Kyruus Blog

Can’t Get No (Referral) Satisfaction

In the event of a medical illness, we all want to feel that we will get to the “right” doctor – someone expert in diagnosing and treating our specific condition using the best and latest in medical knowledge. Not surprisingly, there is a large and growing body of evidence suggesting that the “right” doctor for a given patient is the one with the appropriate training and clinical experience in that patient’s condition. This is particularly true in specialist care where procedures are more common and conditions more narrowly defined.

In one study, patients at risk for stroke and in need of an advanced procedure called a carotid endarterectomy (CEA) were significantly more likely to survive the procedure if their surgeon had performed 4 or more CEAs a year compared to those who performed 3 or less. [1] Another well-publicized example suggests that patients with ovarian cancer experience significantly better 5-year survival rates when treated by physicians who perform more than 10 ovarian cancer procedures per year. [2]

Unfortunately, patients, and even their physicians, rarely have information about the clinical experience and competency of other providers. In the absence of transparent and searchable data on physician expertise, we rely on the healthcare system – and the various players therein – to route us according to our specific needs. However, a new survey of specialists suggests that we may not be doing that well in getting patients to the “right” provider after all.

Kyruus surveyed 100 physicians across 11 specialties including cardiology, orthopedic surgery and neurology on the referrals they receive; how they receive them; how “appropriate” those referrals are; and why they believe inappropriate or “misdirected” referrals make it to their offices.

On average, respondents claimed that nearly 60% of their new patients come from referrals each year. The vast majority of these referrals come directly from the referring physician offices (66%), while 12% come from a referral call centers, as many large medical centers begin to centralize their referral and scheduling efforts. 20% of cases are patient self-referrals.

Slight variations exist, depending on the primary setting of the physician (fig. 1). Physicians at academic medical centers, for example, saw a slightly higher proportion of referrals come from centralized call centers – a finding consistent with what we have experienced and measured with our clients in that care setting.

Fig. 1

referral methods2

In our survey, we asked physicians to grade the clinical “appropriateness” of the referrals they received in the last year as “completely appropriate”, “somewhat appropriate” or “completely inappropriate”, as defined below:

  • Completely appropriate – Based on his or her condition, the patient would be best diagnosed and / or treated by someone with my clinical training and expertise.
  • Somewhat appropriate – While someone with my clinical training and expertise would be capable of seeing the patient, another physician with either more general or more specialized training than me would be more appropriate.
  • Inappropriate – The patient either (A) did not require a referral at all or (B) required referral to another type of specialist to be treated.

Physicians reported that 26.2% of all referrals they received were either “somewhat appropriate” or “completely inappropriate”. In other words, over a quarter of all patients referred potentially went to the wrong type of specialist. The reason for the perceived inappropriateness ranged from the respondent feeling over-qualified, under-qualified or qualified in a non-related area with respect to the patients’ conditions:

  • “[The patient] could have been managed by a PCP.” – Cardiologist
  • “They did not have to be seen by someone with my expertise.” – Cardiologist
  • “Should have been seen by physicians with further sub-specializations.” – Neurologist
  • “Often, patients are referred with non-operative problems who have not had any attempt at non-operative primary care treatment attempts.” – Orthopedic Surgeon
  • “[The patient] might have a urology condition relating to the prostate.” – Uro-Oncologist
  • “A few should have been seen by psychiatry first.” – [Neurologist]
  • “They did not have IBD.” – [Gastroenterologist specializing in IBD]

Surveyed specialists felt that the primary reason for misdirection was a lack of information access at the referral source – whether that be the referring physician, the referring physician’s office, or a centralized hospital call center. A percentage of respondents also felt that inappropriate referrals are the result of “pre-existing personal relationships between physicians.” (fig. 2), meaning that social connections may be leading to potentially risky biases in referral behavior.

Fig. 2

sources of misdirection2

Referral misdirection has significant consequences for both patients and for the system as a whole. 63% of respondents reported that when a patient is incorrectly referred to them, they still see the patient and then re-refer them to another more suitable physician. These double bookings result in non-actionable patient encounters for the system, unnecessary co-pays for the patient, and delays in diagnosis and care. In an effort to prevent this, physicians can often develop overly restrictive schedule templates – seeking to lock out inappropriate patients but simultaneously resulting in major blockers to efficient patient access.

To solve this problem, all stakeholders in the patient referral process – administrators, physicians, referral coordinators, and patients – need a data-driven approach to specialist search and selection. By leveraging information about the clinical experience, demographics, availability, and business rules of the entire physician network, plus precise decision support and triage tools, systems can aim to deliver:

  • Higher quality care for patients
  • Less patient dissatisfaction and outmigration
  • Higher physician productivity and satisfaction
  • More efficient patient access operations

To download the full Kyruus Physician Referral Survey, click the button below.

Download Whitepaper

[1] Statistical modeling of the volume-outcome effect for carotid endarterectomy for 10 years of a statewide database, available at:
Topics: Referral Management Analytics Patient Access Provider Networks