Impact of provider education and implementation of an electronic order on follow-up after food bolus impaction

Introduction

Over the past few decades, there has been a significant increase in the incidence of esophageal food bolus impactions (FBI) with eosinophilic esophagitis (EoE) becoming the most common etiology of FBI (1). In fact, EoE-associated emergency department (ED) visits have tripled and are projected to double by 2030, representing a substantial burden to EDs and healthcare resources (2). Beyond management of the acute event itself, essentially all patients who experience FBI have underlying esophageal diseases, such as EoE or gastroesophageal reflux disease, that warrant longitudinal care (1,3,4).

Appropriate post-FBI care is lacking despite the large body of literature demonstrating the prevalence of esophageal diseases in this population. Previous studies have demonstrated that approximately 40–50% of patients are lost to follow-up after FBI requiring endoscopic disimpaction (5,6). In addition, those who have non-endoscopic resolution (NER) of their FBI (approximately 20–25% of cases) are more likely to be lost to follow-up compared to those who require endoscopic disimpaction (3). In another study, one quarter of patients presenting with FBI requiring endoscopy did not receive appropriate post-endoscopy care, and nearly 1 in 5 patients had a change in diagnosis at follow-up, including 3 new cases of esophageal cancer (7). While lack of follow-up has also been associated with recurrent food impaction, this outcome alone does not capture the burden of esophageal symptoms on patient quality of life (8,9).

In order to mitigate the potential burden on patients and the healthcare system, establishing standardized protocols for post-FBI management is paramount for early diagnosis and management of chronic esophageal conditions. Written recommendations for follow-up have been shown to improve follow-up rates, but to date, no system level intervention has been attempted to improve follow-up among patients presenting with FBI (8). Therefore, our objective was to develop and implement a standardized program involving provider education and an electronic order set to standardize care and improve follow-up rates for patients presenting to the ED with FBI. We present this article in accordance with the STROBE reporting checklist (available at https://aoe.amegroups.org/article/view/10.21037/aoe-24-5/rc).

Methods

Setting and implementation

A multidisciplinary team of ED and gastroenterology providers conducted a quality improvement initiative at a single center tertiary care hospital to standardize care and improve referrals from the ED to gastroenterology care after FBI. An electronic order set was created with nested pre-selected orders comprised of acute management orders (i.e., chest X-ray, glucagon), detailed discharge instructions (which auto-populated in the patient’s discharge paperwork), and an automated referral to gastroenterology clinic (Table 1). Discharge instructions described common causes of food impaction and emphasized the importance of follow-up after FBI (Appendix 1). The referral to gastroenterology clinic was expedited to one of our three esophageal experts. Patients were then contacted three times by a scheduler, and if no contact was made, an automated message sent to the patient with instructions on how to schedule an appointment. The order set was designed with the involved stakeholders to be simple and efficient so as to not increase the click burden of ordering providers. The inclusion of the automated referral aimed to streamline the post-FBI referral process, eliminating ambiguity regarding the responsible party for initiating the referral to gastroenterology and obviating the requirement for patients to see their primary care provider for a referral.

To implement the order set, ED and gastroenterology providers were emailed and received a dedicated training lecture on the order set. Quarterly reminder emails were sent and at 6 months, reeducation at department lectures was performed. The order set went live on January 1^st, 2022.

Study population

Adult patients (age ≥18 years) who presented to the University of Wisconsin ED with an FBI between January 1^st and December 31^st, 2022 were included for analysis after implementation of the FBI order set. FBI was defined as a recent food ingestion and inability to tolerate oral secretions or liquids. Patients were prospectively identified using the International Classification of Diseases (ICD) 10 codes of “foreign body in esophagus” T18.1, “food in esophagus” T18.12 or “food impaction of the esophagus” T18.128. Historical controls from January 1^st, 2018 to December 31^st, 2019 were identified using the same search strategy. We elected to use these dates because of the impact of the coronavirus disease 2019 (COVID-19) pandemic on clinical care in 2020 and to prevent our results from being influenced by any changes in practice in the lead up to the order set implementation, as the development of this project had been discussed at divisional meetings. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional board of the University of Wisconsin (No. 2021-0890) and individual consent for this analysis was waived as the study posed no more than minimal risk for participants based on study design.

