William Check, PhD
July 2017—Inappropriate referrals to rheumatologists and months-long wait times led pathologists to start a service at Harris Health in Houston of consultative-algorithmic workups for rheumatologic disease.
“Everyone liked it. Rheumatologists were happy to get patients they could treat and who were already worked up,” Robert L. Hunter, MD, PhD, says of the service that gave primary care providers the option of selecting algorithmic testing with pathologist consultation rather than order individual tests when signs and symptoms suggested rheumatologic disease. He and his colleagues in the Department of Pathology and Laboratory Medicine at the University of Texas Health Science Center at Houston instituted the service in 2014 to help primary care providers decide when to refer patients for a rheumatology consult.
Dr. Hunter
The service was a success: In two years it reduced the number of unnecessary referrals, the rate of referrals rejected for incomplete testing, the average number of visits to phlebotomists, and the waiting time for a rheumatology consult.
Speaking in February at the Diagnostic Management Team Conference in Galveston, Tex., and in a recent interview, Dr. Hunter, who is distinguished professor and department chair, explained why they set up the service, how they got clinicians to accept it, why it was discontinued and how it has been resurrected, and whom they’re working with now.
First, how is the consultative-algorithmic workup like a diagnostic management team? Michael Laposata, MD, PhD, chair of the Department of Pathology, University of Texas Medical Branch-Galveston, and chief promoter of the concept, defines a DMT as one that meets frequently and regularly and provides patient-specific reports that are delivered before or during the time when treatment decisions are made. The report must consider the clinical context in which diagnostic tests are ordered and attempt to synthesize all test results, and it must be entered into the patient’s medical record.
The consultative-algorithmic workups are similar, Dr. Hunter tells CAP TODAY. In addition to the algorithmic testing and consults, the service the clinical pathologists provide consists of integrating lab data with clinical and imaging data and issuing comprehensive consultative reports, usually within one to three days. “Basically it is a way to put some intelligence into a lab workup,” he says.
Dr. Hunter and his group are looking to develop other similar services, ones that add value they can document. “We’re headed more toward bundled payments. The people making decisions will be the leaders of the medical system,” he says, “and if they don’t know what pathologists do, they will not pay us.”
Dr. Hunter and colleagues approached Harris Health, a safety net organization serving a mostly indigent and underserved population in Harris County. They—one of two academic groups providing pathology service to Harris Health—spoke with administrators and physicians. “At one of the first meetings,” Dr. Hunter says, “the CEO happened to be sitting there, and he said, ‘That’s a good idea. We should do it.’ So we were able to start moving ahead.”
When they talked to the chief of staff at Harris Health, she said her biggest problem was a six-month wait for rheumatology consults. One patient went into renal failure and ended up in the ICU while waiting for a consult. When patients did get to see a rheumatologist, half were found to not need to see a rheumatologist and the other half had not been worked up appropriately.
“Everyone was unhappy,” Dr. Hunter says. “The primary care physicians said, ‘They won’t see my patients.’ The rheumatologists said, ‘This patient has something wrong but I can’t treat them.’”
Dr. Hunter’s group told them, “Most of that is laboratory. Why don’t you send an order for a consultation to pathology? We have access to clinical data, and we can do the tests to determine whether a person needs a rheumatology consult.”
With clinicians, they—faculty and residents—developed algorithms that include a review of imaging and clinical factors and the selection and sequencing of tests. They generated one pathway for rheumatoid arthritis and another for systemic lupus erythematosus. “It is easier to train a few pathologists to do this than to train all providers to understand lab tests and keep current on them,” Dr. Hunter notes. Semyon Risin, MD, PhD, was “a key person doing the legwork and getting this going at the hospital,” he adds. Dr. Risin had retired from pathology; Dr. Hunter persuaded him to return.
One click initiated consultation, making it easy for clinicians to order. Clicking to order prompted a request for essential information and drawing of samples. After the algorithm was executed, a report was prepared that was sent to the patient’s chart. An email reminder was sent to the ordering provider, along with brief conclusions and recommendations.
“We sent the doctor a report saying that, according to ACR [American College of Rheumatology] criteria, this patient probably does, or does not, have rheumatoid arthritis. We also commented on the significance of other findings of our workup.” They made it clear they were not making a diagnosis. “We just let people order an agreed upon workup by clinical indication and drew attention to its likely significance.”
