Too Much Surgery: Overcoming Barriers to De-Implementation of Low-Value Surgery

Unnecessary healthcare in the United States represents one of the greatest sources of excessive spending and preventable harm to our patients. By some estimates, over $100 billion is spent (i.e., wasted) on unnecessary services each year.¹ Recognizing an opportunity to improve care and reduce spending, the Choosing Wisely® campaign galvanized a movement to reduce low-value, unnecessary care. With recommendations against routine use of over 600 low-value healthcare services, this effort is credited with modest success in reducing some imaging, laboratory testing, and medical care.²

While 17 surgical societies participated in Choosing Wisely®, less than 5% of recommendations target low-value surgical procedures.^3,4 Not operating on patients with little or nothing to gain from surgery might seem like an obvious win as we combat wasted healthcare spending. However, even in cases where surgery is clearly low-value per guidelines or randomized trials, de-implementation is not assured. For example, despite recommendations against contralateral prophylactic mastectomy for average-risk women with unilateral breast cancer rates continue to increase.⁵ Similarly, knee arthroscopy continues to be commonly performed for patients with osteoarthritis, meniscal tears, and knee pain despite multiple randomized trials showing no benefit as compared with medical management.⁶ In most circumstances, the multilevel factors (patient, provider, and system) contributing to persistent overuse are unknown. These could be supply-side driven (e.g., surgeon habits and training, financial, concerns about malpractice, and industry influence), demand-side driven (e.g., patient expectations and preferences, perceived pressure from referring providers), or a combination of the two.

The higher stakes for patients undergoing low-value surgery compared to non-procedural services create mounting pressure to address these uncertainties. The objectives of this surgical perspective are to discuss the unique aspects of surgical care delivery acting as barriers to de-implementation and suggest potential strategies to reduce low-value surgery in the United States.

Specific considerations for de-implementation of low-value surgical practices

Earlier widespread adoption of procedures before establishing efficacy

In contrast to the pathway to market for new drugs, new surgical procedures undergo limited to no efficacy-based scrutiny prior to clinical use. This can lead to early adoption and entrenched practice before efficacy or value is known. In these cases, surgeons may be reluctant to accept emerging evidence of limited efficacy or value in favor of their anecdotal personal experiences with the procedure. For instance, use of robotic-assisted procedures has increased nearly three-fold despite limited high-quality evidence demonstrating superiority over other minimally invasive and less costly approaches.⁷ Although the robotic platform may be of high-value in some cases, there is substantial evidence that its use has encroached upon or replaced higher value alternatives for a number of surgical indications.

The Lake Wobegon Effect

In interpreting surgical trials, surgeons may more readily attribute a lack of efficacy to the inferior surgical skill of the participating surgeons as compared to their own. In the Social Psychology literature, “The Lake Wobegon Effect” is the natural tendency to overestimate one’s abilities as compared to others, and is more likely to occur when performance is difficult to measure objectively, such as driving ability.⁸ This can lead to overuse if surgeons believe the procedure to yield superior value “in their hands.” In contrast, drug effectiveness, or lack thereof, is generally not felt to be due to the prescribing ability of particular providers. The Lake Wobegon Effect can also contribute to overtreatment when it’s applied to the patient - surgeons may continue to offer ineffective procedures believing that the unique or special nature of their patient will lead to them benefiting despite no benefit for the average patient. It is important to note that certain subgroups of patients may benefit from these surgical interventions, therefore surgeons and investigators should evaluate for heterogeneity of treatment effects in both clinical trials and observational studies.

Financial incentives to do more, irrespective of value

Perhaps the most significant barrier to de-implementation of low-value surgery in the United States is the prevailing fee-for-service payment system. This approach incentivizes surgeons to operate without an explicit requirement for value. In contrast, payment models for non-proceduralists are based on evaluation and management coding and generally not affected by billing of additional services. For example, a pediatrician seeing a child with an upper respiratory tract infection is not financially incentivized to prescribe antibiotics. In contrast, a surgeon seeing a patient with an asymptomatic inguinal hernia is financially incentivized to offer surgical repair in favor of a watch-and-wait strategy. While de-implementation of low-value surgery may have secondary effects of increasing access for higher value surgery or reducing surgeon burnout, in our current payment system they are generally perceived as negatively impacting the bottom line of surgeons and hospitals.

