.
Statistical analysis
Continuous variables were summarised using mean (standard deviation) or median (interquartile range), while categorical variables were expressed as count (percentages). Comparative analyses of patient profiles between pre- and post-implementation cohorts were conducted using Fisher’s exact test for categorical variables and Mann–Whitney U test for continuous variables. Study outcomes between pre- and post-implementation cohorts were compared using Fisher’s exact tests. For binary postoperative outcomes, univariable and multivariable binary logistic regressions were performed to estimate unadjusted and adjusted odds ratios (ORs), respectively. In the multivariable analysis, comparisons between pre- and post-implementation cohorts were adjusted for age category, sex, and clinical division (medical or surgical). Analyses were performed using Stata (Version 18.0, StataCorp, College Station, Texas, USA) at a significance level of α = 0.05.
Results
Out of 3217 patients screened, 850 patients met the inclusion criteria for assessment (pre-intervention group: 458, post-intervention group: 392) (Fig. 1). Patient demographics are shown in Table 1. There was a significantly greater proportion of medical and older patients in the post-intervention group.
Fig. 1.
Patient flow diagram outlining patient selection
Table 1.
Demographic information
| 2022 (n = 458), (%) | 2023 (n = 392), (%) | p-value | |||
|---|---|---|---|---|---|
| Sex | |||||
| Female | 261 | 57.0 | 209 | 53.3 | 0.283 |
| Male | 197 | 43.0 | 183 | 46.7 | |
| Age | 0.037 | ||||
| 18 to 49 | 91 | 19.9 | 53 | 13.5 | |
| 50 to 64 | 78 | 17.0 | 74 | 18.9 | |
| 65 to 74 | 104 | 22.7 | 79 | 20.2 | |
| 75 or older | 185 | 40.4 | 186 | 47.4 | |
| Median age | 71 | 74 | 0.159 | ||
| Medical | 182 | 39.7 | 185 | 47.2 | 0.029 |
| Surgical | 276 | 60.3 | 207 | 52.8 | |
| Minor Surgical VTE Risk | 107 | 71 | |||
| Moderate Surgical VTE Risk | 119 | 89 | |||
| Major Surgical VTE Risk | 50 | 47 | |||
| Total | 458 | 392 | |||
The rates of VTE risk assessment documentation and guideline-compliant VTE prescribing are outlined in Table 2. In the pre-intervention group, 22.9% (105 out of 458) of patients had a documented VTE risk assessment, which increased to 52.8% (207 out of 392 patients) post-intervention, resulting in an overall increase of 29.9% (95% CI 23.6%–26.1%, p < 0.001). Improvements were observed for prescribing of both pharmacological & mechanical prophylaxis, except in the case of mechanical prophylaxis prescribing among medical patients, where improvements were not statistically significant (Table 2).
Table 2.
VTE risk assessment documentation and VTE prophylaxis prescribing guideline compliance pre- and post-intervention
| 2022 pre-intervention Group (n = 458) | 2023 post-intervention group (n = 392) | Difference (95% CI) | p-value | |
|---|---|---|---|---|
| Rate of VTE risk assessment documentation | ||||
| Overall | 105 (22.9%) | 207 (52.8%) | 29.9% (23.6–26.1%) | < 0.001 |
| Medical | 6 (3.3%) | 78 (42.2%) | 38.9% (31.3–46.4%) | < 0.001 |
| Surgical | 99 (35.9%) | 129 (62.3%) | 26.4% (17.8–35.1%) | < 0.001 |
| Rate of guideline-compliant VTE prophylaxis prescribing | ||||
| Overall | 255 (55.7%) | 273 (69.6%) | 13.9% (6.4–14.0%) | < 0.001 |
| Medical | 102 (56.0%) | 121 (65.4%) | 9.4% (− 0.05–19.3%) | 0.066 |
| Surgical | 153 (55.4%) | 152 (73.4%) | 18% (9.6–26.4%) | < 0.001 |
| Rate of guideline-compliant pharmacological VTE prophylaxis prescribing | ||||
