Abstract
Background and Aims
Resumption recommendations for anticoagulants (ACs) and antiplatelet agents (APAs) after endoscopy are inconsistently documented, which has been associated with a small but clinically significant number of adverse events (AEs). Clinical decision support systems must be implemented properly to ensure high adoption and clinically appropriate use. This study aimed to improve the rate of documentation of resumption recommendations for ACs and APAs using an electronic health record–based intervention integrated into the existing clinical workflow.
Methods
We implemented a soft-stop endoscopy software alert to trigger if resumption recommendations were not initially documented in patients on ACs and APAs. The rates of documentation of resumption recommendations were compared between pre- and postintervention periods.
Results
Across 1238 encounters for elective outpatient endoscopy, there was no difference in AC and APA use between pre- and postintervention periods. More endoscopies were performed in patients taking ACs (77.9%) compared with those taking APAs (25.3%). The intervention led to a significant increase in documentation of resumption recommendations postintervention from 47.1% (95% confidence interval [CI], 42.6-51.6) and 41.6% (95% CI, 34.1-49.6) to 70.5% (95% CI, 66.3-74.4; P < .001) and 73.1% (95% CI, 65.6-79.4; P < .001) in patients on ACs and APAs, respectively. Patients who were older, undergoing colonoscopy, or female were more likely to have documentation of recommendations, whereas those undergoing endoscopic ultrasound (EUS) or endoscopic retrograde cholangiopancreatography (ERCP) on ACs were less likely to have recommendations documented. Sedation type and the presence of a fellow did not impact documentation rates.
Conclusions
Implementing an electronic health record–based intervention into the existing endoscopy workflow improved the documentation of resumption recommendations for antithrombotic medications after elective outpatient endoscopy.
Gastroenterologists are tasked with providing recommendations on the resumption of anticoagulants (ACs) and antiplatelet agents (APAs) after endoscopy. An estimated 10% of patients on ACs each year require a temporary interruption, with gastrointestinal (GI) endoscopy representing one of the most common indications for interruption.1,2
Current recommendations for periendoscopic management of ACs and APAs are based on American Society for Gastrointestinal Endoscopy and American College of Gastroenterology–Canadian Association of Gastroenterology guidelines.2,3 Despite this, documentation on resumption of these medications is inconsistently performed, which may result in significant adverse events (AEs) if therapy is either withheld for longer or restarted earlier than necessary. A recent review at our large academic medical center of patients on antithrombotic agents (excluding aspirin monotherapy) undergoing elective outpatient endoscopy across an approximately 4-year period identified that only 34% of patients had recommendations for antithrombotic agent management documented in the procedure report.4 Although these recommendations may have been communicated verbally, written on discharge paperwork, or documented in nursing notes, this information can be forgotten, misplaced, or lost in an era of electronic health record overload. Recent studies in patients undergoing outpatient endoscopy found a small but significant number of patients who experienced AEs related to interruption of antithrombotic agents within 30 days of endoscopy, namely GI bleeding and thrombosis.4,5 However, strategies to optimize documentation rates for resumption recommendations have not been well characterized.
Clinical decision support systems can introduce challenges, including alert fatigue and derailing of the existing clinical workflow.6 As such, the optimal intervention is process oriented, triggers at the appropriate time, and provides a clinically meaningful information. In this study, our aim was to improve the rate of documentation of resumption recommendations for ACs and APAs using an endoscopy software–based intervention integrated into the existing workflow.
Methods
All patients on single or multiple ACs and APAs (excluding aspirin monotherapy) who underwent elective outpatient endoscopy between July 1, 2022 to October 31, 2022 (preintervention) and July 1, 2023 to October 31, 2023 (postintervention) at our large academic center were included. Approval for this study was granted by the Northwestern Digestive Health Center Quality Improvement Committee.
We designed a soft-stop reminder within the recommendations section of the endoscopy documentation software to trigger a popup alert if recommendations were not initially recorded by the endoscopist (Fig. 1). This alert included the name and dosage of the specific antithrombotic medication by cross-referencing the preoperative medication list. Our intervention was implemented on all outpatient endoscopies on July 1, 2023. Data from a 4-month period postintervention were compared with the same period 1 year before control for confounders such as seasonal variation or new trainees.
