At every stage of surgery—preoperative, intraoperative, postoperative—data is collected. But if you don’t have strategies to provide surgical insights and intelligence, the data remains discrete, isolated, and of limited use.

Data-driven surgery, based on the fast growing field of surgical data science (SDS), applies AI, automation, and analytics to deliver insights for surgeons, perioperative leaders, schedulers, and risk managers.

Data-driven surgery principles apply to all types of surgical and interventional procedures across General, Orthopedic, Neuro and Cardiovascular procedures. Data is synthesized from a number of sources and stages of care, including imaging data, video recordings, device output, the EHR, patient risk factors, vitals, length of stay, and readmissions. It also incorporates data from inventory and financial systems. SDS methodologies are then applied to create clinical, operational, and financial objectives and key results (OKRs) to streamline workflows, enhance care delivery, allocate resources more efficiently, and improve patient outcomes overall.

The Evolution of Data-driven Surgery

In 2004, researchers in computer-aided surgery organized “OR2020”: a workshop to define the OR of the future. The researchers concluded, “The operating room of the future will require integrated systems and technological developments to improve surgical workflow and provide better patient care.”^[1] A similar workshop in 2016 defined “surgical data science,” noting, “data may pertain to any part of the patient care process (from initial presentation to long-term outcomes)” and “may concern the patient, caregivers, and/or technology used to deliver care.”^[2] Although the seeds for data-driven surgery had been planted, “in 2019, an international poll revealed that no commonly recognized surgical data science success stories exist to date,” and many of the challenges identified in 2004 persisted.^[3] For data-driven surgery to take root and succeed, the traditional surgical mindset had to shift. As researchers observed in 2017, “a data-centric cultural shift is necessary to effectively integrate SDS into surgical patient care and training.”^[4]

We are now at a point where this shift is happening, as more hospitals worldwide implement data-driven surgical platforms. Surgeons are using a data-driven approach to improve their surgical decision-making based on their evolving practice. Data-driven surgical solutions store subjective and objective surgery assessments for analysis, which allows care delivery that is personalized for individual patients.

By collecting and synthesizing disparate data elements, this modern methodology provides enriched patient and surgical records. Performance data can be analyzed by hospital, surgeon, case type, and/or a number of other parameters, using AI and automation to provide benchmarking and predictive analytics to drive process improvements for healthier patients, more efficient OR teams, and better reimbursement.

Reducing Unnecessary Surgical Variation

The term “digital surgery” is most closely associated with robotics and proprietary technology. To fully incorporate the latest tech and AI into the surgical process, we need to embrace “data-driven surgery,” which moves beyond robotics and vendor-specific data puddles to encompasses the entire surgical continuum in a vendor-neutral data lake.

A key benefit of this approach is that it uncovers unnecessary variation in the surgical process. Unnecessary variation can result in significant financial losses.^[5] Variation also affects a hospital network’s brand and reputation. While individual hospitals made the US News & World Report “honor roll” for best care, the networks they belonged to showed substantial variation in care.^[6] A recent study on practice variation in JAMA Health Forum states that “understanding the sources of these variations”—what a data-driven surgery approach makes possible—”may inform efforts to improve the value of care.”^[7] These knowledge gaps affect not only patients and surgeons, but also payers and healthcare provider organizations.^[8]

“Clinicians and policy makers are increasingly exploring strategies to reduce unwarranted variation in outcomes and costs,” according to a 2023 nationwide analysis. “Adequately accounting for case mix and better insight into the levels at which variation exists is crucial for such strategies.”^[9] A data-driven surgical approach gives healthcare providers the tools to uncover these insights.

Data-driven surgery analyzes inefficiencies in the operating room (OR)—the source of both a hospital’s greatest revenue and highest costs. A recent study on addressing financial challenges in the OR setting states, “Efficient operating room (OR) management is a key driver of a hospital’s economic success.”^[10] To achieve this financial goal, “OR planning should be based on real-life data at every stage and should apply newly developed algorithms.” ^[11] By collecting and analyzing data in real time before, during, and after surgical procedures, a data-driven approach generates actionable, timely insights.

For example, maximizing operating room utilization is an ongoing challenge for OR scheduling efficiency and a hospital’s profitability. Block marketplaces, data, and AI democratize the OR schedule and create an open and transparent market to balance supply and demand for OR time. It increases surgeon access to operating time by proactively sending surgeons and schedulers alerts and a workflow to release or schedule a case. This automated process regularly monitors OR blocks for efficiency.