The process was implemented systemwide. Harris Health has 10 large clinics with 10 to 20 physicians per clinic. Results were tabulated after more than 3,800 reports had been issued. About 65 percent of all providers in the Harris Health clinics used the service regularly.
The benefits were clear (Risin SA, et al. Ann Clin Lab Sci. 2015;45:239–247). The algorithm eliminated about 90 percent of unneeded referrals, defined as patients who had a condition that needed attention but not one a rheumatologist was able to treat. “We went back and verified this by reviewing the charts and looking at what happened with patients afterward,” Dr. Hunter said.
Waiting time dropped from six months to one to three months. The rate of rejected referrals for medical necessity dropped from 40 percent to less than three percent. The average number of visits to phlebotomists for a diagnosis decreased from 2.7 to one. “Provider and patient satisfaction were excellent,” said Dr. Hunter.
He identified the key factors in making this program successful. “One, we had buy-in from the highest level,” he said. The resistance or reluctance among primary care providers to try the service was overcome by pressure from above. “The biggest piece that made them use it was that the chief of ambulatory care services started badgering people to use it. That was important for the first time. Once they used it, most liked the results and used it again. But it was hard getting them started. They were leery of us making a diagnosis that boxed them into something they didn’t want. We were careful not to do this.
“Another key factor,” he continued, “is that we were starting with what we call a DSRIP program, which is a CMS-funded program to improve health care delivery.” Delivery System Reform Incentive Payment, or DSRIP, programs are part of section 1115 Medicaid waiver programs that provide states with funding that can be used to support hospitals in changing how they care for Medicaid patients. “It was set up so that Harris Health got significant payments if predefined milestones were met.”
Looking back, Dr. Hunter believes that using referral or no referral as an endpoint was a good choice and was appreciated by primary care providers and rheumatologists. Rheumatologists were pleased about seeing more patients they could treat and fewer they couldn’t. The chief of family medicine told Dr. Hunter that prior to the start of the program, he had received only a note that his patient was not eligible for the rheumatology clinic. “With our program, he received a detailed description of the testing done with reference to the ACR-endorsed Criteria for Rheumatic Diseases. He gave the report to his patient saying, ‘Look, you have been reviewed by an expert and do not have rheumatoid arthritis.’” Like most physicians, he was enthusiastic about the program and wanted it to continue.
“I got letters from clinicians across the system saying, ‘We like it,’” Dr. Hunter said.
But the way they were billing the work, through the DSRIP program, proved their undoing, for a time. Texas in 2010 put in for the 1115 waiver and was approved, and the pathology consultation service was funded through that waiver. However, a silo problem arose: The people who pay physicians are in a different silo from the people who run these projects and there were many bureaucratic requirements. “The new person in charge of projects was an ophthalmologist who said, ‘Why would I ever want a consult from a pathologist?’ He had other priorities rather than renewal of the project.” Dr. Hunter’s group has now revived that project at a private hospital where they can be paid for clinical pathology consults, using CPT code 80502. “The service has been resurrected and is growing in UTHealth,” he tells CAP TODAY.
“I think we did a good job of proving the concept that what the primary care physician appreciated most was saving them some time. Helping them get through difficult patients more effectively was the key to that.”
Dr. Hunter and his colleagues are now working with an endocrinologist to create more than a dozen panels that would save him time. One example is amenorrhea in a young woman, which requires a complex endocrine workup. “The endocrinologist wants to be able to order the workup with one click and get a one-page report back,” he said. Other physicians who occasionally see similar patients could order the same workups. “However, they would like an explanation. The endocrinologists approve of this.” Like the rheumatologists, the endocrinologists would like more of the patients referred to them to be patients they can treat. “We are working on that now,” Dr. Hunter said.
“I think the way the world’s headed this will become more important, and we will find more effective ways to pay for it. Pathologists in general are trained primarily in anatomic pathology. Laboratory medicine has been focused on getting the right answer quickly and not on how it should be used. There is a real need for that attitude to change.”
Their experience, he says, is that physicians generally do very well with laboratory testing in their comfort zones. “Faced with a patient whose condition is outside of their comfort zone, they all need help. If an activity is perceived as more rules and regulations, they resist. If it is perceived as help, most want it.”
[hr]
William Check is a writer in Ft. Lauderdale, Fla.