| Overall | 350 (76.4%) | 330 (84.2%) | 7.8% (2.5–13.1%) | 0.005 |
| Medical | 116 (63.7%) | 137 (74.1%) | 10.4% (0.9–19.7%) | 0.033 |
| Surgical | 234 (84.7%) | 193 (93.2%) | 8.5% (3.0–13.9%) | 0.004 |
| Rate of guideline-compliant mechanical VTE prophylaxis prescribing | ||||
| Overall | 337 (73.5%) | 331 (84.4%) | 10.9% (5.5–16.3%) | < 0.001 |
| Medical | 168 (92.3%) | 168 (90.1%) | 2.2% (− 7.2–4.2%) | 0.606 |
| Surgical | 169 (61.2%) | 163 (78.7%) | 17.5% (9.5–25.5%) | < 0.001 |
CI, Confidence interval; VTE, Venous thromboembolism
In the post-implementation group, when a VTE risk assessment was completed, 55 (26.6%) patients had a VTE risk assessment that did not align with local protocol (i.e. assessed as low, however should be high). Despite any misclassification, having any documented VTE risk assessment led to a 19.3% improvement in guideline-compliant prescribing (59.5 vs 78.7%, 95% CI 10.3–28.3%, p < 0.001). Furthermore, guideline-compliant VTE prophylaxis prescribing was significantly better in patients with an accurately documented VTE risk assessment compared to those with a misclassified VTE risk stratification (61.3 vs 83.6%, difference = 22.3%, 95% CI 13.8–30.8%, p < 0.001).
Results from univariate and multivariate logistic regression analysis are shown in Tables 3 and 4. In the univariable logistic regression analysis, post intervention, the odds of having a documented VTE risk assessment were 3.76 times higher (95% CI 2.80–5.05, p < 0.001), while the odds of receiving overall guideline-compliant VTE prophylaxis was 1.85 times higher (95% CI 1.39–2.46, p < 0.001). Patients were 1.64 times more likely to receive guideline-compliant pharmacological prophylaxis (95% CI 1.16–2.32, p = 0.005), and nearly twice as likely to receive guideline-compliant mechanical prophylaxis (OR 1.95, 95% CI 1.38–2.75, p < 0.001).
Table 3.
Univariable logistic regression analysis of factors affecting outcomes
| Variables | VTE risk assessment documentation | Guideline-compliant mechanical prophylaxis | Guideline-compliant pharmacological prophylaxis | Overall guideline-compliant VTE prophylaxis | ||||
|---|---|---|---|---|---|---|---|---|
| OR (95%CI) | p-value | OR (95%CI) | p-value | OR (95%CI) | p-value | OR (95%CI) | p-value | |
| Year | ||||||||
| 2022 | Reference | – | Reference | – | Reference | – | Reference | – |
| 2023 | 3.76 (2.80–5.05) | < 0.001 | 1.95 (1.38–2.75) | < 0.001 | 1.64 (1.16–2.32) | 0.005 | 1.85 (1.39–2.46) | < 0.001 |
| Age category | ||||||||
| 18–49 | Reference | – | Reference | – | Reference | – | Reference | – |
| 50–64 | 0.84 (0.53–1.33) | 0.456 | 1.27 (0.65–2.50) | 0.487 | 0.58 (0.30–1.13) | 0.107 | 0.87 (0.51–1.47) | 0.591 |
| 65–74 | 0.80 (0.51–1.24) | 0.315 | 0.37 (0.21–0.64) | < 0.001 | 0.64 (0.33–1.22) | 0.175 | 0.37 (0.23–0.60) | < 0.001 |
| 75 + | 0.61 (0.41–0.90) | 0.013 | 0.57 (0.34–0.97) | 0.038 | 0.35 (0.20–0.62) | < 0.001 | 0.39 (0.25–0.60) | < 0.001 |
| Sex | ||||||||
| M | Reference | – | Reference | – | Reference | – | Reference | – |
| F | 0.93 (0.70–1.23) | 0.615 | 1.14 (0.82–1.58) | 0.436 | 0.56 (0.39–0.79) | 0.001 | 0.82 (0.62–1.09) | 0.176 |
| Clinical division | ||||||||
| Medical | Reference | – | Reference | – | Reference | – | Reference | – |
| Surgical | 3.01 (2.23–4.07) | < 0.001 | 0.20 (0.13–0.31) | < 0.001 | 3.44 (2.41–4.90) | < 0.001 | 1.12 (0.84–1.48) | 0.440 |
OR Odds ratio; CI Confidence interval; VTE Venous thromboembolism
P-values for significant results are shown in boldface font
Table 4.
