Wednesday, February 1, 2012

How many referrals is too many?

Most AFP review articles about conditions that may require co-management of specialists contain a short section or Table titled "Indications for Referral." For example, the January 1st article on prevention and care of outpatient burns includes a list of criteria from the American Burn Association for considering the transfer of a patient to a burn center. This and other lists generally represent expert consensus on appropriate reasons to refer a patient in a typical primary care setting; obviously, availability and accessibility of specialists has a large influence on a family physician's practice with regard to management of "referable" conditions. Clinicians' training and expertise also affect their comfort levels in caring for patients with complex problems and, as previous studies have shown, these factors lead to variations in referral rates.

Despite variations in referral rates among individual physicians, there is a clear trend in the U.S. toward more referrals. An analysis of ambulatory care survey data from 1999 to 2009 recently published in the Archives of Internal Medicine found that the probability that an office visit resulted in a referral nearly doubled during this time period, from 4.8% to 9.3%. It isn't clear why this is happening, or what percentage of those referrals are appropriate. Medicine may be becoming more complex, or patients may be presenting with more problems that cannot be effectively dealt with in an office visit that is the same length as it was 10 years ago. What is clear is that at a time when a coalition of national primary and specialty care organizations is leading a campaign to reduce overuse of health care resources, the impact of this dramatic increase in referrals cannot be ignored. But in the absence of evidence-based standards for when to refer, how many referrals is too many? Is this even an answerable question? And if it is, what can be done about it?


  1. Insightful question by Kenny Lin regarding the issue of whether or not it is "answerable" to determine how many referrals is too many? This probably is not an "answerable" question. Too many difficult-to-assess (some potentially incriminating ...) factors may be operative - including as Kenny mentions whether ambulatory problems that previously were dealt with in the office can still be handled given time constraint realities of current practice. Additional factors include potential increasing complexity of various primary care disorders - patient-initiated requests (demands) for referral - medicolegal concern by the primary care provider if adverse outcome occurs - reduced clinician comfort in management of various issues - changing 'standard-of-care' expectations in various communities - less continuity of care than in years past - and inadvertent overdiagnosis of many clinical disorders by ordering of unwarranted tests. There may be other factors ... How to objectively study and address this issue will be essential to any hope for longterm cost-containment. What is clear - is that overuse of health care resources is increasingly occurring (obvious from dramatic sometimes multi-fold differences in prevalence of various procedures and operations in different health care communities separated only by a few hundred miles ...). What should seem equally obvious is that resolution of this problem will NEVER occur unless many aspects of the current health care paradigm begin to shift.

  2. For additional perspectives on building collaborative relationships with subspecialists, AAFP members and subscribers can also read the Feb. 1st Curbside consultation on this topic: