Patient Safety Tip of the Week


December 2, 2008

Playing without the ball…the art of communication in healthcare        






Sports fans, how often have you heard the accolade for an athlete in a team sport “he’s really good at playing without the ball (or puck)”? This might mean a wide receiver in football making a downfield block for one of his teammates or a basketball player setting a pick for one of his teammates to go to the basket. Or it might mean a lacrosse player or hockey player moving or slowing down or even stopping to get into a position away from a defender so he can get a pass from a teammate and be in position to shoot on goal.


Medicine, of course, is the consummate team sport and lots of the key teamwork takes place away from the patient and the bedside. Much of the critical communication between healthcare workers takes place behind the scenes and it is important to ensure effective communication systems are in place. Just as the athlete needs to get in the best position to score, the effective clinician needs to get into the best position to influence a positive, safe outcome for the patient.


Let’s start by looking at the inpatient consultation. It is striking how often consultants’ recommendations are never implemented. Though there is considerable variation by both the consulting service and the service requesting the consult, many studies show only about half of all recommendations by some consulting services ever get implemented.


Unfortunately, most of the literature on consultations dates back to the 1980’s and a lot has changed since that time. Goldman et al (Goldman 1983) wrote “the ten commandments for effective consultations”:


According to Wachter et al (Wachter 2005), citing some of the above work, consultants’ recommendations are more likely to be followed if:


But Salerno et al (Salerno 2007) almost 25 years later reviewed the literature on consultation guidelines and updated those guidelines based upon results of a survey they performed. They addressed an important issue that many hospitals still struggle with: “can consultants write orders?”. The old school of thought has been that the primary service should write all orders since that best ensures continuity and coordination of care. However, Salerno et al found somewhat differing views based upon the service requesting consultation. Only 37% of surgeons preferred to retain order writing authority, compared to 59% of primary care physicians. Also, whereas 69% of nonsurgeons wanted the consultant to have a narrow focus, only 37% of surgeons wanted the consultant to have such a narrow focus. And while most (75%) of physicians desired direct communication with the consultant, the value of educational references was not perceived as important as it had been in the 1980’s studies. In their discussion, they note how the changing practice of medicine may have changed needs. Economic pressures may have caused surgeons to spend more time in the OR at the same time when the increasing age and complexity and comorbidities of patients necessitates more time in nonsurgical aspects of care. Even things like contracted housestaff work hours may interfere with desirable direct communication. They recommend the “ten commandments” be rewritten to allow for more of a comanagement relationship between surgeons and nonsurgical consultants and the decision about order writing should be a mutual decision.


Lo et al (Lo 2004), looking at adherence to recommendations by infectious disease consultants, found an overall adherence rate of about 80%. However, they did not find any relationship between adherence and the number of recommendations, severity of illness, or presence of followup notes. Though adherence was less when notes were of poor legibility or disorganized, notes of these qualities were very rare. They did find that adherence was much better for therapeutic recommendations than for diagnostic ones. Calling the primary service increased adherence at a private hospital but not at a public hospital (but they speculate this may have been due to fewer resources available at the public hospital to comply).


(Our own comment on the observation that diagnostic recommendations are less likely to be followed: it is very clear that specialists in the US  are much more likely to order diagnostic studies than primary care physicians in general. In some cases that may be appropriate, in others not. So it should not be at all surprising that the primary service is less likely to comply with all the consultant’s recommendations for diagnostic studies.)


Weiner et al (Weiner 2006) looked at implementation of recommendations made by geriatrics consultants. They found their recommendations implemented completely in only 52% of cases. They did find higher rates of implementation when verbal rather than written communication took place and for those recommendations that facilitated things like discharge planning.


We are fans of the direct communication approach. That has a value not only in the current consultation but helps instill a trust and understanding about consultations in the future as well. Back to the sports analogy: the better you get to know your teammate, the more you will anticipate how he will be able to help achieve your goals and your patient’s goals.