Multivariable logistic regression analysis of factors affecting outcomes
| Variables | VTE risk assessment documentation | Guideline-compliant mechanical prophylaxis | Guideline-compliant pharmacological prophylaxis | Overall guideline-compliant VTE prophylaxis | ||||
|---|---|---|---|---|---|---|---|---|
| AOR (95%CI) | p-value | AOR (95%CI) | p-value | AOR (95%CI) | p-value | AOR (95%CI) | p-value | |
| Year | ||||||||
| 2022 | Reference | – | Reference | – | Reference | – | Reference | – |
| 2023 | 4.89 (3.54–6.76) | < 0.001 | 2.04 (1.41–2.95) | < 0.001 | 1.95 (1.35–2.81) | < 0.001 | 1.98 (1.48–2.66) | < 0.001 |
| Age category | ||||||||
| 18–49 | Reference | – | Reference | – | Reference | – | Reference | – |
| 50–64 | 0.76 (0.46–1.27) | 0.298 | 1.06 (0.53–2.14) | 0.860 | 0.53 (0.27–1.07) | 0.075 | 0.76 (0.44–1.30) | 0.320 |
| 65–74 | 0.70 (0.43–1.15) | 0.160 | 0.32 (0.18–0.57) | < 0.001 | 0.56 (0.29–1.10) | 0.090 | 0.33 (0.20–0.54) | < 0.001 |
| 75 + | 0.68 (0.43–1.08) | 0.100 | 0.28 (0.16–0.50) | < 0.001 | 0.42 (0.23–0.76) | 0.004 | 0.33 (0.21–0.52) | < 0.001 |
| Sex | ||||||||
| M | Reference | – | Reference | – | Reference | – | Reference | – |
| F | 0.97 (0.71–1.33) | 0.849 | 1.05 (0.73–1.50) | 0.787 | 0.52 (0.36–0.75) | < 0.001 | 0.76 (0.57–1.02) | 0.067 |
| Clinical division | ||||||||
| Medical | Reference | – | Reference | – | Reference | – | Reference | – |
| Surgical | 3.79 (2.68–5.35) | < 0.001 | 0.16 (0.10–0.25) | < 0.001 | 3.24 (2.21–4.73) | < 0.001 | 0.99 (0.73–1.35) | 0.956 |
AOR Adjusted Odds ratio; CI Confidence interval; VTE Venous thromboembolism
P-values for significant results are shown in boldface font
After adjusting for demographic differences, the odds of achieving positive outcomes increased significantly across all measures. Post-intervention, patients were nearly five times more likely to have documented VTE risk assessments (AOR 4.89, 95% CI 3.54–6.76, p < 0.001) and nearly twice as likely to receive overall guideline-compliant VTE prophylaxis (AOR 1.98, 95% CI 1.48–2.66, p < 0.001). Patient were 1.95 as likely to receive guideline-compliant pharmacological prophylaxis (95% CI 1.35–2.81, p < 0.001), and 2.04 times more likely to receive guideline-compliant mechanical prophylaxis (95% CI 1.41–2.95, p < 0.001).
Discussion
Statement of key findings
This study showed that rates of VTE risk assessment documentation and guideline-compliant VTE prophylaxis prescribing significantly improved following implementation of an integrated electronic alert system linked to a CPOE-based order set for VTE risk assessment within the eMMS.
In typical practice, VTE risk assessments are documented separately within clinical forms or progress notes, often leading to oversight as physicians must prescribe VTE prophylaxis separately. Embedding VTE risk assessment into eMMS workflow through CPOE functionality ensures that documentation of risk assessments and medication prescribing can be completed at the same time, enhancing transparency and streamlining care. Additionally, electronic documentation of VTE risk assessments eliminates the need for manual auditing of paper-based medical records to assess compliance. Our service aims to generate monthly compliance reports to monitor organisational performance and formulate targeted interventions, ensuring continual optimisation of VTE prevention measures.