Figure 1.
Electronic health record alert prompting the endoscopist to document resumption recommendations in the endoscopy report.
Data extraction included patient demographics, endoscopy indication and type, medications, and recommendations regarding resumption of ACs and APAs as per review of the procedure report. Generalized linear models (PROC GENMOD, SAS v9.2; SAS Institute, Cary, NC, USA) were used to estimate the rate of use and rates of documentation for continued use for each agent while adjusting for repeated measures on patients by assuming an exchangeable variance structure. Least-square mean estimates and 95% confidence intervals (CIs) were used for the rate of use and documentation for each agent pre- and postintervention. Individual generalized linear models were fit to determine if any patient or encounter descriptives might predict documentation.
Results
During the study period, there were 1238 endoscopic encounters (605 preintervention and 633 postintervention) for elective outpatient endoscopy on 1121 unique patients on ACs and/or APAs. Of these encounters, 965 (77.9%) were for patients taking ACs and 313 (25.3%) occurred on patients taking APAs (Table 1). More patients on ACs in both the pre- and postintervention groups underwent endoscopy compared with those on APAs. Apixaban and clopidogrel were the most common ACs and APAs, respectively. Most patients underwent colonoscopy, although patients who underwent endoscopic ultrasound (EUS) or endoscopic retrograde cholangiopancreatography (ERCP) accounted for 10.7% of preintervention encounters and 8.9% of postintervention encounters. Surveillance, including for colon polyps or cancer, Barrett’s esophagus, and inflammatory bowel disease, was the most common procedure indication in the preintervention group (37.7%), whereas diagnostic studies were the most common in the postintervention group (34.0%). A fellow was present in 20.7% of preintervention procedures compared with 22.4% of postintervention procedures. There was no difference in AC or APA use between pre- and postintervention (Table 2).
Table 1.
Encounter descriptives in pre- and postintervention groups
| Variable | Preintervention (n = 605) | Postintervention (n = 633) |
|---|---|---|
| Encounters with anticoagulants | 468 (77.4) | 497 (78.5) |
| Apixaban | 230 | 274 |
| Rivaroxaban | 126 | 107 |
| Other | 83 | 84 |
| Multiple | 29 | 32 |
| Encounters with antiplatelet agents | 159 (26.3) | 154 (24.3) |
| Clopidogrel | 132 | 132 |
| Other | 25 | 20 |
| Multiple | 2 | 2 |
| Procedure type | ||
| Colonoscopy | 370 (61.2) | 375 (59.2) |
| Upper gastrointestinal endoscopy | 201 (33.2) | 256 (40.4) |
| Endoscopic ultrasound/endoscopic retrograde cholangiopancreatography | 65 (10.7) | 56 (8.9) |
| Miscellaneous | 36 (6.0) | 64 (10.1) |
| Sedation type | ||
| Moderate | 317 (52.4) | 300 (47.5) |
| Monitored anesthesia | 265 (43.8) | 310 (49.1) |
| None | 0 (0) | 13 (2.1) |
| General anesthesia | 3 (0.5) | 9 (1.4) |
| Not specified | 20 (3.3) | 0 (0) |
| Indication type | ||
| Biliary obstruction | 43 (7.1) | 27 (4.3) |
| Diagnostic | 163 (26.9) | 215 (34.0) |
| Surveillance | 228 (37.7) | 202 (31.9) |
| Screening | 132 (21.8) | 129 (20.4) |
| Therapeutic | 27 (4.5) | 26 (4.1) |
| Ultrasound-guided biopsy sampling | 6 (1.0) | 9 (1.4) |
| Not available or multiple | 6 (1.0) | 25 (4.0) |
| Fellow present | 125 (20.7) | 142 (22.4) |
Values are n or n (%).
Table 2.
