Let me say from the onset, I am no proponent of pay-for-performance for veterinarians, yet when requested by the practice, I try to meet their desires. When examining a variety of pro-pay, many issues arise that complicate a seemingly straight-forward and intuitive premise: pay should somehow be tied to performance. The pay-for-performance experience, both in United States and abroad, has identified a number of unintended consequences that should be considered carefully by anyone advocating specific pay-for-performance approaches.
One of the challenges of all incentive programs is creating rules that can accidently create the wrong results. To help increase the likelihood that pay-for-performance approaches will accomplish the desired goals without inadvertent problems, this article present a number of lessons-learned from incentive programs that currently exist.
KNOW THE PITFALLS
A number of studies in human healthcare settings document the problems with well-intentioned pay-for-performance programs. Four categories are discussed below:
Gaming the System. In the U.K. a 2004 pay-for-performance program for primary care physicians was initiated; the program tracked 146 quality indicators related to clinical care, organization of care, and patient experience. Significant incentives were made available, giving primary care physicians a chance to increase their income by 25 percent. A review of the program’s first-year results showed that excluding patients from the denominator used in calculating performance was a common practice for increases in the performance factors. In this case, it was the documentation, rather than actual use of preventive services, that reflected improved performance.
In veterinary medicine, it was as simple as providers selling more large bags of pet food to increase the average client transaction (increase in gross sales but decrease in net income, which most of the veterinary practice management software systems do not report by provider or EOM summary reports).
Focusing on the Wrong Thing. This category was often known as “teaching to the test.” Providers who attempted to maximize the measured attributes of care often wasted resources at the expense of effective medical actions. A human healthcare example was the retinal eye exams for diabetes. The standards of care called for an annual screening, and many providers focused on patients who were two or three years between screenings. However, very few providers focused on those positive cases who never had any screening (approximately one third; this population would usually show greatly improved outcomes with the first screening.
The survey showed the lack of timely follow-up was the cause for suboptimal treatments, not the screening procedures themselves. Factors like poor follow-up (32 percent), delay in scheduling tests/treatments (14 percent) and rapid disease progression (23 percent), showed the relationship between a single metric (e.g., exam) vs the desired outcome(s).
In veterinary medicine, the excessive laboratory screening for Anesthetic Risk Level 1 patients, rather than the minimal required as listed by VIN, is such an example. Clients ended up paying far more for pre-anesthetic screening than they would for the VIN recommended laboratory level.
Overtreatment Bias. The standard of care stated Medicare pneumonia patients be treated for pneumonia within 4 hours of admission; 22 percent of the patients had been treated BEFORE the diagnosis of pneumonia. Treating patients due to a standards of care metric showed many patients were treated before a final diagnosis made, and therefore many were treated in error.
In veterinary practice, the lack of a readily available problem list in the EMR has often caused overtreatment, according to established metrics, that may not be available within the fiscal restraints of the client/patient relationship. Also, it is a rare practice that uses veterinary nurse telephonic follow-up for dispensed medications, to ensure duration of treatment is maintained and/or signs/symptoms have been mediated.
Setting the Wrong Targets. The fourth challenge is setting the proper level of performance. How high should this bar be? Should the relative improvement be scored, or absolute performance? How big should the reward be? If the bar is set to low, it may make the program attractive but unsustainable; this allows the provider to win the battle but lose the war.
In veterinary medicine, our experts state 50% of patients are overweight, and 85% of adult animals have a dental care need, yet it is a rare practice that even approaches a 50% active follow-up surveillance of either preventive condition. So we settle for the percentage of EMRs with Body Condition Scores (BCS 1 to 9) and/or Dental Grades (DG 1+ to 4+); again, many veterinary electronic medical record systems do not provide a historical trend chart for these two conditions at each visit to prompt better provider awareness/response.
AVOID THE PITFALLS
Several lessons about using financial incentives to improve performance can be learned from both inside and outside of healthcare settings.
