For value-based care in oncology, keep these metrics in mind

In oncology, value-based care means transforming patient outcomes by emphasizing personalized and efficient health care. A poignant example of this approach comes in the story of a patient with metastatic pancreatic cancer. His diagnosis was grim, with a predictable and limited life expectancy. Yet armed with a bucket list and determination, the patient sought to make the most of his remaining time.

One day, though, the patient’s caregiver reported that he was confused and unresponsive. This led to a quick intervention by Barbara L. McAneny, MD, and her team at New Mexico Cancer Center, which revealed low blood pressure with no fever or severe oxygen deprivation, prompting further investigation. He was administered fluids and started on a course of antibiotics because the triage protocol includes sepsis without a fever—a common, but often overlooked, issue in older patients with metastatic pancreatic cancer.

The patient’s condition improved rapidly, and he was admitted to the hospital for a day. Instead of languishing in an emergency room or ending up in intensive care, he returned home. He lived for another year and achieved many of his bucket-list goals, thanks in part to the rapid response of Dr. McAneny and her team.

“To be tuned in, to figure it out and have the processes in place to respond really quickly … that’s value-based care,” said Dr. McAneny, an oncologist and former founder and board chair of National Cancer Care Alliance—now part of ONCare Alliance—which is a member of the AMA Health System Program that provides enterprise solutions to equip leadership, physicians and care teams with resources to help drive the future of medicine.

In an interview with the AMA, Dr. McAneny offered guidance on how value-based care should—and should not—be used in oncology.

“The way value-based care needs to work in oncology is to pay the oncologist for managing all of the problems that occur with cancer and collaborating with other physicians to manage the noncancer illnesses,” said Dr. McAneny, who is also co-chair and chief officer of advocacy and government affairs for the ONCare Alliance. “People don’t stop having diabetes when they get a cancer diagnosis, for example. But when we give treatment and we include steroids, we can mess up their diabetes.

“So, we need collaborative care, and that requires time and effort because collaboration and care coordination is really one doctor picking up the phone and talking to one another,” she added, noting that for such serious illnesses communicating through the EHR is not always the optimal way for physicians to work together.

“The coordination also has nothing to do with everyone being in the same employed system,” said Dr. McAneny, who served as AMA president in 2018–2019. “It is really making it easy for physicians to communicate with one another and to recognize that that takes time and effort.”

The problem? Our current fee-for-service “health care system is focused on profits for insurers, hospitals and others instead of increasing the value of the care for patients.”

Quality metrics play a crucial role in value-based care by providing a way to measure and assess the effectiveness, efficiency and overall quality of health care services. These metrics help ensure physicians and their care teams deliver high-quality care that leads to better patient outcomes.

One bucket for quality measures in oncology is technical, which means, “Do I know the right regimen to do?” Dr. McAneny said. “That to me is pathway compliance. And the pathways should be written by academic institutions, constantly updated and based on science not on the economics.”

“In oncology, we have National Comprehensive Cancer Network, which started as a compendium of all the evidence, which is a huge step forward,” Dr. McAneny said. “Now we have groups like the Dana-Farber Cancer Institute that is creating Dana-Farber Pathways, which we’re subscribing to and implementing in practice because this is a great way to democratize the information, but also be able to measure individual physicians in different disease states on whether or not I’m actually making the right suggestions and recommendations to the patients.

“If I’m on pathway 80% of the time or so, that’s a good quality metric,” she added.

The other quality metrics have to do with patient experience. For example, timely access to care. This is a big concern because when a patient calls with a new cancer diagnosis to be seen, it should not take months to be seen.

“That’s not good quality,” Dr. McAneny said, noting time from initial phone call referral to the visit is a key indicator of access to care.

Beyond that is looking at the efficiency of getting the care underway, which includes first visit to treatment.

This “means that I’ve done all the staging things. I’ve communicated with everybody I need to communicate with, and we’ve come up with a treatment plan, we know what we’re doing and we’re ready to start,” Dr. McAneny said. Telling a patient that they need a positron emission tomography (PET) scan next month and then they will need a biopsy the month after that is “just not acceptable in cancer.”