Outcomes

The primary outcome was the proportion of patients receiving appropriate follow-up after FBI defined as an elective upper endoscopy or outpatient gastroenterology clinic appointment within 6 months of FBI. Patients were excluded from the study if outpatient follow-up was planned outside of our medical system or if patients were not medically homed within our system. All patients had a total of 1 year of follow-up data.

Statistical methods

Descriptive statistics were used to summarize baseline demographic characteristics. Comparison between groups were tested using the Pearson χ² and Student t-test. Comparisons were made between the prospective cohort and historical controls as well as between those in the prospective cohort in whom the order set was utilized or not utilized.

Results

Demographics, food impaction management and order set utilization

We identified 116 unique patient encounters during the study period of which there was a total of 76 FBI meeting inclusion criteria (13 excluded due to follow-up planned outside the health system and 27 non-FBI encounters). Of the 76 FBI, 43 (56.6%) required endoscopic disimpaction. Patients were predominantly Caucasian males with a median age of 54.6±21.0 years and all patients had health insurance (Table 2). The order set was utilized for 26 patients (34.2%) and was more frequently used in those requiring endoscopic disimpaction (73.1%) vs. NER (26.9%, P=0.04).

For the 43 subjects who required endoscopic disimpaction, reflux-related disorders such as peptic stricture or reflux esophagitis was the most commonly suspected etiology in 21 subjects (48.9%). EoE was the next most commonly suspected etiology in 15 subjects (34.9%) followed by other in 4 (9.3%), prior esophageal surgery in 2 (4.7%), and esophageal dysmotility in 1 (2.3%). Biopsies were obtained in only 8 (18.6%) of cases. Among those whom were biopsied, EoE was diagnosed in 6 (75%) of cases.

Prospective cohort follow-up

Forty patients (52.6%) had clinic or endoscopic follow-up within 6 months with no significant difference between those for whom the order set was utilized (61.5%) and for whom it was not (48.0%, P=0.26) (Figure 1). Additionally, there was no significant improvement in follow-up rates with use of the order set of those with endoscopic disimpaction [68.4% (13/19) vs. 66.7% (16/24), P=0.90] or NER [42.9% (3/7) vs. 30.8% (8/26), P=0.36]. However, follow-up at 6 months was significantly higher in the group that required endoscopic disimpaction (72.5%) compared to those with NER (27.5%, P<0.01). Lastly, when follow-up data was expanded to 1 year after FBI, only one additional patient in whom the order set was not used and had NER of their FBI followed up within that timeframe.

Figure 1 Six-month follow-up rates after food bolus impaction for the prospective cohort and historical controls.

Comparison to historical controls

Between 2018–2019, there were 156 instances of FBI. Compared to the prospective cohort, there was no significant difference in patient demographics. Similar to the prospective cohort, endoscopy was required in 63.5% of cases (P=0.31). Moreover, there was no difference in rates of follow-up at 6 months after implementation of the order set compared to historical controls (52.6% vs. 64.1%, P=0.09).

Discussion

Our study re-demonstrates the significant gap in appropriate post-FBI care, underscoring the critical need for structured quality improvement initiatives to address this deficiency. Using a conservative definition of appropriate care (upper endoscopy or clinic follow-up within 6 months), only 52.6% of patients had appropriate post-FBI care despite our quality improvement plan and repeated provider education. Similar to our previous study, patients with NER had significantly lower rates of post-FBI care, emphasizing the importance of including this population in future initiatives (3). This study is the first to date that aimed to address the gaps in post-FBI follow-up care by introducing a systems-based, simple electronic order set for ED and gastroenterology providers. Our study failed to demonstrate any improvement in follow-up rates for those in whom the order set was employed or even compared to historical controls. Despite these less-than-favorable outcomes, our study provides valuable insights that can inform future endeavors aimed at investigating this important issue.

