Why Clinical Guidelines Do Not Change Medical Practice Overnight

A clinical guideline can summarize evidence, but it does not function like a light switch. In a study of more than one million patients with knee osteoarthritis, publication of historical treatment guidelines was followed by measurable changes in injection use, but the changes were modest, gradual, and different across specialties.

The study is useful because it shows the distance between publishing evidence and changing everyday care.

Patients often hear that a treatment is "recommended" or "not recommended" and assume the issue is settled for every person. Clinicians may also assume that once a professional society publishes a guideline, practice patterns will change quickly.

In reality, medical decisions are influenced by the quality of the evidence, the strength of the recommendation, physician training, prior experience, patient expectations, reimbursement, competing guidelines, and whether the recommendation applies cleanly to the patient in front of the clinician.

A guideline is an organized interpretation of evidence. It is not a substitute for diagnosis or judgment.

The investigators used an administrative claims database to identify 1,065,175 patients with knee osteoarthritis from 2007 through the third quarter of 2015. They measured how often patients received corticosteroid or hyaluronic acid injections during each quarter.

The researchers then examined changes around publication of the first and second editions of the American Academy of Orthopaedic Surgeons knee osteoarthritis guidelines, published in 2008 and 2013. They also analyzed whether injection trends differed among orthopedic surgeons, primary care physicians, nonoperative musculoskeletal specialists, and pain specialists.

These were historical guidelines. This study should not be used as a summary of current recommendations.

During the study period, 38.0% of the patients received at least one corticosteroid injection, and 12.9% received at least one hyaluronic acid injection.

Corticosteroid use was increasing before the first guideline. The rate of increase slowed after the first edition and later plateaued after the second edition. Hyaluronic acid use also changed: its rate of growth slowed after the first guideline, and the overall trend turned downward after the second guideline.

The changes were not the same across medical specialties. Hyaluronic acid use decreased or plateaued after the second guideline among orthopedic surgeons and pain specialists, while the trend did not significantly change among primary care physicians or nonoperative musculoskeletal providers.

The guideline publications appear to have influenced practice, but they did not produce an immediate or complete shift. That is not surprising. A recommendation based on conflicting or limited evidence usually changes behavior differently than a strong recommendation supported by consistent evidence.

The specialty differences also should not be turned into a contest over which group practiced better medicine. The claims data could identify who billed for an injection and when. It could not determine the clinician's reasoning, the patient's prior treatment, the severity of symptoms, or whether the treatment was appropriate for that individual.

The study therefore describes utilization. It does not grade the quality of each decision.

• It does not determine whether a particular injection was appropriate or effective.
• It does not measure pain relief, function, satisfaction, or complications for individual patients.
• It does not establish that one specialty followed evidence better than another.
• It cannot determine why a clinician did or did not perform an injection.
• It depends on historical insurance claims and coding accuracy.
• It discusses guideline editions from 2008 and 2013 and should not be presented as a current guideline summary.

Evidence-based care requires more than posting a guideline. The clinician must understand the evidence, determine whether it applies to the patient, explain the expected benefit and limitations, and make a recommendation that fits the diagnosis.

A guideline can help prevent habit from replacing thought. It can also identify treatments that deserve a more selective approach. But the patient still needs a clear diagnosis, and the treatment should still serve a defined purpose.

For patients and referring clinicians in Oklahoma City, the useful question is not simply, "Is this treatment in a guideline?" It is, "What is the quality of the evidence, how strong is the recommendation, and does it apply to this patient's pathology and goals?"

• Is this the current guideline or a historical version?
• How strong is the recommendation, and how good is the underlying evidence?
• Does the recommendation apply to this patient's diagnosis and disease severity?
• What outcome was measured: pain, function, complications, cost, or treatment utilization?
• What reasonable exceptions or alternatives should be discussed?

This study found that historical knee osteoarthritis guidelines influenced injection use, but practice changed slowly and unevenly. That is a reminder that publishing evidence is only the first step.

Higher-value care requires accurate diagnosis, informed clinical judgment, clear patient communication, and a willingness to change established habits when the evidence supports it.

Usually not. Guidelines organize evidence and make recommendations, but they still require clinical judgment. The strength of the recommendation, the quality of evidence, and the patient's diagnosis all matter.

No. It measured how often injections were used before and after historical guidelines. It did not measure symptom relief or determine whether an individual injection was appropriate.

The study showed different utilization trends but could not prove why. Training, awareness, patient populations, competing recommendations, and established practice patterns may contribute.

Source: Bedard NA, DeMik DE, Glass NA, et al. Impact of Clinical Practice Guidelines on Use of Intra-Articular Hyaluronic Acid and Corticosteroid Injections for Knee Osteoarthritis. J Bone Joint Surg Am. 2018;100:827-834. doi:10.2106/JBJS.17.01045.

This article is for general educational purposes and does not replace an individualized diagnosis or treatment recommendation. Medical decisions should be made with a qualified clinician who has reviewed the patient's history, examination, imaging, and goals.