Surrogates under scrutiny: fallible correlations, fatal consequencesBMJ 2011; 343 doi: https://doi.org/10.1136/bmj.d5160 (Published 15 August 2011) Cite this as: BMJ 2011;343:d5160
- Ray Moynihan, conjoint lecturer
- 1University of Newcastle, Australia
We live in a time when much disease is measured not by symptoms but by numbers, determined by biomarkers in our blood or bone.1 Transforming a healthy person’s risk of disease into a chronic condition has been a key characteristic of modern medicine, creating vast new markets for “preventive” pills designed to reduce suffering and extend life. The annual global spend on cholesterol lowering drugs alone has exceeded £10bn (€11bn; $16bn), while more generally widening definitions and lowering thresholds continue to expand the patient pool.2 Well funded campaigns urge the public to know their numbers, and professionals are rewarded for treating to target. Yet the grand assumption underpinning this approach—that helping a person’s numbers will automatically improve their health—is a delusion as dangerous as it is seductive.
Use of flecainide to reduce the number of irregular heart beats, for some people also raised their risk of an early death, killing tens of thousands just decades ago.3 4 Long term hormone replacement therapy lowered “bad” cholesterol and raised “good” cholesterol for generations of women, but it also lifted their chances of heart attacks and strokes.5 Prescribing pills to aggressively decrease blood sugar in high risk diabetes patients has been increasing their risks of disease and premature death rather than reducing them. 6 Yet decisions to approve and prescribe drugs based on success with surrogate end points continue apace, as do their sometimes deadly consequences.
Since at least the 1950s studies showing correlations like those between high blood pressure and heart disease have led us to believe that if we can modify people’s biomarkers, we can lower their risks of disease or early death. While the theory sometimes works, its logical flaw is obvious. Whether …