Secondary Logo

Show Me the Data! A Perioperative Data Warehouse of Epic Proportions

Wanderer, Jonathan P. MD, MPhil; Poler, S. Mark MD; Rothman, Brian S. MD

doi: 10.1213/ANE.0000000000001321
Editorials: Editorial
Free

From the *Departments of Anesthesiology and Biomedical Informatics, Vanderbilt University School of Medicine, Nashville, Tennessee; Division of Anesthesiology, Geisinger Health System, Danville, Pennsylvania; and Department of Anesthesiology, Vanderbilt University School of Medicine, Nashville, Tennessee.

Accepted for publication March 2, 2016.

Funding: Dr. Wanderer is supported by funding from by the Foundation for Anesthesia Education and Research (FAER, Schaumburg, IL) and Anesthesia Quality Institute’s (AQI, Schaumburg, IL) Health Service Research Mentored Research Training Grant (HSR-MRTG).

The authors declare no conflicts of interest.

Reprints will not be available from the authors.

Address correspondence to Brian S. Rothman, MD, Department of Anesthesiology, Vanderbilt University Medical Center, 1301 Medical Center Dr., 4648 TVC, Nashville, TN 37232. Address e-mail to brian.rothman@vanderbilt.edu.

“Show me the money!

I need to feel you, Jerry!

Show me the money!

Jerry, you got to yell!

[screaming] Show me the money! Show me the money!”

Jerry Maguire, 1996

Over the last few decades, the development of electronic health records (EHRs) has unleashed waves of change across our health care environment, ultimately transmogrifying into a tsunami of digitization after the passage of the American Recovery and Reinvestment Act in 2009. Fueled by $19 billion of incentives, EHRs quickly rushed into medical centers across the nation, attempting to wash away paper charts, leaving keyboards and mice in their wake. We are witnessing a transformation of our health care landscape where tectonic changes in reimbursements are shifting fee-for-service to value-based purchasing and other payment models. We must now demonstrate how well our patients have done, not what we have done to them. We will be paid according to measurements of our performance, navigating the poorly charted waters of “no outcome, no income.”1

The push for digital records has contributed to the rapid adoption of anesthesia information management systems (AIMS), which are now implemented in 75% of US academic anesthesiology departments.2 These AIMS are a necessary, but by no means sufficient, requirement for understanding and reporting on perioperative patient care and outcomes. The central construct of most AIMS is an anesthetic record, which has an intrinsic relationship to a specific patient and encounter. The simplicity ends here, however, because 1 patient may have multiple medical record numbers (i.e., an emergent trauma admission), encounters may or may not include a hospital admission, hospital admissions may be comprised multiple encounters, overnight stays may occur as an outpatient or as an inpatient, and encounters may be associated with multiple anesthetics. These variable relationships make answering simple questions such as, “How many gastric bypass patients had postoperative nausea and vomiting?” potentially very difficult. Additionally, merging AIMS and EHR data add additional layers of complexity, requiring multiple, distinct sources of information (tables) to be knitted together (joined) in just the right way (query) for the resulting output (report) to be interpretable and correct. Building accurate reports takes time, money, and informatics expertise, and it is no longer an optional activity—unless, of course, your practice is willing to take reimbursement decreases that incrementally reach 9% by 2021.3

To be sure, there are centers of excellence in anesthesia informatics, where innovation has enabled remarkable projects to take place. We have witnessed in recent years, for instance, the emergence of large-scale perioperative outcomes research,4 near real-time automated feedback on anesthetic performance for trainees,5 automated systems for evaluating cancelled cases,6 dynamic risk-profiling systems leveraging large-scale historic data,7 and a variety of real-time decision support systems.8 Most of these innovations, however, have occurred in centers using best-of-breed systems that were customized in some way, not with off-the-shelf software. A refrain heard frequently at the Society for Technology in Anesthesia annual meetings: “That’s great for you, but we have Epic” (Dr. Wanderer, personal communication).

