Referral process needs standardization, protocols

Primary care physicians make millions of referrals to specialists each year, yet there is little protocol to follow and few tools to rely on when determining who will take their patient’s care to the next level.

In 2009, there were more than 100 million referrals made during ambulatory visits-that’s roughly 1 in 10 visits resulting in referrals. There is a lot of variation in how and when physicians seek specialist intervention-physician training and expertise, as well as the severity of the patient’s illness and their expectations for care, all factor into referral decisions.

MORE COVERAGE: What you need to know about referrals and liability

Yet when physicians make the call to send a patient to a specialist, there is no standard practice for evaluating the fit between the patient and the referred physician.

Niteesh K. Choudhry, MD, PhD, and Joshua M. Liao, MD, both of Brigham and Women’s Hospital and Harvard Medical School in Boston; and Allan S. Detsky, MD, PhD of the University of Toronto, Mount Sinai Hospital and University Health Network made the case in a recent issue of the Journal of the American Medical Association (JAMA) for standardizing the referral process.

“Physicians must often base their referral recommendations on little or no objective information,” the authors write. “Physicians have few mechanisms for personal performance feedback and little or no training in how to evaluate the quality of care that their peers provide.”

Standardization could affect both the cost and quality of care, they argue, due to the fact that there is currently little consistency across the profession, with a variation of up to five-fold. The issue is even more apparent in the inpatient setting, when ever-changing on-call specialists are used.

Even in ambulatory settings, the authors argue that patients are often referred to generic clinics or departments, with little consideration made by the referring physician as to which particular specialist would best suit the needs of the patient. Of course, authors note, there is a another end of the spectrum where physicians practice much more control over the referral process, but often availability of appointments, who works within certain networks, geographic locations, and the patients’ ability to pay are key factors in the referral process as well. In terms of patient preference, physicians may be apt to refer patients who value thoroughness to specialists who are “liberal” with diagnostic testing, or to those who have similar cultural beliefs as the patient.

 The authors suggest that some of the metrics currently reported for various industry initiatives, such as pay-for-performance or other federal programs, could also be used to help physicians select specialists for their patients.

But that system would still have drawbacks, the authors note.

“Although acquiring more granular and detailed data about physician performance maybe helpful, it alone will be insufficient for improving crucial aspects of the referral and recommendation process,” they write. “Knowing that a consultant’s patients generally achieve good glycemic control also does not indicate how easy it is for patients to have their blood drawn, how effectively results are communicated to patients, or how collegial or collaborative consultants and their staff are in co-management along with referring physicians.”

With this in mind, patient satisfaction scores may be a useful tool in making referrals. But the key to finding a better method for the referral process will likely be a combination of clinical metrics and patient feedback, the authors conclude.