Recently, the California Healthcare Foundation (2014) defined variation in care as “the spectrum of approaches used by a defined group of practitioners to address a specific medical condition.” Variation in care may include differences among providers in utilization of diagnostic tests and procedures, medications, frequency of visits, length of hospital stay, surgery, utilization of complementary/integrative medicine and specialty referrals.
Recognition of remarkable variability in healthcare practice is not new. More than 30 years ago Wennberg at Dartmouth in a classic study demonstrated marked geographical variation in use of health care services for no apparent reason. From spine surgery to tonsillectomy, more than 10-fold variations in surgery rates were found in different regions of the U.S., even within the same state.
Variation extends to the outpatient arena as well. In a paper entitled, “Who you see is what you get”, published more than a decade ago, the authors surveyed physicians of different specialties – neurosurgery, family medicine, rheumatology, and neurology – about appropriate tests for clinical scenarios related to back symptoms. Recommendations for diagnostic testing were markedly divergent and determined primarily by the physician’s specialty. Neurosurgeons recommended more MR testing, neurologists performed more electromyography, and rheumatologists suggested more blood tests.
Clearly the drivers for the observed variability in practice patterns include one’s specialty training. Other factors may include availability of resources, e.g. ready access to MR imaging and the tendency to mirror the strategies of immediate colleagues.
Not surprisingly, variation in the use of diagnostic and therapeutic procedures translates to inconsistent--often poor--clinical outcomes and striking differences in cost of care among patients with the same clinical problem. Evidence-based or informed clinical pathways or carepaths have been suggested as an approach to reduce inappropriate and unnecessary utilization of health care services while improving patient outcomes. Most specialty societies and other professional organizations have developed, endorsed, and published clinical guidelines for management of a variety of common medical disorders. Typically these recommend diagnostic strategies and treatment approaches based on the best available published evidence or expert consensus (if published clinical studies are lacking).
Unfortunately, in isolation, published guidelines don’t seem to have much impact on clinician behavior. More than 10 guidelines for management of back pain have been published internationally in the last decade but studies demonstrate little effect on clinical practice or adoption by primary care providers. While the promise of the electronic medical record (EMR) was to facilitate improvements in patient care (meaningful use), to date there is little evidence of reduction in inappropriate variation in process of care.
The challenge for health care providers and institutions is to facilitate development of practical, evidenced-based clinical workflows which can provide decision support within the EMR without increasing the already excessive burden on the clinician for documentation and nonclinical work. Such an approach is not meant to eliminate physician autonomy. There will always be a need for “appropriate variation” from a guideline in selected patients based on the physician’s clinical judgement.