Modern medicine is built on averages. Clinical trials test treatments against control groups, statistical significance is calculated across populations, and the result is a recommendation aimed at a hypothetical median patient. The system works reasonably well for many conditions and produces real progress against many diseases. It also misses, sometimes badly, when an individual patient lives outside the average that the recommendation was built for.
This is not an argument against evidence-based medicine. It is a case for understanding when standard treatment is doing more for the dataset than for you.
Why averages can mislead
A clinical trial that shows a drug works has shown that, on average, the treated group did better than the control group. The average can hide enormous variation. Some patients responded dramatically. Some did not respond at all. Some responded badly. The headline result is the net of these effects, weighted by sample size, and the resulting recommendation gets applied to individual patients who may resemble none of the responders the trial measured.
For some conditions, the variation is small and the average is a fine guide. For others, it is large enough that the standard recommendation is closer to a coin flip than a directive. Depression treatment is a well-studied example. Antidepressants show modest average benefit and substantial individual variation, and the actual question for a given patient, which drug works for me, is not what the trials are designed to answer. The patient ends up cycling through medications until one fits, which is personalized medicine by trial and error.
Where one-size-fits-all breaks worst
Three categories tend to suffer most. The first is psychiatric medication, where individual neurobiology, comorbidities, and side-effect tolerance vary so widely that the population-level guidance is only a starting point. The second is chronic pain, where the underlying mechanism varies more than diagnostic categories suggest, and treatments that work for one patient sometimes worsen another. The third is metabolic conditions like type two diabetes, where standardized dietary advice has been repeatedly shown to produce wildly different responses across patients.
In all three areas, the gap between standard recommendations and individual outcomes is large enough that a thoughtful patient-clinician relationship outperforms strict adherence to guidelines. That is not anti-medicine. It is medicine catching up to its own data.
Pushing for personalization without falling for hype
Personalized medicine is also a marketing category, and not every test or service that promises individualized treatment delivers. Genetic testing for psychiatric drug selection, for example, has been heavily promoted and shows modest evidence. Consumer wellness panels often produce dramatic-sounding results with weak clinical actionability.
The reasonable middle is to ask better questions. Does this guideline reflect strong evidence in patients like me? Is response to this treatment usually consistent or highly variable? What does the data say about non-responders, and what is the next step if I am one? A clinician who engages seriously with these questions is more useful than one who quotes guidelines without context.
The bottom line
One-size-fits-all is a feature of how clinical evidence is generated, not a description of biology. Most patients do well on standard treatment. Some do not, and recognizing that you might be in the second group is the first step toward better care.
Leave a Reply