How Should Outpatient Wound Clinics Honestly Measure Success?

How we evaluate ourselves today will impact whether patients can see us tomorrow. 

Everyone is probably familiar with TV advertisements for cancer treatment centers. Do any of them advertise that they cure more than 90% of all cancers? If they did, they would lose all credibility. Everyone knows that some types of cancer are more serious than others and that some cases are more advanced. That is why oncologists report cancer survival by type and stage, which is an example of reporting outcomes by risk stratification. However, it is not possible to determine if a cancer center is doing a “good” or a “poor” job if the only information available is the survival rate of the patients treated there. We would also need to know the expected survival rate for similar patients. To understand “success” of treatment, the severity of the condition, the expected outcome, and how the reported outcome compares to the expected outcome must be known. Because national data are available on cancer survival rates, it is possible for the best treatment centers to make credible claims that their outcomes are better than the national average and for centers to compare results to their peers. These performance data can then be used by patients to decide where they want their treatment. There’s another benefit to risk stratification: the oncology field continues to be well funded and pharmaceutical companies can justify the cost of better drugs when nationally reported data show that outcomes for a specific cancer are poor and need improvement. Nobody sees the low survival rate of pancreatic cancer and assumes that oncologists who treat that type of cancer must be especially poor doctors. Instead, when outcomes are honestly reported, a disease with consistently poor outcomes becomes a target for innovation rather than evidence of poor performance. It is time for wound care practitioners to adopt the same approach to measuring success for their clinics. Reporting outcome by risk stratification allows valid comparisons of practitioner performance across the country, enables us to identify where investment is most needed, and provides an opportunity for innovative products to demonstrate their value proposition. 

PUBLIC REPORTING OF HEALING RATES

Healing rate seems like an obvious thing for wound centers to measure. That is why most have publicly reported healing rates as a measure of success for decades. Although patients may be content to “manage” diabetes or high blood pressure, they want wounds to heal. The problem is that wounds are not a disease, they are a symptom of a disease (or many diseases). So, whether a wound heals is often determined by how sick the patient is, not just by wound type or severity. Without a way to stratify both the wound(s) and the patient, clinicians caring for the most difficult cases will appear to have worse outcomes than those caring for “easier” cases. In the absence of a reliable method of risk stratification, wound centers devised a clever strategy –  report that all wounds heal. Recently, a systematic analysis was performed of publicly reported wound healing rates. Data were available from wound centers in 44 states, 85% of which reported a mean wound healing rate of 92% (usually within five weeks).¹ This, frankly is not possible. We know this because a systematic analysis was then performed on the healing rates of control subjects in clinical trials, using them as the “best possible case” for real-world healing. Among 48 randomized controlled trials (RCTs), there were 2,624 control wounds with an overall healing rate of only 40%, which is less than half the healing rate reported by nearly every wound center in the United States. However, the truth is worse than that. “Real-world” patients are not as healthy as control subjects. Patients treated at a consortium of six wound centers were compared to subjects enrolled in RCTs of cellular and/or tissue-based products (CTPs) performed in those centers.²  The “real” patients living with venous leg ulcers (VLUs) were 10 years older and often lived with either peripheral arterial disease (PAD) or congestive heart failure, both of which were excluded in the trials. Also, the mean size of real-world  VLUs was five times the size of  VLUs enrolled in the clinical trials. RCTs for diabetic foot ulcers (DFUs) were performed on Wagner Grade I and II ulcers, but 44% of real-world patients had Grade III DFUs that were three times larger than those in RCTs. Table 1 compares the healing rate of DFUs, VLUs, and pressure ulcers in controlled trials with healing rates of real-world ulcers in the U.S. Wound Registry (USWR) at 12 weeks. The average healing rate for all wounds in the USWR is only about 35%. In other words, there is simply no way that the 92% rate is possible. Healing rate, as it is currently reported by U.S. wound centers is meaningless as a measure of success. These “fantasy” healing rates are a much bigger problem than simply misrepresenting the facts. Wound center directors have been doing exactly the opposite of what is best for them and the field of wound care. Their misguided logic has been that because very sick patients have poor healing rates, the data on these patients should not be reported because it will make the clinic and practitioners look bad. Meanwhile, the majority of healthcare dollars are being used on those very sick patients to improve their (not reported) wound outcome. Under the new Quality Payment Program (QPP), “cost” is an increasingly important component of the Merit-Based Incentive Payment System (MIPS), and the Centers for Medicare & Medicaid Services (CMS) can monetarily penalize practitioners for “overspending.” It is vital that we report data on the sickest patients, especially those who do not heal, in order to justify the Medicare dollars spent on their care. If wound care practitioners want to secure payment for future services, they must report outcomes of the sickest patients, including those who do not heal. 

HOW SHOULD WE MEASURE & REPORT HEALING RATES?