Nevertheless, we have entered a different age of communication. Just look at your kids “texting” each other (or multiple others). Email allows us to communicate back and forth with our colleagues effectively rather than play frustrating telephone tag. And, given the comments in the Salerno paper about some physicians’ limited availability because of time and economic constraints, these forms of asynchronous but effective communication are here to stay.


Enter the computer age and some potential solutions to these dilemmas. Were et al (Were 2008) at the Regenstrief Institute have developed an innovative approach using CPOE to improve implementation of recommendations by consultants. They identified a low rate of compliance with recommendations (about 50%) and put together an interdisciplinary quality improvement team to come up with suggested improvements. Focus groups including both referring and consulting physicians agreed that use of CPOE could provide a good solution. So they developed a tool with which consultants could create “consultant-recommended orders (CRO’s). These are orders input by the consultant using the typical CPOE entry tools they would use to write orders on their own patients. However, they are stored separately from other orders and do not become actual orders until they are activated by the primary (referring) physician. When the primary physician logs on to the system, the CRO’s are flagged and he/she can review, edit and turn those CRO’s into actual approved orders.


Using a before/after study design, the Were team demonstrated a 30% improvement in compliance with consultant recommendations and demonstrated satisfaction by both the referring and consulting physicians with the CPOE CRO tool. You’re going to see a lot more of useful tools like this.


The Were article contradicts the prior finding that suggested a larger number of consultatnt recommendations reduces the likelihood of implementation of those recommendations since they found no such correlation.


Also, remember that the consultation is a two-way process. Just as when diagnostic imaging studies are ordered, an order for a consult should clearly convey what question is being asked. CPOE can also facilitate this. A good CPOE order entry tool for consults would allow the referring physician to click on a drop-down list of the most common reasons for consulting the specialty being consulted and a text box to enter specific information or questions being asked. The standardized items in the drop down list are useful because you can tie alerts to them. For instance, if a particular consulting service always wants a specific study done for a specific condition, an alert could convey that to the referring service so the result of that study could be available by the time the consultant sees the patient. Another alert might actually suggest that consultation for a specific reason might be better done as an outpatient.


It is amazing how hi-tech tools can overcome some of the limitations in our complex medical systems. So use them judiciously. But don’t forget the value of face-to-face communication with your colleagues even when you are using these hi-tech tools.





Goldman L, Lee T, Rudd P. Ten Commandments for Effective Consultations. Arch Intern Med. 1983; 143(9):1753-5


Wachter RM, Goldman L, Hollander H. Ch. 28 Medical Consultation. in Hospital Medicine. 2nd Edition. Philadelphia: Lippincott Williams & Wilkins 2005


Salerno SM, Hurst FP, Halvorson S, Mercado DL.  Principles of Effective Consultation: An Update for the 21st-Century Consultant. Archives of Internal Medicine 2007; 167(3):271-275


Lo E. Rezai K. Evans AT. Madariaga MG. Phillips M. Brobbey W. Schwartz DN. Wang Y. Weinstein RA. Trenholme GM. Why don't they listen? Adherence to recommendations of infectious disease consultations.[see comment]. Clinical Infectious Diseases 2004; 38(9):1212-8


Weiner M, Lamar V, Munger S, Perkins A, Fultz B, Wheeler M, Counsell S, Sennour Y, Callahan C. Variation In Implementation Of Recommendations Of Inpatient Geriatrics Consultation. J Am Geriatr Soc 2006; 54 (S4): S94-95 (abstract)




Were MC, Abernathy G, Hui SL, Kempf C, Weiner M. Using Computerized Provider Order Entry and Clinical Decision Support to Improve Referring Physicians’ Implementation of Consultants’ Medical Recommendations. First published October 24, 2008 as JAMIA PrePrint; doi:10.1197/jamia.M2932









Patient Safety Tip of the Week Archive


What’s New in the Patient Safety World Archive