A healthy 86-year-old woman presented to my office with vague post-prandial epigastric pain. Her history and physical examination were consistent with symptomatic cholelithiasis and, given her otherwise excellent health, I referred her for surgical evaluation. Following her initial visit with the surgical consultant, an ultrasound examination confirmed cholelithiasis. However, the surgical consultant thought that an abdominal bruit was also present and requested an ultrasound examination of the abdominal aorta. The ultrasound was normal, but an undifferentiated mass in the region of the right adrenal gland was detected. A computed tomographic (CT) scan revealed a homogeneous 4.5 × 5.0 cm adrenal mass. A metabolic work-up was unremarkable, and the patient underwent laparoscopic cholecystectomy without complications. I learned of the adrenal mass during the postoperative period. My patient's adrenal mass will be followed clinically. I was miffed to learn of the additional work-up, but when I contacted the attending physician to assert my displeasure he did not acknowledge a conflict.
What is the role of a consultant in the referral process? What constitutes a “good” referral? Was the primary care physician's displeasure justified in this case? The issue is framed by the primary physician as the consultant having “taken over,” while the consultant does not see it that way. I have practiced medicine in three systems on three continents and, although it is cold comfort, I know firsthand that similar problems occur in other places.
A particularly irritating scenario occurs when a consultant sends a referred patient to another subspecialist without the involvement of the primary physician. It is possible that the primary physician might be able to address the problem or may even have already done so, so this action results in unnecessary duplication of effort. This is the so-called “black box” phenomenon that occurs frequently in teaching hospitals. The patient disappears into the system only to reappear in the primary practice after having undergone consultations or procedures that are sometimes unrelated to the original consultation and frequently before the consultant's letter arrives in the mail. It is particularly irksome when the consultant, having completed a subspecialty work-up, with or without a definitive diagnosis, arranges a secondary referral without the involvement of the primary physician. When I have brought such occurrences to the attention of my subspecialist colleagues, they have been duly chagrined and, hopefully, educated by the exchange.
In this case scenario, the surgeon's behavior is ambiguous because it is not clear if the second ultrasound examination to investigate the abdominal bruit should be considered a prudent part of the preoperative work-up for this patient. If one assumes that the surgeon was unsure whether the abdominal bruit might represent some other problem (e.g., an arterial stenosis or aneurysm) that could either explain the patient's postprandial abdominal pain or adversely affect the surgical procedure, it is reasonable to give him the benefit of the doubt for having pursued further testing. However, the discovery of an undifferentiated adrenal mass illustrates the risks involved in performing diagnostic tests that may not be clearly indicated. All of the surgeon's findings resulted in additional testing which has now become necessary to ensure the patient's safety during surgery and to avoid any medicolegal risks in the event of an adverse outcome. Although it can be argued whether the abdominal bruit should have been worked up in the first place,1 the surgeon's diligence cannot be faulted in working up the abnormal findings that were then uncovered. After all, he was conducting an appropriate evaluation of the patient consistent with the role of an expert consultant. Regardless of the particulars of this case, there is no doubt that a larger problem exists with the referral process as it currently exists.
It may help to consider why primary physicians refer patients2:
To establish a diagnosis.
For a specified investigation when the diagnosis is reasonably clear.
For treatment or surgery when the diagnosis is known.
For advice on management when the diagnosis is known.
For a consultant to take over management when the diagnosis is known (e.g., dialysis).
For a second opinion to reassure the primary physician that all that was required has been done.
For a second opinion to reassure patients or their families that all that was required has been done.
For other reasons, including an organizational requirement for a second opinion or for medicolegal concerns.3 The primary physician may have more than one reason for making a referral.
The consultant is expected to provide an expert answer to the primary physician. To do that, the consultant may need to obtain additional diagnostic tests or even a secondary consultation. It is unreasonable to expect that the primary physician will be consulted at every step of the way, as the physician who posed the question here seems to suggest. If, however, a secondary referral is considered to address an unrelated problem found in the course of a work-up, it would be preferable to involve the referring physician in that decision.
Repeated interactions over time are necessary for a referring physician and a consultant to develop a satisfactory level of communication. Certainly, this is easier to accomplish in a smaller, more personal community hospital, where the primary physician and the sub-specialty consultant serve on the same staff and rub shoulders in the corridors and at meetings. A large teaching hospital is perforce more impersonal, and poor communication is more likely to result, even among those of us who work as generalist primary physicians in academic settings. The primary physician and the subspecialist need to be cognizant of this potential pitfall and take steps to ensure that they are aware of each other's needs and expectations. At the very least, the more precisely a referral letter states the reason for referral, the more likely it is that a consultant will provide the appropriate feedback.
I am sure that there is room for improvement on the part of the primary physician and the subspecialist consultant that could increase physician and patient satisfaction with the referral process.4 Primary care physicians provide continuous, comprehensive and coordinated care to a population of undifferentiated patients. A collegial interaction between primary and subspecialty physicians is essential to ensure quality patient care. To date, research has attempted to identify the physician and organization variables that affect referral rates.5 Because continuity of care appears to have value in terms of outcomes6 and cost control,7 more research should be undertaken to identify interventions that improve the referral process.