About 15 years ago, rheumatologists at the University of Colorado School of Medicine, Aurora, implemented a preappointment consult triage system as a way to identify rheumatology patients who require timely evaluation and treatment.
Over time, the endeavor caused some soul searching. They wondered how effective their consult triage system was in identifying all patients with inflammatory rheumatic diseases and in ensuring they were seen promptly. They also wondered about the revenue implications of routine outpatient care of autoimmune and inflammatory rheumatic disease (AIRD) patients, compared with that of non-AIRD patients.
“Hospital leadership is very interested in making sure that all patients have access to specialty care in a timely manner,” said
Year-long follow-up finds high sensitivity, more revenue
What started out as a quality improvement project morphed intothat was published online July 12, 2018, in Arthritis Care & Research. Using data recorded in the electronic medical record, Dr. West and his colleagues retrospectively reviewed 961 new outpatient rheumatology consults sent during a 9-month period for final diagnosis and revenue generation for routine outpatient care over 1 year following consult review or initial evaluation. The first step of the consult management involves an intake access coordinator, Ryan Goecker, obtaining information about the patient from the referring clinician. Next, one of three experienced rheumatologists at the university – , , or – reviews the information about the case. “We do request that labs and x-rays be done ahead of time, depending on what the consult question is, so that we have the data to be able to decide whether a patient likely has an inflammatory process that we need to see or not,” Dr. West explained. “It takes somewhere between 5 and 20 minutes per consult. In total, it takes our rheumatologists about 5 hours a week to do the consults. However, even after subtracting the physician time spent screening consults, the time saved by the consult triage system enabled over 200 more time slots per year to be available to see new AIRD patients than would have been possible without the triage system.”
Following review of the data supplied about the case, patients who may have acute inflammatory monoarticular arthritis or some other rheumatologic emergency are seen within 24-48 hours, while consults with a possible AIRD are approved and seen within 1-4 weeks. Priority is given to those with suspected vasculitis, systemic lupus erythematosus, myositis, and disabling inflammatory arthritis. Patients with probable noninflammatory conditions such as osteoarthritis, fibromyalgia, and mechanical low back pain are not scheduled for an in-person visit. “Instead of simply declining the consult, we try to give some direction, or we say, ‘Why don’t you try this, this, and this, sort of a miniconsult,” Dr. West noted. “One of the potential adverse reactions to doing consult triage would be that the referring provider would be upset if we decline their consult. We acknowledge that. But over time we’ve been able to convince them that this is best for all concerned, so that the inflammatory disease group gets in to see us in an appropriate amount of time.”
Of the 961 consults, 673 were scheduled for an in-person AIRD evaluation and 288 consults were not scheduled. Patients were seen an average of 13 days after consult review. Of the 673 approved consults, 597 (89%) came for evaluation. Of these, 357 were diagnosed as having an AIRD, while 240 were diagnosed as having a non-AIRD. Among the patients not scheduled for a rheumatology visit, 128 had 1-year follow-up data, with 6 patients eventually diagnosed as having an AIRD. This translated into a consult triage sensitivity of 98% and a positive predictive value of 60%. “There’s a fair number of people with noninflammatory disease who do get into see us,” Dr. West said. “That has to do with our desire to not be overly rigid. The sensitivity, specificity, and positive predictive value were not quite as high as some people would like to see, but we really weren’t missing people who really needed to be seen.”
In their conservative cost analysis, revenue data for outpatient care was available for 318 of the 357 AIRD patients and 192 of 240 non-AIRD patients. It demonstrated that care of AIRD patients generates 44 times more revenue, compared with non-AIRD patients ($5,877 vs. $134 per patient, respectively; P less than .001).
“Our consult triage protocol appears to be an effective method to assure that patients with inflammatory rheumatic diseases get expedited access to appropriate rheumatologic care,” Dr. West concluded. “Using conservative measures, caring for patients with inflammatory rheumatic diseases results in significantly more revenue generation.” Generalizability of such a protocol to other practice settings depends on the time rheumatologists are willing to commit to it. “That’s the big thing,” he said. “The question comes up, could a nurse or a nurse practitioner learn the skills over time to be able to do efficient consult triage, to free up the doc from that? Yes. I think that’s absolutely possible.”
Going forward, Dr. Pearson and Dr. Kolfenbach, who are authors on the paper, have launched a pilot project in which the rheumatologist is compensated for doing a record review and e-consult on patients not recommended to be seen in the rheumatology clinic. “For consults who do not have inflammatory disease, we can say [to the referring clinician], ‘If we did see them, this is what we would be doing,’ ” Dr. West said. “Sometimes if there are still questions, we’ll say something like, ‘Once you get the results back, let us know and we can look at it again.’ That way they are getting similar compensation as telemedicine and other forms of consultation.”
At the Zuckerberg San Francisco General Hospital (ZSFG), use of a novel electronic consultation and referral system that was implemented in 2007 by University of California, San Francisco, providers continues to thrive for rheumatology and other specialty services. With this electronic system, previously called e-referral, all provider communication is captured in real time and recorded within the electronic health record. In a 2015 article from Arthritis Care & Research, researchers led by, reviewed 2,105 e-referrals made between 2008 and 2012 ( ). The main outcome of interest was use of preconsultation exchange, defined as back-and-forth communication between referring and specialty care providers, facilitating triage of referrals, requests for more information, or resolution of questions without a visit. The researchers found that about 25% of e-consults were resolved without a clinic visit, and that the proportion of e-consults undergoing preconsultation exchange increased over time, from 55% in 2008 to 74% in 2011.
“We’ve had situations in which somebody within our hospital system has requested a rheumatology referral but did not realize the urgency of the patients’ clinical signs and symptoms,” Dr. Yazdany, a rheumatologist and health services researcher at UCSF, said in an interview. “E-referrals are often responded to within that day. When we see an e-consult for a young woman with a rash and protein in her urine, a rheumatologist will recognize that is very concerning for a serious diagnosis like lupus of the kidney or systemic vasculitis. In that situation, we would pick up the phone and sometimes even call the patient and have them come to the emergency room for an urgent evaluation. Over the years, there have been many situations in which we’ve been able to intervene much earlier than we otherwise would have. In some cases that early intervention was lifesaving, or at least organ-preserving for the patient.”
Two studies from the early 2000s examined the use of preappointment management in rheumatology (see