Selecting a specialist: adding evidence to the clinical practice of making referrals.

Referring patients to other physicians is one of the most fundamental and frequently performed tasks in clinical practice. In 2009, referrals to other physicians weremadeduringalmost 1 in 10ambulatoryvisits in the United States for a total of more than 100 million referrals.1 Despite the routine nature of referrals, there is significant variation in how and when physicians choose to ask for specialist involvement. Rates of referral appear tovaryupto5-fold,withbothoverreferral and underreferralbeingcommon.2Decisionsaboutwhether to refer appear to be influenced by both patient factors, suchas illness severity andexpectations, aswell as physician training and expertise.2 As a consequence, standardizing and optimizing the referral process may affect the cost and quality of care.3 Evenwhenappropriate referral decisionsaremade, there is little consistency in how physicians select consultants. Some physicians have little or no control over the choice, especially in the inpatient settingwhereoptions are limited to on-call specialists who change frequently and, sometimes, unpredictably. In other environments, the culture is simply to refer to a particular clinic or department. In these circumstances, clinicians spend little time consideringwhich particular specialist their patients will see. However, in many cases, physicians have substantial influence over the referral process and could consider abroad rangeof factors in their decisionmaking (Box).4