Reasons for non-compliant VTE prophylaxis prescribing were not formally reviewed, however common reasons included VTE prophylaxis being omitted or missed entirely, inappropriate delays in initiation of post-operative VTE prophylaxis among surgical patients or incorrect dosing of pharmacological prophylaxis.
Interpretation
Across all outcomes, effects of the intervention were more pronounced after accounting for potential confounders such as age, sex and clinical division, suggesting that demographic differences may have initially led to an underestimation of the intervention’s full impact.
Post-intervention, patients who had completed VTE risk assessments were significantly more likely to receive guideline-compliant VTE prophylaxis, compared to patients without completed assessments. Moreover, when VTE risk was accurately stratified, rates of guideline-compliant prophylaxis further improved. This finding highlights the role of the intervention in promoting comprehensive risk assessment documentation and its direct impact on subsequent clinical decision-making, contributing to a more systematic approach to patient care.
Our findings are consistent with studies examining CDS and CPOE-based interventions within EMR systems, supporting their use as effective quality improvement initiatives for VTE prevention [14–19].
In contrast, a study conducted in Switzerland found that electronic alert systems did not improve guideline-compliant VTE prophylaxis prescribing among hospitalised patients [22], most likely because alerts were ignored by ordering physicians due to alert fatigue. Similarly, another study reported no improvement in prescribing practice following the introduction of VTE CDS, likely due to additional time required by physicians to interact with the system [23].
A major shortcoming of our own previous system was the fundamental disconnect between paper-based and electronic systems; the alert appeared each time the eMMS was accessed, regardless of whether a VTE risk assessment was already documented on paper. Concern was noted surrounding this redundancy and its associated potential for inefficiencies, frustration and risk of alert fatigue. These studies, along with our experience, highlight key barriers to physician compliance including alert fatigue caused by the nonspecific, “blanket” nature of alerts and clinical workflow disruption. To address these challenges, our health service transitioned from unstandardised, paper-based documentation to a fully integrated electronic alert system that avoids redundant prompts and utilises CPOE functionality to enable electronic documentation of VTE risk assessments.
While the intervention led to significant improvements in the surgical patient population, results did not achieve statistical significance among medical patients. Higher compliance in both pre-and post-intervention surgical patient cohorts likely stemmed from heightened awareness among surgical teams of surgery as a known risk factor for VTE, prompting VTE risk assessment and thromboprophylaxis prescribing as part of routine surgical care. In contrast, poorer performance in the medical patient cohort, both before and after intervention, may be due to several factors. Medical patients often have multiple comorbidities and acute presenting conditions requiring prioritisation, causing physicians to potentially overlook VTE prophylaxis. Additionally, medical patients may have varied VTE risk profiles and stratifying risk as low, intermediate or high can be challenging for physicians resulting in reluctance to document a VTE risk assessment. This highlights a limitation of the system, as it lacks flexibility for cases where physicians exercise clinical judgement and choose to deviate from the local VTE prevention protocol based on individualised patient assessments. Although the system provides a free-text box for users to document override reasons, this feature was poorly utilised, likely due to physicians finding it time-consuming. This information was not formally reviewed as part of the study.
Pre- and post-intervention, non-compliant mechanical prophylaxis prescribing among medical patients was most evident in patients who had contraindications to pharmacological prophylaxis and therefore required mechanical prophylaxis instead. While it is common practice for physicians to withhold pharmacological prophylaxis due to concerns such as bleeding risk, they may overlook ordering of mechanical prophylaxis in its place. The proportion of these patients remained largely unchanged post-intervention, highlighting an additional gap with the intervention’s effectiveness.
To optimise CDS utility in this patient population, further improvements to the alert system could include implementing predefined options for alert overrides and an additional order within the orderset which prompts physicians to consider mechanical prophylaxis in cases where pharmacological options are contraindicated.