One of the primary findings of this study, and a major limitation, was the low utilization of the order set (34.2%), particularly among those with NER of FBI, despite repeated re-education efforts and simplification of the order set to encourage use. The low utilization among ED providers can likely be explained by their infrequent exposure to FBI, prolonging periods between education of the order set and actual FBI events. Additionally, ED providers may think NER portends a less significant disease, thereby underestimating the significance of FBI and association with underlying esophageal pathologies. However, use among gastroenterology (GI) providers was unexpectedly low despite repeated re-education efforts and in-depth knowledge of esophageal disorders. Guo et al. hypothesized that this may be partly explained by anchoring bias when GI providers, typically task-focused on disimpaction, fail to identify an obvious cause of FBI on abbreviated endoscopic exam and become less likely to pursue follow-up investigation or treatment (7). Nevertheless, when ED or GI providers fail to recognize the importance of follow-up after FBI, patients are likely inadequately educated on the significance of the event and need for follow-up. These findings demonstrate that simple provider education (and need for continual re-education) is insufficient to improve follow-up after FBI and that system-based approaches are necessary.

Order set utilization and referral to outpatient GI referral may have been improved by changing how the order set was implemented. The decision to make the order set voluntary was based on discussions with key stakeholders, particularly ED providers, who were concerned about alert burden by placing hard stops in the electronic medical records using ICD codes. An alternative option considered was an automated referral to GI clinic upon chart closure using ICD codes not requiring provider approval or signature, but this approach raised concerns about inappropriate referrals if true FBI did not occur. Consequently, our approach with a voluntary order set aimed to strike a balance between encouraging its use, avoiding excessive alerts and avoiding unnecessary referrals; however, our study demonstrates that a voluntary system such as ours is unlikely to improve post-FBI follow-up. Importantly, while 23.3% of the encounters identified by the ICD codes used were deemed to not meet the clinical definition of a FBI, the raw total was only 27 encounters over a calendar year, which is unlikely to result in a significant burden of alert fatigue in the ED or overwhelm GI clinic triaging of unnecessary referrals.

While provider factors play a significant role in follow-up after FBI, patient factors are likely just as, if not more, important. Even among those in whom a referral was placed and contact made by our clinic to schedule follow-up, only 61.5% completed follow-up within 6 months, which was no better than those in whom the order set was not ordered or compared to historical controls. This suggests that we were unlikely to see significant increase in follow-up rates even with greater utilization of the order set, and that loss to follow-up in this group may therefore be patient-dependent. While our study was not specifically designed to explore patient behaviors among this group, there are several plausible explanations. For some, FBI may be perceived as a singular, isolated event, rather than an indication of underlying esophageal disease. For others who have lived with chronic esophageal symptoms, they may have become accustomed to their condition, developed adaptive behaviors, and may be complacent with their symptoms rather than seek immediate follow-up (10). In fact, Murray et al. found that among 19 patients who failed to follow-up after FBI, the main reasons for not attending a follow-up appointment were lack of symptoms (40%), finding an appointment time consuming (30%), feeling an appointment was unnecessary (24%), and cost (8%) (11). As discussed earlier, these issues can be addressed with proper patient education, which we attempted to achieve by including detailed patient instructions in the ED discharge paperwork. Our study suggests that patient education at time of the FBI event may be insufficient (possibly complicated by anesthesia related amnesia), and that re-engagement/re-education after the event may be necessary to improve follow-up rates.

A major limitation to our study was the low utilization of the order set which was discussed previously. Another major limitation of the study was our low sample size. With a longer study duration and greater subject recruit, we may have been able to identify greater differences in follow-up. Lastly, our study was limited in its ability to identify patient factors associated with loss to follow-up which, as discussed previously, appears to be one of the greatest factors impacting follow-up after FBI.

Future studies can draw valuable lessons from our attempt at improving post-FBI care. One of the greatest strengths of this study was the inclusion of patients who experience NER of FBI, as this population is frequently excluded from FBI studies. Given the higher rates of loss to follow-up in this group, future studies must deliberately include patients who experience NER. Secondly, hard stop alerts or automated referrals should be used to eliminate provider discretion in the process, as simple provider education and a voluntary order set resulted in low utilization and follow-up rates. Additional tools could also be integrated, such as follow-up appointment scheduling before patients leave the ED or text message appointment reminders, both of which have been shown to improve follow-up after ED visits (12,13). The feasibility of such tools, however, may be constrained by important infrastructure requirements (e.g., after hours appointment scheduling), which are unlikely to be implemented specifically for an infrequent event like FBI. Additionally, patients were unlikely to follow-up beyond 6 months, highlighting the critical importance of timely intervention at the time of, and immediately after, the FBI. Lastly, studies must place a greater emphasis on patient education after FBI, as patient-dependent factors are likely a main factor for loss to follow-up after FBI.

