Why Medical Education Strategy Needs Instructional Design
More content won't fix your outcomes problem. Better design might.
A composite scenario drawn from conversations I have had over the past two years with pharmaceutical companies, nonprofit organizations, and early-stage healthcare startups: a team is scoping their annual medical education plan. They have identified the therapeutic areas they want to address, the audiences they want to reach, and the rough budget envelope. They have selected an external partner. They have set timelines for development, review, and launch. They have decided whether each program will be live, enduring, or hybrid. The strategic plan is detailed, well-structured, and ready to execute.
What is missing from the plan is any deliberate decision about how learning will happen.
That gap is not unusual. Most medical education strategic plans I see allocate significant attention to two layers, content and execution, and very little to a third layer, instructional design. Content decisions answer the question "what should this program teach?" Execution decisions answer "who builds it, when, and how?" Instructional design decisions answer something different: "what has to happen in the learner's brain and behavior for this program to achieve its intended outcome, and how do we engineer the program to produce that?"
Skipping that third layer is the most common reason high-effort medical education programs fail to change practice. It is also, in my experience, the most underestimated risk in strategic planning for medical education.
What instructional design actually is
Instructional design has a definition problem in medical communications. The term is often used loosely to mean slide design, eLearning aesthetics, or LMS configuration. Those are downstream activities. Instructional design proper is the systematic application of learning science to the design of educational programs, with the goal of producing measurable changes in knowledge, skill, or behavior.
A working definition: instructional design is the discipline that decides how a program teaches, based on evidence about how adults learn, and matches the design choices to the program's intended outcome.
In practice, this includes decisions about how learning objectives are written and at what cognitive level; how content is sequenced, chunked, and reinforced; which activity types (case-based, problem-based, simulation, didactic) match which objectives; how learners are assessed in ways that predict practice change, not just recall; how spaced reinforcement, retrieval practice, and feedback loops are built in; and how the program will be evaluated against intended outcomes.
These choices draw on a body of evidence accumulated over decades, including adult learning theory, Bloom's taxonomy of cognitive objectives, multimedia learning principles, and Miller's pyramid of clinical competence. None of this is new. What is new, or at least newly visible, is how often it gets skipped.
Why it gets skipped
There are three structural reasons that instructional design tends to fall out of medical education strategic planning, none of which reflect bad intent.
The first is that instructional design is largely invisible when it is done well. A program with strong instructional design feels intuitive to the learner. The content unfolds in a way that makes sense, the activities feel relevant, and the assessment fits the learning. Because the design is invisible, it does not attract attention in planning conversations the way content selection or budget allocation does. The work of instructional design tends to be noticed only when it is absent and the program fails to land.
The second is that many education teams do not have instructional designers on staff. They have medical writers, project managers, accreditation managers, and content reviewers. Each of these roles is essential, and each carries implicit assumptions about how learning happens. Those assumptions are usually inherited from prior programs, not derived from learning science. When the team assembles to scope a new program, no one in the room is specifically accountable for the instructional design layer, so it defaults to whatever the format conventionally looks like.
The third is that instructional design can appear, on its surface, to add time and cost to development without producing visible deliverables. A well-designed needs assessment, a thoughtful learning objectives review, an activity-to-objective mapping exercise: these are upstream investments. The benefit shows up in outcomes data months later, not in the next deliverable on the project plan. In environments where development cycles are compressed, upstream work is often the first thing cut.
The cost of skipping it
The cost shows up in outcomes data. Moore's Outcomes Framework, the most widely used model for evaluating CME and accredited medical education, defines seven levels of impact, from participation through community health. Most programs measure at Levels 1 through 3 and report respectable numbers. Many fewer programs measure at Level 5 (performance in practice), and among those that do, the practice-change effect is often small and inconsistent.
This finding is not new. A widely cited synthesis of systematic reviews on CME effectiveness concluded that CME does improve physician performance and patient health outcomes, but that the effect is moderate and depends substantially on the design of the educational intervention. Programs that use interactive techniques, multiple exposures, and case-based methods consistently outperform programs built on passive delivery and one-shot exposure. The variable that explains much of the difference is instructional design.