We cannot continue reporting fantasy healing rates. The best way to measure healing rate is in relation to the predicted likelihood of healing. That is why the Wound Healing Index (WHI)³ was developed in a joint project between the Institute for Clinical Outcomes Research, a not-for-profit organization devoted to research focused on improving the quality of evidence to guide surgical treatments, and the USWR. Factors in the WHI model include patient- and wound-related variables such as patient age, wound age, wound size, wound depth, wound severity, and whether there is evidence of infection or bioburden. Other factors found to be significant in the model were the total number of wounds of any type present on the body, the method of arrival to clinic (eg, unaided ambulation, wheelchair, stretcher), and whether the patient required hospitalization. Specific patient factors were predictive in certain wound types (eg, dialysis, renal transplant, PAD). 

Honest reporting of healing rates allows us to think about wound care a bit like cancer treatment. It is possible to create three categories within each “model” that roughly represent the following: wounds highly likely to heal (be cured), wounds highly unlikely to heal (probably cannot be cured), and wounds that could heal (see Figure). 

Providers may not be able to do much to change the outcome of wounds in the category of “unlikely to heal.” Think of this category as pancreatic cancer. On the other hand, if a treatment or product did improve the outcome of wounds in this category, it would be easy to demonstrate value. As for wounds in the category “likely to heal,” most providers should be successful with these wounds, so we are not really impressed if a provider has a high healing rate with them since they were expected to heal. That means the real test for practitioners is how they perform with the middle group, whose outcomes can be changed with aggressive, protocolized care and/or with advanced therapeutics. Using this method of reporting, it is possible to compare practitioner and/or wound center performance fairly. How could we implement standardized outcome reporting across the country? Most programs that involve outcome reporting at the national level are those mandated by the federal government. Unfortunately, none of the quality measures that a hospital-based outpatient department (HOPD) is required to report to CMS are relevant to wound care.⁴ Other specialties, including cardiology, have developed national standards and accreditation programs thanks to the availability of quality measures for relatable conditions. Wound care is not a recognized medical specialty⁵ and has no national physician quality measures, so it would be difficult for wound centers to replicate this opportunity. Recognizing that there were gaps in the field of quality measures, CMS created qualified clinical data registries (QCDRs) in 2013 that are empowered to develop specialty-specific measures. Using data from quality projects it had already pilot tested, the USWR developed a suite of wound care-relevant measures in conjunction with the Alliance of Wound Care Stakeholders and its member organizations. For Year 2 of MIPS, CMS has approved 13 wound care-relevant quality measures through the USWR. Some of these measures would work equally well for the HOPD. Table 2 lists the measures available in 2018. 

DOING “THE RIGHT THING”

In 2010, an analysis of data showed that only 17% of VLUs got evidence-based compression at each visit during 108,000 visits to 18 outpatient wound centers in 16 states, and most patients were treated with tubular bandages, elastic bandages, or were told to elevate.⁶ Even worse, only 6% of DFUs had any kind of offloading recorded. These patients did get expensive interventions such as hyperbaric oxygen therapy (HBOT) and CTPs. When DFU offloading wasn’t documented, the cost of care doubled. In response, the USWR initiated its Do the Right Thing^™ project that targeted arterial screening, DFU offloading and  VLU compression as part of a quality initiative.⁷ It is likely that wound care practitioners will see these three metrics as a very low bar by which to measure a wound care program’s success. However, when considering the ProvenCare model by Geisinger Health System, success can be obtained when identifying some low standards as a starting point.⁸ 

Variations in care are a problem in HOPDs. For example, poor arterial supply is a common reason for failure to heal, and it is now well accepted that ischemia can be at the level of the angiosome. Thus, the presence of palpable pulses is not a sufficient vascular evaluation in a patient with multiple risk factors for PAD who also lives with a nonhealing lower extremity wound. Furthermore, Medicare does not accept a pulse examination as sufficient evidence of perfusion in the workup for HBOT or the application of a CTP. Many methods of arterial screening are acceptable, and one should be performed prior to placing a patient living with a VLU in compression. Any facility being reimbursed by Medicare as an HOPD should be able to provide some method of arterial screening onsite (even if only an ankle-brachial index) and should attempt to do this at the initial visit. Similarly, any HOPD should be able to provide evidence-based compression for a VLU when there are many options available. Even though arterial screening and VLU compression are the mainstay of services for venous ulcers, there remains a high degree of variability in their delivery, with compression often delayed for days, or even weeks, pending arterial studies scheduled elsewhere. This sometimes results in patients being lost to follow up or, worse, delaying diagnosis of limb-threatening arterial disease. 

