Medicine, Mass Interventions, and Outcomes
There are two interesting medical studies out today, both tied to large populations of patients being given interventions best for smaller groups.
The first is a new meta-analysis of total knee and total hip replacement (TKR/THR), showing that 58% of such implants last 25 years. That may or may not sound like a large number, but it is not markedly better than the original Charnley replacements. This is despite improvements in materials, techniques, and training since then.
There are many factors here, not least of which is a growing number of non-ideal candidates as TKR/THR is more widely done. Patients are younger, patients are often heavier, and patients are often more active, all of which lead to less ideal outcomes, whether it is through excessive wear, implant loosening, or something else. Revision is a costly and serious exercise, and should be avoided a priori, which is best done via careful implant choice, good technique, and proper patient selection.
The second study looks at the proportion of patients presenting with lower back pain (LBP) who receive imaging studies. This practice has been cautioned against repeatedly — most sudden onset lower back pain doesn’t show up on imaging, & what does appear tends to be incidental, which can lead to unrelated procedures and worse outcomes.
The study shows that a striking 25% of patients who present to primary care with LBP received imaging, as 33% who presented to an emergency department. The rate has increased over 21 years, despite the many cautions against it. If imaging is available, and a diagnosis isn’t certain, physicians will use it, especially given patients’ demands for imaging, and despite the risks associated.
Both studies point to why improving medical outcomes is not simply a question of improved access to doctors, tests, and procedures. Large numbers, mass screening, and non-ideal candidates change the risk-return tradeoff in ways that don’t always favor population health.