In another example, optimizing clinical effectiveness and teamwork dynamics is a key challenge facing clinical and operational leaders. Creating a high-fidelity surgical record that included video, imaging and data and applying surgical data science techniques can identify key opportunities for optimizing surgical techniques, maximizing throughput and improving teamwork dynamics.

AI and machine learning have transformed care in a number of clinical settings. As this technology is increasingly deployed in ORs around the globe, HCOs will realize the potential to transform surgical outcomes, scheduling and staffing, the payer-provider relationship, hospital administration, quality and risk management, and patient outcomes. The end result being a reduction in costs, increases in productivity, and healthier patients.

References:

[1] Kevin Cleary, Ho Young Chung, Seong K Mun. OR2020 workshop overview: operating room of the future. International Congress Series, Volume 1268. 2004. Pages 847-852. ISSN 0531-5131. https://doi.org/10.1016/j.ics.2004.03.287.

[2] Maier-Hein, L., Vedula, S.S., Speidel, S. et al. Surgical data science for next-generation interventions. Nat Biomed Eng 1, 691–696 (2017). https://doi.org/10.1038/s41551-017-0132-7

[3] Maier-Hein L, Eisenmann M, Sarikaya D, et al. Surgical data science – from concepts toward clinical translation. Med Image Anal. 2022 Feb;76:102306. doi: 10.1016/j.media.2021.102306. Epub 2021 Nov 18. PMID: 34879287; PMCID: PMC9135051.

[4] Vedula SS, Hager GD. Surgical data science: The new knowledge domain. Innov Surg Sci. 2017 Apr;2(3):109-121. doi: 10.1515/iss-2017-0004. Epub 2017 Apr 20. PMID: 28936475; PMCID: PMC5602563.

[5] Nouhi M, Hadian M, Jahangiri R, Hakimzadeh M, Gray S, Olyaeemanesh A. The economic consequences of practice style variation in providing medical interventions: A systematic review of the literature. J Educ Health Promot. 2019 Jun 27;8:119. doi: 10.4103/jehp.jehp_386_18. PMID: 31334271; PMCID: PMC6615132.

[6] Sheetz KH, Ibrahim AM, Nathan H, Dimick JB. Variation in Surgical Outcomes Across Networks of the Highest-Rated US Hospitals. JAMA Surg. 2019;154(6):510–515. https://doi.org/10.1001/jamasurg.2019.0090

[7] Song Z, Kannan S, Gambrel RJ, et al. Physician Practice Pattern Variations in Common Clinical Scenarios Within 5 US Metropolitan Areas. JAMA Health Forum. 2022;3(1):e214698. https://doi.org/10.1001/jamahealthforum.2021.4698

[8] Song Z, Kannan S, Gambrel RJ, et al. Physician Practice Pattern Variations in Common Clinical Scenarios Within 5 US Metropolitan Areas. JAMA Health Forum. 2022;3(1):e214698. https://doi:10.1001/jamahealthforum.2021.4698

[9] Nèwel Salet, Vincent A. Stangenberger, Rolf H. Bremmer, Frank Eijkenaar. Between-Hospital and Between-Physician Variation in Outcomes and Costs in High- and Low-Complex Surgery: A Nationwide Multilevel Analysis. Value in Health, Volume 26, Issue 4. 2023. Pages 536-546. ISSN 1098-3015. https://doi.org/10.1016/j.jval.2022.11.006.

[10] Corina Bello, Richard D. Urman, Lukas Andereggen, Dietrich Doll, Markus M. Luedi. Operational and strategic decision making in the perioperative setting: Meeting budgetary challenges and quality of care goals. Best Practice & Research Clinical Anaesthesiology, Volume 36, Issue 2, 2022, Pages 265-273, ISSN 1521-6896, https://doi.org/10.1016/j.bpa.2022.04.003.

[11] Corina Bello, Richard D. Urman, Lukas Andereggen, Dietrich Doll, Markus M. Luedi. Operational and strategic decision making in the perioperative setting: Meeting budgetary challenges and quality of care goals. Best Practice & Research Clinical Anaesthesiology, Volume 36, Issue 2, 2022, Pages 265-273, ISSN 1521-6896, https://doi.org/10.1016/j.bpa.2022.04.003.