Inability to leverage existing de-implementation strategies

Strategies that have been successful in achieving de-implementation for non-procedural services may not apply to low-value surgical procedures. Among the most effective are restrictive ordering and point-of-care decision support tools, neither of which is easily adapted to current perioperative workflow. Formulary restriction can significantly reduce antibiotic overuse while point-of-care decision support can help clinicians identify low-value imaging before they order it.⁹ The current surgery corollary—insurer restriction through pre-authorization—often lacks the detailed clinical discrimination needed to significantly drive de-implementation. In fact, fewer than 2% of surgical procedures are declined by payers.¹⁰ The ability of other strategies to reduce the use of low-value surgery is simply unknown. In a systematic review of trials testing de-implementation strategies directed at low-value care, only 10% of studies evaluated a strategy targeting a surgical procedure.⁹ Finally, due to a perceived likelihood of “surgeon shopping” by patients demanding a low-value procedure, the threshold to offer the procedure may be lower in competitive health care markets.

Potential solutions and future directions

Understand determinants of low-value surgery

To design and employ effective strategies for de-implementation of low-value surgery, we must first understand the factors influencing its utilization. An improved understanding of factors influencing both surgeon and patient decision-making is needed. Multilevel factors (patient, provider, and system) also contribute to overuse, some of which may be outside of surgeons’ control. Application of consolidated behavioral and organizational frameworks from implementation science can help identify determinants of overuse and guide development and evaluation of strategies for de-implementation across diverse practice settings. For example, adapting the Tailored Implementation of Chronic Disease (TICD) framework to understand persistent use of a low-value procedure takes into consideration a range of potential determinants including guideline factors, surgeon and patient-level factors, professional interactions such as referral patterns, incentives and resources, capacity for organizational change, and other social, political or legal factors. Understanding these determinants of low-value surgery ensures that we use the most effective strategies for de-implementation.

Leverage quality collaboratives to monitor and reduce low-value surgery

Quality collaboratives such as the National Surgical Quality Improvement Program (NSQIP) have focused on improving quality through an emphasis on the structures and processes that may lead to improved outcomes for a given surgical procedure. As these quality improvement processes continue to mature, the existing infrastructure of these collaboratives could also be leveraged to measure surgical appropriateness and overuse to support de-implementation of low-value surgery. These groups may serve as a platform for testing strategies for de-implementation. Similar approaches are currently applied to improve quality of care for patients receiving percutaneous coronary interventions in Michigan through the Blue Cross Blue Shield Cardiovascular Consortium. Furthermore, the Surgical Care and Outcomes Assessment Program (SCOAP) of Washington State effectively scaled a benchmarking and education initiative to improve appropriateness of elective colon resection for diverticulitis across 49 hospitals.¹¹

Implement clinical decision support systems for surgeons and health systems

Clinical decision support tools may reduce use of low-value surgical procedures. These tools aim to promote care that is consistent with treatment guidelines, to provide clinicians with evidence-based protocols, or to require second opinions or oversight when surgical decisions are discordant with established appropriateness criteria. Mechanisms for decision support often integrate within electronic health records or utilize portable applications or decision trees. To be effective for surgical decision-making, these tools must fit into the clinical workflow and be acceptable to surgeons and other practitioners. Mandatory second opinions may be effective to reduce utilization of elective low-value surgery. For example, in a randomized trial, requiring a second opinion from another physician was demonstrated to reduce rates of cesarean sections without affecting maternal or perinatal mortality.¹²

Design, implement, and evaluate incentives for de-implementation

As we transition from fee-for-service towards value-based payment models, local health systems and individual surgeons may be financially incentivized to monitor and limit low-value care. However, the impact of these policies will be limited on low-value care as long as surgeons remain accountable to and incentivized by productivity-based compensation models at the institution level. In order to ensure that the profession retains necessary autonomy in surgical decision-making, surgeons and their societies should be actively involved in designing, implementing, and evaluating those systems. Innovative models of reimbursement, such as Value Based Insurance Design (V-BID), will adjust levels of patient cost sharing with perceived value of medical services (i.e. high-value services will cost less and low-value services will cost more for patients). It behooves the surgical profession to be actively involved in designing those systems to promote appropriate care and to minimize any unintended consequences for patients.

Although the Choosing Wisely® campaign has inspired a movement to identify and reduce the use of low-value healthcare services, a more specific understanding of the barriers that exist within surgery is overdue. Moving forward, successful de-implementation of low-value surgery will depend on understanding why low-value procedures persist, leveraging quality collaboratives to monitor appropriateness, implementing clinical decision support systems, and developing incentives for de-implementation. As stewards of the healthcare system and advocates for patient care, surgeons must lead in identifying and overcoming barriers to de-implementation of low-value surgery.

Abbreviations:

TICD

Tailored Implementation of Chronic Disease

NSQIP

National Surgical Quality Improvement Program

V-BID

Value-Based Insurance Design