All QCDRs must have at least one outcome quality measure, which must be risk-stratified. The USWR incorporated the WHI into the quality measures for DFU and VLU healing rate. For 2017, as part of their MIPS quality reporting, a relatively small group of courageous wound care practitioners across the country will brag about DFU healing rates of 50%. Why? because a 50% healing rate will put these clinicians in the ninth decile on that outcome measure, and that’s because the average healing rate for DFUs is about 42% when all DFUs are grouped. MIPS quality measure performance rates are translated to decile rankings based on the performance of all U.S. practitioners, and the decile ranking is used to allocate points for the quality measure. Therefore, physicians whose healing rates are better than the mean achieve a higher decile ranking. Even though CMS has endorsed this measure and created a national benchmark rate that acknowledges that the majority of DFUs do not heal, reporting honest healing rates still requires courage because the vast majority of wound centers continue to report healing rates > 90%. It will take time for the field to adjust to honest reporting. However, the high decile performance of these practitioners will translate to high rankings via Physician Compare, which now has an agreement with the social media platform Yelp to translate quality performance to their star rating system. This means that reporting honest rates to CMS (even rates of 50%) will translate to high ratings on Yelp.

MANAGING WHAT YOU MEASURE

The point of measurement is comparison. Part of the role of a QCDR is to allow clinicians in similar practices (or who care for similar patients) to compare data. There is a compelling economic reason for HOPDs to use the same quality measures as practitioners. HBOT utilization is down by 50% nationally as a result of various audit programs, the newest of which is Targeted Probe and Educate (TPE), which is affecting nearly the entire U.S.⁹ The USWR has a quality measure that contains most of the elements needed to pass a Medicare audit for the use of HBOT in DFUs. A recent article discusses how registry participation could help practitioners prepare for and survive these audits.¹⁰ This is a difficult quality measure to pass. However, TPE failure rates are purportedly at 80-100%, so clinics and practitioners need a way to improve their documentation. Passing this measure requires that, prior to initiating HBOT for a DFU, the patient must have met the following criteria: 1) not achieved 30% closure after four weeks of treatment, 2) wound bed preparation with debridement of necrotic material, 3) arterial screening performed and documented, 4) adequate offloading of the DFU at each visit for four weeks of treatment documented, and 5) nutritional assessment performed and documented. The QCDR process is flexible enough that new measures could be created for each HBOT indication, and measures can be updated annually.

MEASURES THAT MATTER

A conservative estimate of annual Medicare spending on DFUs alone is $6.9 billion.¹¹ Preliminary data from the USWR suggest that point-of-care decision support and DFU quality measure reporting, even in the absence of a fiduciary incentive for practitioners (in fact, even when practitioners must pay to participate) can improve DFU healing rates by 10% when healing rates are honest. Given the annual cost of DFUs, the financial impact of even a modest improvement in healing is enormous. Unfortunately, neither manufacturers nor wound care practitioners have the same level of commitment to quality reporting that exists in other areas of medicine. Nevertheless, a growing number of practitioners and an increasing number of HOPDs have committed to quality reporting through the USWR. How will you measure your success? Strangely enough, reporting our failures may be the best way to ensure
future success. 

Caroline E. Fife is chief medical officer at Intellicure Inc.; executive director of the U.S. Wound Registry; medical director of St. Luke’s Wound Clinic, The Woodlands, TX; and co-chair of the Alliance of Wound Care Stakeholders. 

References

1. Fife CE, Eckert KA, Carter MJ. Publicly reported wound healing rates: the fantasy and the reality. Adv Wound Care. 2017;6(9):1-18.

2. Serena TE, Fife CE, Eckert KA, Yaakov RA, Carter MJ. A new approach to clinical research: integrating clinical care, quality reporting, and research using a wound care network-based learning healthcare system. Wound Repair Regen. 2017;25(3):354-65.

3. Horn SD, Fife CE, Smout RJ, Barrett RS, Thomson B. Development of a wound healing index for patients with chronic wounds. Wound Rep Regen. 2013;21(6):823-32.

4. Fife CE. Thoughts from the eye of the healthcare storm part 1: the HOPD’s role in reporting quality measures. TWC. 2017;11(3):10-14. 

5. Darrah J. Measuring the value of wound care certification in a quality-based healthcare system. TWC. 2016;10(10):27-9.

6. Fife CE, Carter MJ, Walker D. Why is it so hard to do the right thing in wound care? Wound Repair Regen. 2010;18(2):154-8.

7. Carey D. Navigating the path toward value-based wound care. TWC. 2015;9(5)26-9.

8. Fife CE. Making the “right thing” the “easy thing” in wound care. TWC. 2018;12(4):3.

9. Gelly HB. The new near future of hyperbaric medicine regulation: targeted probe and educate. TWC. 2018;12(3):18-21.

10. Fife CE, Eckert KA. The hyperbaric oxygen therapy registry: driving quality and demonstrating compliance. UHM. 2018;45(1):1-8. https://carolinefifemd.files.wordpress.com/2018/02/uhm-45-1-jan-feb-2018-fife-eckert-hbotr.pdf

11. Nussbaum SR, Carter MJ, Fife CE, et al. An economic evaluation of the impact, cost, and medicare policy implications of chronic nonhealing wounds. Value Health. 2018;21(1):27-32.