“Even if the treatment is going to be hospice because the patient is 95 years old and has metastatic disease, you still need to explain that to patients and families,” she said.

The other part is being genuinely responsive to patients and family members. For example, do you answer or respond to their phone call promptly?

At New Mexico Cancer Center, “if you leave a symptom call, you’ll get a call back within an hour. If you’re really sick … it’ll be a warm handoff to the nurse who’ll figure out next steps,” Dr. McAneny said. “But getting patients to the right site of service is an important quality metric.”

It is all about “making sure that patients are responded to when they need to be responded to and not when it’s convenient for me to make that response,” she said. “For primary care, quality is defined as what percentage of patients get their mammograms. But for me it’s not that because I’m seeing them after they’re diagnosed.”

Patient-reported outcomes measures are another type of measure that is becoming increasingly prevalent. This is because it can be important to know how well a physician practice or health care organization is doing in its work to help patients with different kinds of cancer live better lives.

But helping patients with cancer live better lives might look different to each patient, leading to more patient-centered care.

In Dr. McAneny’s experience, though, not all value-based care quality metrics are created equal in terms of being an accurate indicator of high quality of care. For example, what value-based care should not include is holding oncologists responsible for the cost of medications.

“Having the cost of the drugs in there completely skewed Capability Maturity Model Integration’s Oncology Care Model and made it very difficult to manage,” Dr. McAneny said. That is “because it put physicians in the absolute wrong position of: If I do the best drug for the patient, which happens to be the new immunotherapy, which is very expensive, but very nontoxic and very effective, then I am doing the right thing for the patient. But I’m harming my practice because I’m going to be held responsible for the cost of this drug.”

Dr. McAneny also feels that no chemotherapy in the last two weeks of life is hard to control.

That is “because we’ll see patients with leukemia who we’re going to give chemotherapy to and some of them won’t make it, but if we don’t give them chemotherapy, all of them won’t make it,” she said. “Therefore, we don’t know which people are in the last two weeks of their life.”

Another quality metric that Dr. McAneny is not fond of is days on hospice. While she understands that to get the maximum benefit, hospice patients “need to be in hospice more than a day or two before they die,” she said.

“But if it were me, I wouldn’t want to spend three months of my life sitting around contemplating my impending doom because I’m on hospice for some quality measure,” Dr. McAneny said. Instead, “let me refer based on the patient’s need and their family’s need and their support and not put an artificial metric on top of it.”

Dr. McAneny argued that “no value-based care model should ever put the physician into the conflict of doing the right thing for the patient versus the right thing for the practice. It should just never happen.”

In addition to continuing to improve the accuracy of quality metrics, improved capacity to pull timely data is needed, especially when it comes to pulling relevant data at the point of care.

This is where Dr. McAneny’s latest venture, ONCare Alliance, can help. ONCare Alliance helps independent practices remain independent by collaborating for research, data projects, group purchasing and several other activities,

“We’re partnering with 32 other oncology practices putting data into a common data warehouse be able to say: How are we doing with patients on this regimen?” she said. But “we’re not there yet, to be able to do it well. EMR data is not research quality, but we are working on it.”

“The regimen outcomes are different. If you have metastatic breast cancer and you’re hormone sensitive and HER2-negative, I ought to be able to keep you alive for many years,” Dr. McAneny said. “On the other hand, if you’re triple-negative and my tools are limited, if I can get you a year or two, I’m doing really well.”

“It has to be very, very granular. You can’t say: ‘We cure 50% of the patients we see.’ Well, if everybody you see has nonmelanoma skin cancers, you ought to be curing 100% of them, or maybe 95%,” she said. “So, we’re not ready for outcome measures, but I have great hope that with data science we will be able to get to what is the optimal treatment for this patient, and what is the optimal cost, and what is the expected longevity.”

Learn more with the AMA about value-based care, including ways to improve data sharing and best practices for payment methods.