It is through these murky and obstacle-filled waters that Hofer et al.9 tread. The University of California Los Angeles’ Epic reporting data warehouse is comprised >15,000 tables. Hofer et al. developed a methodology to reduce the time, cost, and informatics expertise necessary to work effectively in this environment by recreating the fundamental constructs necessary for perioperative outcomes work in AIMS systems: the anesthetic, the patient, and the encounter. They did this by writing queries to join tables in specific and useful ways, validating their approach with real-world examples as their work progressed. On top of the cornucopia of underlying data structures, they built a layer of base tables and on that a series of reporting tables. The result? If you want to answer the question about postoperative nausea and vomiting in gastric bypass patients, instead of carefully and tediously lining up 17 tables in just the right manner, you only need to look in 1.

We are in the era of the triple aim: reduce health care costs, improve health care quality, and improve patient outcomes. “Show me the data!” is a cry heard everyday in our medical centers, shouted on high by our public payers, private payers, and hospital administrators. The importance of being able to do this will only grow with time. However, do it we must if we plan to stay relevant in these changing times. Hofer et al. demonstrate that having Epic is not an excuse for avoiding innovation and that much can be done within that environment. If it can be done in Epic, it can be done at scale. It will be epic.

There is a saying within our EHR world: once you know one Epic installation, you know one Epic installation. It remains to be seen if Hofer et al.’s methodology is generalizable or if there are other approaches that would work as well or better. With other institutions seeking similar access to perioperative data, it is essential for this approach to be replicated at other institutions. With >50% of the US population having a medical record in an Epic system, this cannot occur too soon. We urge Hofer et al. to share their underlying query code, the secret sauce that has revealed their patient outcomes. Show us the data, and show us how!

Back to Top | Article Outline

DISCLOSURES

Name: Jonathan P. Wanderer, MD, MPhil.

Contribution: This author helped write the manuscript.

Attestation: Jonathan P. Wanderer has approved the final manuscript.

Name: S. Mark Poler, MD.

Contribution: This author helped write the manuscript.

Attestation: S. Mark Poler has approved the final manuscript.

Name: Brian S. Rothman, MD.

Contribution: This author helped write the manuscript.

Attestation: Brian S. Rothman has approved the final manuscript.

This manuscript was handled by: Maxime Cannesson, MD, PhD.

Back to Top | Article Outline

REFERENCES

1. Denham CR. The no-outcome no-income tsunami: surviving pay-for-performance. Focus on patient safety. Nat Patient Saf Found 2004;7:1–3.
2. Stol IS, Ehrenfeld JM, Epstein RH. Technology diffusion of anesthesia information management systems into academic anesthesia departments in the United States. Anesth Analg 2014;118:644–50.
3. Manchikanti L, Staats PS, Boswell MV, Hirsch JA. Analysis of the carrot and stick policy of repeal of the sustainable growth rate formula: the good, the bad, and the ugly. Pain Physician 2015;18:E273–92.
4. Kheterpal S. Clinical research using an information system: the multicenter perioperative outcomes group. Anesthesiol Clin 2011;29:377–88.
5. Ehrenfeld JM, McEvoy MD, Furman WR, Snyder D, Sandberg WS. Automated near-real-time clinical performance feedback for anesthesiology residents: one piece of the milestones puzzle. Anesthesiology 2014;120:172–84.
6. Starnes JR, McEvoy MD, Ehrenfeld JM, Sandberg WS, Wanderer JP. Automated case cancellation review system improves systems-based practice. J Med Syst 2015;39:134.
7. Was A, Wanderer J. Matching clinicians to operative cases: a novel application of a patient acuity score. Appl Clin Inform 2013;4:445–53.
8. Wanderer JP, Sandberg WS, Ehrenfeld JM. Real-time alerts and reminders using information systems. Anesthesiol Clin 2011;29:389–96.
9. Hofer IS, Gabel E, Pffeffer M, Mahbouba M, Mahajan A. A systematic approach to creation of a perioperative data warehouse. Anesth Analg 2016;122:1880–4.
© 2016 International Anesthesia Research Society