Strengths and weaknesses
This study investigated co-prescribing of VTE risk assessment and VTE prophylaxis prescription orders, presenting an innovative approach for incorporating VTE risk assessment documentation into eMMS workflow. Our tool was developed within a commercial eMMS software, making it easily replicable and scalable for use within other hospitals. Additionally, the statistical analysis approach accounted for demographic differences between pre- and post-intervention populations. This study has several limitations. A major limitation is inherent in the retrospective observational study design. Results may be prone to misclassification bias due to potential for incomplete documentation of relevant data in medical records. The 10-month gap between the intervention and post-intervention audit may introduce potential confounding factors, however this was deliberately chosen to assess the long-term impacts and sustainability of the CDS intervention. As data were collected from a single tertiary site, this limits the external validity of our results and findings may not be applicable to other sites. Lastly, our study does not directly evaluate whether improved guideline compliance translates into reduction in VTE events within our health service, which is the intended goal of prophylaxis.
Further research
While this study demonstrated improvements in VTE prevention guideline compliance, future research should assess patient-centred outcomes such as actual incidence of VTE events, bleeding complications rates, length of stay and mortality rates. Exploring these broader clinical outcomes will provide valuable information as to whether increased compliance rates translate into meaningful reductions in VTE-related complications. Subgroup analyses to assess performance across different subspecialities could provide insights into how the system performs across different clinical areas, identifying opportunities for further adjustments according to specific patient needs.
Further research should assess the economic impacts of CDS interventions within eMMS, including potential cost savings from eliminated VTE events and reduced length of hospital stay. Exploring user experience through factors such as ease of use, clinical value of the tool and impact on daily workflow could help identify key barriers to healthcare provider acceptance and optimise workflow integration. Establishing user feedback mechanisms would facilitate real-time improvements to enhance CDS system functionality and effectiveness. Future work is also required to explore the long-term sustainability and scalability of VTE CDS integration across other health services utilising eMMS to determine its feasibility for more widespread adoption.
Conclusion
An integrated electronic alert system linked to a CPOE-based VTE risk assessment order set within an eMMS significantly improved VTE risk assessment documentation and guideline-compliant VTE prophylaxis prescribing. Its functionality allows for standardised documentation and enhanced compliance monitoring, providing a valuable tool for health services utilising eMMS to enhance VTE prevention efforts.
Supplementary Information
Below is the link to the electronic supplementary material.
Funding
Open access funding provided by Grampians Health Service, Victorian Health Libraries Consortium (VHLC). The authors declare that no funds, grants or other support were received during the preparation of this manuscript.
Conflicts of interests
The authors have no relevant financial or non-financial interests to disclose.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Jha AK, Larizgoitia I, Audera-Lopez C, et al. The global burden of unsafe medical care: analytic modelling of observational studies. BMJ Qual Saf. 2013;22(10):809–15. 10.1136/bmjqs-2012-001748. [DOI] [PubMed] [Google Scholar]
- 2.Khan F, Tritschler T, Kahn S, et al. Venous thromboembolism. Lancet. 2021;398(10294):64–77. 10.1016/S0140-6736(20)32658-1. [DOI] [PubMed] [Google Scholar]
- 3.Wang C, Cui F, Li J, et al. Risk factors for venous thromboembolism in hospitalized patients in the Chinese population. Open Life Sci. 2018;13:82–9. 10.1515/biol-2018-0012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Falck-Ytter Y, Francis CW, Johanson NA, et al. Prevention of VTE in orthopedic surgery patients: antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 Suppl):e278S-e325S. 10.1378/chest.11-2404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gould MK, Garcia DA, Wren SM, et al. Prevention of VTE in nonorthopedic surgical patients: antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 Suppl):e227S-e277S. 10.1378/chest.11-2297. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kahn SR, Lim W, Dunn AS, et al. Prevention of VTE in nonsurgical patients: antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012;141(2 Suppl):e195S-e226S. 10.1378/chest.11-2296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Lester W, Freemantle N, Begaj I, et al. Fatal venous thromboembolism associated with hospital admission: a cohort study to assess the impact of a national risk assessment target. Heart. 2013;99(23):1734–9. 10.1136/heartjnl-2013-304479. [DOI] [PubMed] [Google Scholar]
