The Prevention Gap
The Distance Between What We Know and What We Do
Over the course of my career in clinical trials and public health, I have realized that the central challenge in prevention is not discovery but implementation. We identify risk factors, measure them precisely, test interventions in large, randomized trials, and publish carefully analyzed results. The scientific enterprise functions as it should. Yet population outcomes often fail to reflect what that body of evidence makes possible. The distance between what we know and what we consistently do is the prevention gap, the space between evidence and action.
Narrowing Gap
This gap is rarely about ignorance. In many areas, the evidence base is mature and internally consistent. Instead, it reflects hesitation, institutional process, competing incentives, and a persistent preference for certainty over probability. It is the interval between early signal and collective response, between statistical confidence and practical adoption.
I encountered this dynamic many times over the years while either serving on or reporting to a Data Monitoring Committee (DMC). A memorable example for me involved our research center serving as the Clinical Trials Center for the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), the largest randomized hypertension trial conducted in North America. Beginning in 1994, more than 42,000 participants were randomized across 623 sites in a double-blinded design comparing different antihypertensive medications (diuretics, calcium channel blockers, ACE inhibitors, and alpha-blockers) to see which would be best for initial treatment of hypertension. Safety oversight was entrusted to an independent DMC appointed by the National Heart, Lung, and Blood Institute.
By mid-1999, during routine review, the DMC noted that compared with diuretic chlorthalidone, the alpha-blocker doxazosin showed no advantage for the primary endpoint of coronary heart disease death and nonfatal myocardial infarction. More concerning, however, was a higher incidence of combined cardiovascular events, driven largely by a substantially increased rate of congestive heart failure in the doxazosin (alpha-blocker) group compared with the chlorthalidone (diuretic) group. Because the combined cardiovascular and heart failure outcomes were secondary endpoints, the DMC requested additional analyses to test the robustness of the finding. Those analyses confirmed the trend, and statistical projections indicated a very low likelihood that doxazosin would demonstrate superiority to chlorthalidone on the primary outcome by the scheduled end of follow-up.
After further review in January 2000, the advisory committee recommended termination of the doxazosin arm, and the NHLBI Director accepted that recommendation. The decision was the product of accumulated evidence, prespecified monitoring guidelines, and disciplined judgment under uncertainty.
The implications extended beyond the trial. After publication of the ALLHAT findings, alpha-blockers were no longer recommended as first-line therapy for hypertension, and diuretics were reaffirmed as preferred initial treatment. Evidence moved deliberately through regulatory and professional channels before it changed practice.
When credible evidence accumulates and oversight structures function as designed, the system can respond. The movement from signal to standard is deliberate, but it need not be paralyzed. The prevention gap does not arise from appropriate scrutiny. It emerges when caution hardens into inertia after reasonable thresholds for action have been crossed.
Outside the structured environment of clinical trials, delay often extends further. Additional confirmation is sought long after reasonable action is justified. Consensus drifts. Implementation stalls. Each step appears modest, yet the cumulative effect is delay measured in years. In prevention, this is not neutral; it translates into events that might otherwise have been avoided.
The same pattern is visible elsewhere. Measles, once eliminated in the United States, has returned in communities where vaccination rates have fallen below levels needed for herd immunity. The biology has not changed. The evidence has not changed. What has shifted is uptake and trust. Preventive systems built over decades can erode quietly until outbreaks reveal their fragility.
The structural forces behind the prevention gap are familiar. The benefits of prevention are statistical and dispersed; the costs are immediate and visible. When prevention succeeds, nothing happens, and nothing happening rarely commands attention. Systems often reward short-term responsiveness more than long-term foresight. Leaders who act early in the face of uncertainty risk criticism for overreaction, whereas those who delay action until harm is unmistakable face less accountability.
The prevention gap is not a failure of science. It is a failure of timing, of translating sufficient evidence into action. History rarely shows that we knew nothing. More often, it shows that we waited. And waiting has consequences of its own.
This recurring pattern of the interval between warning and response, and the individuals who recognized signals early and acted despite resistance, are explored more fully in my book, The Preventioneers: Diseases, Disasters, and the Discoveries That Changed Our World (Johns Hopkins University Press, May 5, 2026). Understanding this framework may help us narrow the distance between knowledge and action.