Make the Plan Easy to Use. This means easy to understand, easy to explain, and easy to compute. The more complex the metrics, the more subject it is to gaming or argument. An inevitable trade-off exists here. Healthcare is a complex activity and trying to translate this complexity into a web of new metrics can lead to an intricacy that is unpersuasive and unintelligible. If the real goal is to change human behavior, the strategist must keep in mind the failings of human recognition. Selecting the right measure is different from finding the perfect measure.
In veterinary medicine, where most of our team has entered this profession as a “calling”, yet we concurrently underpay (veterinary nurse tenure in the profession is only about 5 years, before they leave for better paying jobs). The Australian award rates for anyone filling a veterinary nurse position (trained or not) start at about $20 per hour, plus a federally directed minimum superannuation (retirement fund), plus 4-week vacation, and they still have a low retention rate after 5 years (unless the spouse/partner has a good income). As such, I look for supplemental recognition systems to increase staff fiscal recognitions.
Focus on principles, not on rules. Taking the patient nutritional needs one step further, it is critical that the provider (veterinarian) refers weight control patients to the nutritional advisor (veterinary nurse). In most veterinary EMR software systems, this referral action is not tracked, so we revert to a small spiral pocket notebook carried by each veterinary nurse, with programs (nutrition, behavior, photonic pen, etc.) on one index and provider initials on the other index. The page shows a “hash mark” for every client referral from the specific provider for a specific program follow-up, and the pages are summarized by the practice manager monthly (weekly in large practices). Provider recognition at a staff meeting may be all that is needed for the most active provider referral (by program), while the nutritional advisors, behavior counselors, pain managers (photonic pen treatments), etc., can receive a fiscal reward from a portion of the increased program income.
Some practices use gift certificates for exceptional ideas or program participation. In practices that use movie or zoo tickets for client “thank you for the referral” notes, they also use the zoo/movie tickets for various staff recognitions.
Understand the Role of Money. The role of money as an incentive should not be overestimated or underestimated. The evidence is clear: some people respond positively to financial incentives, while others respond negatively. We just do not know in advance who are influenced and in what way. Fortunately, in our profession, personal recognition by the leadership is usually very effective. Unfortunately, any pay-for-performance program gets designed to effect a large number of providers and operate under the assumption that all providers are equally receptive to financial incentives in the same way the program’s designers think they are.
In veterinary practice, many primary providers respond positively to public recognition and seems sufficient to stimulate improved practice performance. Back to new graduates, one pay-for-performance system that seems to usually work is giving them 6 months to learn the practice flow and standards of care, then use their second six months as a personal measurement of performance (e.g., 20 percent of their personal production is the minimum expectation, and any percent over that amount becomes a baseline for establishing the following year’s base pay). When this method is also used for tenured associates, the stress of individual salary negotiations seems to disappear.
FUTURE OF PAY-FOR-PERFORMANCE
The veterinary industry payment system will eventually evolve into something more sophisticated than simple activity-based payment systems. It will move to reflect the relationship between money spent and value received, with a primary focus on client-center patient advocacy within the scope of the practice Vision, Mission Focus, and Core Values. The practice’s written Standards of Care must also be in alignment with these three factors. Practice staff (veterinarians and veterinary nurses) actually only have two options: (1) wait for a program to be forced on the practice culture or (2) or assist in the development of the practice culture that supports the practice’s vision. A savvy consultant will understand and assist in the development of a meaningful practice vision (see the Signature series monograph, Leadership Action Planner, in the VIN Bookstore, www.VIN.com), as well as meaningful standards of care (see the Signature series monograph, Standards of Care, also available from the VIN Bookstore). The best way for any incentive program to achieve benefits for it’s people and the practice, without exaggerating the pitfalls of unintended consequences, is to enhance the team participation in developing the practice program(s) and then for the leadership to examine how the fiscal results should look.
