Changing Physician Behavior – When "Incentives" Go Wrong

R. Myles Riner , MD, FACEP

Partner Emeritus

Published January 18, 2013

The recent 60 Minutes broadcast – “The Cost of Admission”, and a comparable article in the NY Times, have raised a very controversial issue in emergency medicine and put this issue prominently in the public eye. The issue is: when is it inappropriate to try to influence physician behavior, and when do incentives become coercive?

Clearly, there are times when protocols, policies, performance goals, and even financial incentives can be helpful in getting physicians to adopt practices that are in patients’ best interests. The bottom line, though, is that if the goals of these incentives are not directed first and foremost to improving outcomes and providing cost effective, efficient, patient directed, quality driven objectives; such incentives deserve to be viewed with great skepticism, if not condemned outright. Unfortunately, the line between appropriate incentives and inappropriate incentives is not always easy to determine without a very thorough review. Thus, a protocol, or software driven set of admission criteria, that incentivizes providers to admit patients who rarely need to be admitted rather obviously steps over the line; but a protocol that encourages a physician to reconsider admission in patients who meet certain criteria in order to reduce the likelihood of an adverse outcome if the patient is discharged home could well be in patients’ best interests.

Likewise, a financial incentive that encourages physicians to over-order tests and treatments that the patient doesn’t need in order to maximize the revenues generated by that physician is almost certainly inappropriate; but a financial incentive that reimburses emergency physicians based, to some degree, on their productivity might be in patients’ best interests in the ED if it results in a lower time to provider and a higher sense of satisfaction and fairness for those physicians who are more productive. I say ‘might’ because financial incentives, in particular, can have unexpected results. For example, too great a ‘productivity’ incentive in an ED with double or triple coverage can encourage the EPs to ‘cherry pick’ patients based on the potential to generate revenue if there are no complementary policies in place to ensure that care is provided to patients in the most appropriate order, and workflow is distributed reasonably.

Linking incentives to quality, cost effective care is complicated, tricky, and some would even say potentially dangerous. So how do we know when a policy or protocol or incentive meets the ‘best interests of patients’ objective? Are there clues or criteria or indicators of incentives that might be unreasonable or inappropriate? The National Health Care Purchasing Institute put out a report on “The Growing Case for Using Physician Incentives to Improve Health Care Quality” that list some of the ‘key factors that influence the extent to which an incentive strategy is effective in improving physicians’ performance’, but the focus of this paper is more about making incentives work than vetting incentives for appropriateness. Nonetheless, one can draw some useful criteria for approaching an incentive policy or process with skepticism, based on these ‘key factors’, such as:

If the providers do not have a high level of trust in those developing or implementing these incentives.
If the financial incentive is substantial, or represents a significant portion of overall compensation.
If there is lack of transparency to providers and patients in the incentive process.
If there is distrust of the accuracy of the data that forms the basis of the incentive.
If there is no perceived need by physicians for adoption of the incentive or policy.
If medical leadership is not fully supportive of the policy (provided they are not themselves compromised by the incentive).
If physicians can not clearly define the benefit to the patient from adoption of the practices that are incented.
If the protocol or policy is complicated, indirect, or black-box.

I would add some more, particular to but not exclusive to, emergency medicine:

If the policy or process appears to try to skirt EMTALA regulations
If the process appears to discriminate based on ability to pay or insurance status
If threats of loss or employment or opportunity to earn a living are used or implied, especially if not mitigated by appropriate due process
If appropriateness of care is not the first and foremost objective of the incentive
If the incentive makes providers feel like they are ‘between a rock and a hard place’ (there’s enough of that in the routine practice of emergency medicine)
If non-physicians created the policy or protocol, or the physicians who did are ‘unnamed’
If the policy or incentive is not subject to modification by the physicians affected
If there is no ongoing mechanism to assess the impact of the policy or incentive on outcomes or care quality
If the policy or incentive has not been vetted by a fully independent third party for compliance with applicable rules and regulations
If there is insufficient evidence-based support for the policy

I am not inclined to draw any conclusions about the validity of the 60 Minutes allegations of impropriety in the HMA admissions incentives and facilitating Pro-med software without a full airing of the details of the case, which will likely come out in the course of the Government’s fraud investigation. The decision to provide and receive care in an inpatient setting carries a lot of financial consequences (and some risks) for patients, and fortunately there is almost always a second opinion rendered on this decision by the hospitalist or specialist or PMD who actually admits the patient to their service. I will say that I do not think even the most sophisticated software should ever supplant physician judgment when it comes to deciding whether to admit or discharge an ED patient. I also suspect that software which automatically orders up a battery of tests on a patient based on their age and chief complaint is likely to run at cross purposes to the objective of cost-effective care in the ED, even if the provider has the opportunity to retroactively cancel these orders, and even if using the software ‘facilitates’ patient flow.

Incentivizing physician behavior is not inherently bad or good, but it can all too easily invoke moral, ethical and legal dilemmas, and demands transparency, careful scrutiny, and thoughtful application, especially in the Emergency Department setting.

Partnering to improve patient lives

Vituity branding orange wave pattern background