Generalized linear model estimates for the rates of use and documentation for ACs and APAs pre- and postintervention
| Medication | Preintervention | Postintervention | P value |
|---|---|---|---|
| Rate of use | |||
| ACs | 25.8 (23.7-28.0) | 26.8 (24.7-29.0) | .505 |
| APAs | 9.0 (6.9-11.8) | 7.8 (5.9-10.4) | .570 |
| Rate of documentation | |||
| ACs | 47.1 (42.6-51.6) | 70.5 (66.3-74.4) | <.001 |
| APAs | 41.6 (34.1-49.6) | 73.1 (65.6-79.4) | <.001 |
Values are % (95% confidence interval).
AC, Anticoagulant; APA, antiplatelet agent.
There was a significant increase in documentation of resumption recommendations postintervention, with rates increasing from 47.1% (95% CI, 42.6-51.6) and 41.6% (95% CI, 34.1-49.6) to 70.5% (95% CI, 66.3-74.4; P < .001) and 73.1% (95% CI, 65.6-79.4; P < .001) in patients on ACs and APAs, respectively (Table 2). After adjusting for study period, patients who were older (odds ratio [OR], 1.02; 95% CI, 1.01-10.3), female (OR, 1.52; 95% CI, 1.15-2.00), or undergoing colonoscopy (OR, 1.75; 95% CI, 1.34-2.30) were more likely to have documentation of resumption recommendations for ACs, whereas those undergoing EUS or ERCP were less likely to have documented recommendations for resumption of ACs (Table 3). Patients on APAs undergoing a diagnostic procedure were less likely to have recommendations documented compared with those undergoing procedures for all other indications (OR, 0.55; 95% CI, 0.33-0.90). Patient race, sedation type, and presence of a fellow did not predict rates of documentation of recommendations for ACs or APAs. Two patients in the preintervention group experienced AEs related to ACs and/or APAs compared with 1 patient in the postintervention group.
Table 3.
Potential predictors of documentation after adjusting for pre- and postintervention
| Demographics | Anticoagulants | Antiplatelet agents |
|---|---|---|
| Age (years) | 1.02 (1.01-10.3) | 0.99 (0.98-1.02) |
| Sex (female) | 1.52 (1.15-2.00) | 0.89 (0.52-1.49) |
| Race | ||
| White | 1.40 (0.91-2.13) | 0.96 (0.44-2.07) |
| Black/African American | 1.33 (0.79-2.24) | 1.42 (0.54-3.68) |
| Other/not specified | Reference | Reference |
| Indication type | ||
| Diagnostic (vs all other) | 0.86 (0.64-1.16) | 0.55 (0.33-0.90) |
| Surveillance (vs all other) | 1.19 (0.90-1.57) | 1.49 (0.86-2.57) |
| Screening (vs all other) | 1.13 (0.82-1.57) | 1.48 (0.83-2.64) |
| Procedure type | ||
| Colonoscopy (vs all other) | 1.75 (1.34-2.30) | 1.53 (0.94-2.49) |
| Upper endoscopy (vs all other) | 0.83 (0.63-1.10) | 0.65 (0.39-1.08) |
| Endoscopic ultrasound/endoscopic retrograde cholangiopancreatography (vs all other) | 0.54 (0.36-0.83) | 0.97 (0.43-2.19) |
| Sedation type | ||
| Moderate (vs all other) | 1.16 (0.89-1.51) | 1.41 (0.87-2.31) |
| Monitored anesthesia (vs all other) | 1.06 (0.82-1.39) | 0.71 (0.43-1.16) |
| Fellow present | 1.16 (0.83-1.60) | 0.90 (0.50-1.63) |
Values are odds ratios (95% confidence intervals).
