I admit that I have a bias when it comes to the evaluation of quality of care and use of guidelines. I believe that measuring quality of care is close to impossible. Quality, as Justice Potter Stewart might have suggested, is almost always in the eye of the beholder. Nonetheless, we should try to come up with some agreed upon markers of quality. Otherwise, we will be without data and then, as Demming says, “just another person with an opinion”. I also believe that heterogeneity in the delivery of care for a single condition is often inappropriate. In general, use of guidelines will allow average practitioner help more patients more effectively.
Evidence Based Guidelines are, however, analogous to a magnetic compass which points to magnetic north, not true north. As long as the journey is going well and you don’t need help near the end of a journey, following the guidelines will most often help with decision-making. As you get nearer the pole and the journey isn’t going as easily as you had predicted, the compass, or the guidelines, may no longer be helpful. At the point when the patient hasn’t responded as had been predicted by the guidelines, specialist care is often invoked. At this point, a specialist/consultant may recommend variations from guidelines, driven now by clinical experience/judgment: the best assessment of evidence, other than that which drove the guidelines; consideration of the patient’s individual values and goals; as well as ethical and family considerations are appropriate.
Sometimes variations from guidelines suggest that utilizing a more expensive therapy may be appropriate. Sometimes using an “older” therapy that has generally been supplanted by “newer” therapies may be better for an individual patient. Not all patients will respond to “Guideline Driven Therapies” the way that had been predicted or hoped for. In these cases an expert is often asked to help with the therapy.
In patients covered by some Pharmacy Benefit Management (PBM) programs, there is often a “peer review” process that looks at a diagnosis and a therapy to see if they “match”. If they don’t, then the prescribing physician is often asked to justify the course. When the prescribing physician is an experienced, well-trained practitioner, who generally hasn’t misused the system, some of this oversight is onerous. Two examples come to mind:
One friend of mine, an experienced rheumatologist had been treating a middle-aged man employed as a general contractor with severe DJD of the knees with intraarticular hyaluronic acid injections on a six to nine month basis for over 5 years. With this therapy, the patient didn’t require more than occasional NSAIDs for comfort, thus avoiding opiates and knee replacement surgery, and was able to work without missing a day. While the therapy was included in the original Guidelines issued by the patient’s insurance company, the most recent edition rescinded approval for the use of hyaluronic acid due to equivocal evidence of efficacy based on a single, new meta-analysis. Because the therapy was not recognized in the revised guidelines, the rheumatologist was challenged by a physician reviewer (who happened to be a pediatric intensivist) working for the PBM. When the rheumatologist asked to discuss the issue with a rheumatologist or at least an adult internist, the pediatrician balked. Eventually, the therapy WAS approved, after about 90 minutes of discussions over multiple phone calls.
Another friend also discussed potential difficulties in getting an effective therapy approved in a complex patient. In this situation, a young woman with long QT syndrome who had been treated with diphenylhydantoin, a well recognized antiarrhythmic drug from the 1970’s to1980’s. The drug had controlled the patient’s symptoms in the past, and when it was stopped, the symptoms returned and were found to be related to the Long QT associated ventricular fibrillation. Other drugs may not have been effective, and when Dilantin was restarted the symptoms disappeared.
In both these instances, an experienced physician deviated from current Guidelines, because they suggested therapies that were thought to be outdated or were otherwise not applicable, even though in the individual patient they did work,. If PBMs could keep a database of physicians and their level of training, in addition to how often they deviated from usual therapies without good reason, they could avoid potentially unnecessary conflicts with physicians. This would save time and could save overall health care costs. In addition, time spent pleading with non-peer physicians could be better spent caring for more patients. This would likely improve physician satisfaction with practice. In addition, patients would likely be happier, being able to be seen more rapidly and not have to discuss insurance “interference” with therapy recommendations. Primary Care physicians would be happier because the specialist could see his/her patients sooner and likely be given better service as well.
[ii] Shrank, W.H; Porter, M.E.; Jain, S.H.; Choudhry, N.K.: A Blueprint for Pharmacy Benefit Managers to Increase Value: Am J Managed Care: 2009, 15, 87-93