- 8.Rowswell HR, Nokes TJC. Significant reduction in hospital-acquired thrombosis: impact of national risk assessment and real-time feedback: open. Heart. 2017;4(2): e000653. 10.1136/openhrt-2017-000653. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bakhsh E. The benefits and imperative of venous thromboembolism risk screening for hospitalized patients: a systematic review. J Clin Med. 2023;12(22):7009. 10.3390/jcm12227009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cohen AT, Tapson VF, Bergman J, et al. Venous thromboembolism risk and prophylaxis in the acute hospital care setting (ENDORSE study): a multinational cross-sectional study. The Lancet. 2008;371(9610):387–94. 10.1016/S0140-6736(08)60202-0. [DOI] [PubMed] [Google Scholar]
- 11.Jaspers T, Duisenberg-van Essenberg M, Maat B, et al. A multifaceted clinical decision support intervention to improve adherence to thromboprophylaxis guidelines. Int J Clin Pharm. 2021;43:1327–36. 10.1007/s11096-021-01254-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Gallier S, Topham A, Nightingale P, et al. Electronic prescribing systems as tools to improve patient care: a learning health systems approach to increase guideline concordant prescribing for venous thromboembolism prevention. BMC Med Inform Decis Mak. 2022;22(1):121. 10.1186/s12911-022-01865-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kahn SR, Morrison DR, Diendéré G, et al. Interventions for implementation of thromboprophylaxis in hospitalized patients at risk for venous thromboembolism. Cochrane Database Syst Rev. 2018;4:008201. 10.1002/14651858.CD008201.pub3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Borab ZM, Lanni MA, Tecce MG, et al. Use of computerized clinical decision support systems to prevent venous thromboembolism in surgical patients: a systematic review and meta-analysis. JAMA Surg. 2017;152(7):638–45. 10.1001/jamasurg.2017.0131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Karajizadeh M, Hassanipour S, Sharifian R, et al. The effect of information technology intervention on using appropriate VTE prophylaxis in non-surgical patients: a systematic review and meta-analysis. Digit Health. 2022;8:20552076221118828. 10.1177/20552076221118828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Amland RC, Dean BB, Yu H, et al. Computerized clinical decision support to prevent venous thromboembolism among hospitalized patients: proximal outcomes from a multiyear quality improvement project. J Healthc Qual. 2014;37(4):221–31. 10.1111/jhq.12069. [DOI] [PubMed] [Google Scholar]
- 17.George B, Gonzales S, Patel K, et al. Impact of a clinical decision-support tool on venous thromboembolism prophylaxis in acutely Ill medical patients. J Pharm Technol. 2020;36(4):141–7. 10.1177/8755122520930288. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Mathers B, Williams E, Bedi G, et al. An electronic alert system is associated with a significant increase in pharmacologic venous thromboembolism prophylaxis rates among hospitalized inflammatory bowel disease patients. J Healthc Qual. 2017;39(5):307–14. 10.1097/JHQ.0000000000000021. [DOI] [PubMed] [Google Scholar]
- 19.Bhalla R, Berger MA, Reissman SH, et al. Improving hospital venous thromboembolism prophylaxis with electronic decision support. J Hosp Med. 2014;8:115–20. 10.1002/jhm.1993. [DOI] [PubMed] [Google Scholar]
- 20.Haut ER, Lau BD, Kraenzlin FS, et al. Improved prophylaxis and decreased rates of preventable harm with the use of a mandatory computerized clinical decision support tool for prophylaxis for venous thromboembolism in trauma. Arch Surg. 2012;147(10):901–7. 10.1001/archsurg.2012.2024. [DOI] [PubMed] [Google Scholar]
- 21.Soo GC, Wong Doo NC, Burrows J, et al. Improving the adoption of an electronic clinical decision support tool and evaluating its effect on venous thromboembolism prophylaxis prescribing at a Sydney tertiary teaching hospital. J Pharm Pract Res. 2019;49:508–16. 10.1002/jppr.1562. [Google Scholar]
- 22.Spirk D, Stuck AK, Hager A, et al. Electronic alert system for improving appropriate thromboprophylaxis in hospitalized medical patients: a randomized controlled trial. J Thromb Haemost. 2017;15(11):2138–46. 10.1111/jth.13812. [DOI] [PubMed] [Google Scholar]
- 23.Eijgenraam P, Meertens N, van den Ham R, et al. The effect of clinical decision support on adherence to thrombosis prophylaxis guidelines in medical patients; a single center experience. Thromb Res. 2015;135(3):464–71. 10.1016/j.thromres.2014.12.012. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.