Conclusions

Our study re-emphasizes the high rates of loss to follow-up after FBI. While we were unable to improve follow-up by educating providers and implementing a simple voluntary order set, future researchers can learn from the valuable insights and pitfalls that our study provides. By addressing these aspects, future quality improvement initiatives will pave the way for more effective strategies, ensuring that all individuals presenting with FBI receive the necessary post-event care, and ultimately improve management of their esophageal disease.

Acknowledgments

Funding: None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://aoe.amegroups.org/article/view/10.21037/aoe-24-5/rc

Data Sharing Statement: Available at https://aoe.amegroups.org/article/view/10.21037/aoe-24-5/dss

Peer Review File: Available at https://aoe.amegroups.org/article/view/10.21037/aoe-24-5/prf

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://aoe.amegroups.org/article/view/10.21037/aoe-24-5/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the institutional board of the University of Wisconsin (No. 2021-0890) and individual consent for this analysis was waived as the study posed no more than minimal risk for participants based on study design.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sperry SL, Crockett SD, Miller CB, et al. Esophageal foreign-body impactions: epidemiology, time trends, and the impact of the increasing prevalence of eosinophilic esophagitis. Gastrointest Endosc 2011;74:985-91. [Crossref] [PubMed]
2. Lam AY, Lee JK, Coward S, et al. Epidemiologic Burden and Projections for Eosinophilic Esophagitis-Associated Emergency Department Visits in the United States: 2009-2030. Clin Gastroenterol Hepatol 2023;21:3041-3050.e3. [Crossref] [PubMed]
3. Hoversten P, Lomeli L, Broman AT, et al. Esophageal disease and loss to follow-up are common among patients who experience non-endoscopic resolution of food impaction. Dis Esophagus 2023;36:doac054. [Crossref] [PubMed]
4. Kerlin P, Jones D, Remedios M, et al. Prevalence of eosinophilic esophagitis in adults with food bolus obstruction of the esophagus. J Clin Gastroenterol 2007;41:356-61. [Crossref] [PubMed]
5. Chang JW, Olson S, Kim JY, et al. Loss to follow-up after food impaction among patients with and without eosinophilic esophagitis. Dis Esophagus 2019;32:doz056. [Crossref] [PubMed]
6. Kassim T, Gapp J, Walters RW, et al. Immediate dilation in esophageal food impaction is safe and effective but performed infrequently: observations from a large Midwest US cohort. Dis Esophagus 2020;33:doz084. [Crossref] [PubMed]
7. Guo H, Hamilton P, Enns E, et al. Postendoscopy Care for Patients Presenting With Esophageal Food Bolus Impaction: A Population-Based Multicenter Cohort Study. Am J Gastroenterol 2023;118:1787-96. [Crossref] [PubMed]
8. Rosenwald K, Pan Z, Andrews R, et al. Follow-up and symptom persistence after esophageal food impaction. Dis Esophagus 2021;34:doab029. [Crossref] [PubMed]
9. Prasad GA, Reddy JG, Boyd-Enders FT, et al. Predictors of recurrent esophageal food impaction: a case-control study. J Clin Gastroenterol 2008;42:771-5. [Crossref] [PubMed]
10. Hirano I, Furuta GT. Approaches and Challenges to Management of Pediatric and Adult Patients With Eosinophilic Esophagitis. Gastroenterology 2020;158:840-51. [Crossref] [PubMed]
11. Murray FR, Kreienbühl A, Straumann A, et al. Natural History of Patients Lost to Follow-up After Esophageal Food Impaction. Clin Gastroenterol Hepatol 2023;21:2413-4. [Crossref] [PubMed]
12. Messina FC, McDaniel MA, Trammel AC, et al. Improving specialty care follow-up after an ED visit using a unique referral system. Am J Emerg Med 2013;31:1495-500. [Crossref] [PubMed]
13. Arora S, Burner E, Terp S, et al. Improving attendance at post-emergency department follow-up via automated text message appointment reminders: a randomized controlled trial. Acad Emerg Med 2015;22:31-7. [Crossref] [PubMed]