The most common pattern I see in programs that have flatlined at Level 3 is this: the content is accurate, the format is on-trend, the production quality is high, the completion rates are good, the post-activity satisfaction scores are strong. And the practice change indicators, when measured, show very little movement. The program looks successful on the metrics it was designed to optimize and fails on the metric it was meant to influence. That is the structural signature of a program where content and execution were planned carefully and instructional design was assumed rather than designed.
What changes when instructional design is in the plan
A few specific examples of what shifts when instructional design enters the strategic planning conversation upstream rather than downstream.
Needs assessment becomes a different document. The conventional needs assessment identifies a knowledge or practice gap and quantifies it. An instructional-design-informed needs assessment also identifies the cognitive nature of the gap. Is the gap a recall problem (clinicians do not know the new threshold)? A judgment problem (clinicians know the threshold but do not know how to apply it in atypical cases)? A workflow problem (clinicians know and can apply, but the EHR or visit time does not support the action)? Each of these gaps requires a different educational intervention. A program designed for a recall problem will not solve a judgment problem, regardless of how well-written the content is.
Learning objectives shift in level. The conventional objective is written at the recall or comprehension level: "At the end of this activity, learners will be able to describe the new staging criteria for X." An instructional-design-informed objective is written at the cognitive level that matches the intended outcome: "Learners will apply the new staging criteria to differentiate stage 2 from stage 3 disease in patients with overlapping presentations." The wording difference looks small. The downstream implications for activity design, assessment design, and outcomes measurement are not small.
Format follows outcome, not the other way around. The conventional approach selects the format early (a webinar, an eLearning module, in person lecture) and then asks how to analyze it. The instructional-design approach selects the format late, after deciding what cognitive level and behavioral target the program needs to reach. A program aimed at Moore's Level 5 will not be delivered as a one-hour passive webinar regardless of how convenient the format is. The format will need to include active retrieval, case-based application, and either spaced reinforcement or longitudinal touchpoints. If the format does not support those mechanisms, the program needs a different format.
Assessment design predicts the outcome metric. A program intended to change practice cannot rely on multiple-choice knowledge checks as its primary assessment. Multiple-choice items measure recall and recognition reliably. They do not measure clinical judgment under uncertainty, and they predict little about whether a learner will change practice. An instructional-design-informed assessment uses case-based items, simulated decision-making, or chart-style scenarios that approximate the cognitive demands of real practice. Miller's pyramid is the foundational framework here: knowing, knowing how, showing how, and doing are different competencies that require different assessment formats.
Where instructional design enters strategic planning
The practical question is not whether to include instructional design in a strategic plan but where to put it. My recommendation is that it enters at four points.
At the gap analysis stage, before format decisions are made, instructional design helps identify what kind of gap the program is addressing and what intervention type is appropriate.
At the outcomes design stage, ideally during initial scoping, instructional design defines what level of outcome the program is targeting and how the design will support that level.
At the activity design and sequencing stage, instructional design determines how the program teaches, including the mix of didactic, case-based, and applied activities, the spacing and reinforcement structure, and the modalities used.
At the evaluation framework stage, before development begins, instructional design defines how the program will be measured against intended outcomes and ensures the assessment is built to detect the change the program is meant to produce.
If a strategic plan includes deliberate decisions at each of these four points, the instructional design layer is in place. If the plan skips two or more of them, the program is likely to perform one or two Moore's levels below its intended target.
Five questions worth asking before finalizing your next medical education strategic plan
These are the questions I use when I am asked to review a strategic plan or pressure-test a program design. They are deliberately simple. Their value is in being asked at all.
What gap is this program designed to close, and what kind of gap is it (knowledge, judgment, or workflow)?
What Moore's outcome level is the program targeting, and is the design capable of producing that level of change?
Are the learning objectives written at the cognitive level the program needs to reach?
Does the assessment format match the cognitive demand of the practice the program is meant to change?
If this program achieves its intended outcome, how will we know, and what will we measure to confirm it?
A strategic plan that can answer these five questions is a plan with instructional design in it. A plan that cannot is a plan that has decided what to teach and how to execute, but not how learning will happen. The first plan changes practice. The second plan, in my experience, produces good completion rates and quiet disappointment.
If you are scoping a medical education strategy and want a structured review of where the instructional design layer fits, I am happy to have that conversation.
