Medical strategy is the work of establishing where scientific knowledge in a therapeutic area is sufficient and where it is not, and planning how that gap might be closed. Its starting point is a question rather than a message.
Defining the evidence gap
When a literature search in a therapeutic area is complete, the picture that emerges is rarely without holes. Those gaps appear in several forms:
- Population gaps — the patient groups studies exclude: older people, renal impairment, pregnancy, multimorbidity.
- Comparison gaps — the absence of any study comparing two options directly.
- Outcome gaps — only surrogate endpoints measured, with outcomes the patient actually experiences left unmeasured.
- Duration gaps — no long-term efficacy or safety data.
Where these gaps go unnamed, it becomes easy to build a narrative that says more than the data supports.
Where insight comes from
Medical insight arises from observing the distance between clinical practice and published evidence. The hesitations clinicians meet in practice, the questions guidelines leave unanswered, and the divergence between real-world data and trial data all indicate which question is worth investigating.
What matters here is the direction of travel: a process moving from practice towards evidence generates questions; a process seeking support for a conclusion fixed in advance generates selectivity.
Setting priorities
Not every evidence gap carries the same weight. Criteria that can be used to prioritise include:
- The number of patients the gap affects
- How far a clinical decision depends on that information
- The potential for harm created by the present uncertainty
- Whether the question is answerable, and whether the required design is feasible
Keeping a plan within the evidence
The credibility of a medical plan depends on its respecting the limits of the evidence. A benefit not yet demonstrated belongs in a plan as an assumption to be tested, not as an expected outcome. That distinction is the line separating a scientific plan from a promotional one.
The place of real-world data
Data collected from routine practice reaches populations and durations that randomised trials do not. Confounding, however, is stronger in such data: patients who receive a treatment and those who do not differ from the outset. Real-world data does not replace randomised evidence; it answers questions randomised evidence cannot reach.
Revising the plan
A plan has to be reviewed as new data is published. A medical plan that stays fixed while the evidence moves has stopped being evidence-based. Writing down which assumption a plan rests on also makes visible what must change when that assumption fails.
This is a general scientific information article; it is not an original research publication and does not constitute individual medical advice.