Jordana Bieze Foster, Editor
Lower extremity researchers are realizing what practitioners already know: Sometimes, average just isn’t good enough.
Anyone who’s familiar with the college admissions process can tell you about the limitations of averages. Sure, averages – in this case, grade point averages – can be useful as a screening tool, and to get a general sense of a student’s performance.
But a grade point average doesn’t tell you anything about whether a specific student is stronger in math than history or prefers memorization to critical analysis—things that might be important to consider when choosing a college and a major.
In the medical literature, an intervention is typically considered to be effective if the average outcome for an experimental group is positive, preferably when compared to the average outcome for a control group or a usual-care group. But as practitioners know, often a given intervention can be very effective in some patients but not effective in others. In situations like this, the average outcome in a group is not likely to be positive, and developers of guidelines will say there is not sufficient evidence to recommend the intervention. Which is really unfortunate for the minority group of patients in whom the intervention actually was effective.
But researchers are beginning to become sensitive to these types of subject-specific variations. An excellent example is described in this month’s article on McConnell taping for anterior knee pain (see “Patellofemoral taping: Pain relief mechanisms.”) National Institutes of Health researchers, trying to identify kinematic effects of taping, found that taping had no significant effect on the study group overall.
But further analysis revealed that taping actually did have significant effects on kinematics, except that those effects were different depending on whether the patella was malaligned in a medial or lateral direction to begin with. Taping had a medializing effect in patients who demonstrated a lateral shift at baseline and a lateralizing effect in those with a medial shift. When averaged across the group, the two effects essentially cancel each other out, giving the impression of no effect at all.
This type of subject-specific or subgroup-specific thinking is becoming more prevalent in other areas of lower extremity research as well. Designers of training programs to prevent anterior cruciate ligament injuries are considering whether protocols will be more effective if tailored to specific sports or specific age groups (see “ACL experts aim to take training to the next level.”) Cerebral palsy researchers reported late last year that strength training improved crouch gait in patients with spastic diplegia who were independent ambulators but not those who used assistive devices (see “In the Moment: Neuromuscular.”) The list goes on.
There’s no doubt that evidence-based recommendations help improve clinical practice. But such recommendations may suffer from an over-reliance on average findings, and consequently may overlook significant benefits that an intervention may have for specific patients.
That’s one more reason why optimal patient management should include elements of both evidence-based medicine and clinical expertise. Because your patients deserve care that’s better than average.