Discussion
This quality improvement intervention significantly increased the rates of resumption recommendations for antithrombotic medications after outpatient endoscopy. Although previous studies have shown improvement in documentation rates for resumption recommendations using multifaceted interventions requiring educational initiatives, nursing interventions, and endoscopy-software alerts, our study demonstrated that a endoscopy-software alert alone was sufficient to significantly raise the rates of documentation of recommendations.7
We designed the alert to trigger in patients who had antithrombotic medications on their preoperative medication list and integrated it into the procedure report, a step already in the endoscopist’s workflow, to minimize additional active steps. To reduce alert fatigue, we implemented the reminder as a soft stop so that it could be bypassed if needed with a single additional click. A minimally time-intensive intervention that leverages the existing resource of an endoscopy documentation software would allow for easier implementation at other centers. Further studies should assess the experience of the endoscopist to determine its effect on endoscopy workflow and alert burden to help overcome additional barriers to implementation.8
Given the potentially catastrophic AEs associated with inappropriate periprocedural management of antithrombotic agents, improving communication about these medications after endoscopy should be prioritized. Although these recommendations may have been communicated through other means, the endoscopy report represents a central location for communication with patients and their care team and should be used to limit errors in management of these medications.9
Although the study was limited to patients undergoing elective outpatient endoscopy, the intervention is currently active in both outpatient and inpatient endoscopies at our large academic center. Further investigation is needed to assess the efficacy of this intervention for inpatient endoscopies and to track the improvement in documentation rates over time. Given the crucial role of these medications, the effect of improvement in the rate of documented management recommendations on the rate of AEs should be studied.
In conclusion, the use of an endoscopy documentation software–based intervention significantly improved the rate of documentation of resumption recommendations for ACs and APAs after elective outpatient endoscopy. Further study is needed to determine whether this reduces AEs and to evaluate the response to this intervention over time.
Disclosure
The following author disclosed financial relationships: S. Komanduri: Consultant for Boston Scientific, Ethicon J&J, Castle, and Lucid. All other authors disclosed no financial relationships.
References
- 1.Sherwood M.W., Douketis J.D., Patel M.R., et al. Outcomes of temporary interruption of rivaroxaban compared with warfarin in patients with nonvalvular atrial fibrillation. Circulation. 2014;129:1850–1859. doi: 10.1161/CIRCULATIONAHA.113.005754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Abraham N.S., Barkun A.N., Sauer B.G., et al. American College of Gastroenterology-Canadian Association of gastroenterology clinical practice guideline: management of anticoagulants and antiplatelets during acute gastrointestinal bleeding and the periendoscopic period. Am J Gastroenterol. 2022;117:542. doi: 10.14309/ajg.0000000000001627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.ASGE Standards of Practice Committee. Acosta R.D., Abraham N.S., Chandrasekhara V., et al. The management of antithrombotic agents for patients undergoing GI endoscopy. Gastrointest Endosc. 2016;83:3–16. doi: 10.1016/j.gie.2015.09.035. [DOI] [PubMed] [Google Scholar]
- 4.Than J., Rodrigues T., Kan A., et al. Management of antithrombotic agents after elective gastrointestinal endoscopy: rates of documentation and adverse events in a large tertiary care center [abstract] Gastrointest Endosc. 2023;97:AB370. [Google Scholar]
- 5.Hansen-Barkun C., Martel M., Douketis J., et al. periprocedural management of patients with atrial fibrillation receiving a direct oral anticoagulant undergoing a digestive endoscopy. Am J Gastroenterol. 2023;118:812–819. doi: 10.14309/ajg.0000000000002076. [DOI] [PubMed] [Google Scholar]
- 6.Wasylewicz A.T.M., Scheepers-Hoeks A.M.J.W. In: Fundamentals of clinical data science. Kubben P., Dumontier M., Dekker A., editors. Springer; 2019. Clinical decision support systems; pp. 153–169. [PubMed] [Google Scholar]
- 7.O’Connell B., Boyd A., Kothari D., et al. Improving documentation of anticoagulation and antiplatelet recommendations after outpatient endoscopy. BMJ Open Qual. 2022;11 doi: 10.1136/bmjoq-2021-001725. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Bisschops R., Rutter M.D., Areia M., et al. Overcoming the barriers to dissemination and implementation of quality measures for gastrointestinal endoscopy: European Society of Gastrointestinal Endoscopy (ESGE) and United European Gastroenterology (UEG) position statement. United Eur Gastroenterol J. 2021;9:120–126. doi: 10.1177/2050640620981366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Yen T., Jones B., Espinoza J.M., et al. Optimizing endoscopy procedure documentation improves guideline-adherent care in upper gastrointestinal bleeding. Dig Dis Sci. 2023;68:2264–2275. doi: 10.1007/s10620-023-07823-6. [DOI] [PMC free article] [PubMed] [Google Scholar]