uring the past decade, regulators and researchers including the FDA, EMA, NIH, and ICH have issued guidance encouraging gender‑inclusive research. Yet in practice, development decisions continue to default to male‑centric designs and post hoc subgroup analyses. The guidance exists. So why does behavior not change?

As the first two articles in this series explored, the issue is not a lack of awareness or policy. It is that behavior is shaped by how regulators’ expectations are interpreted, prioritized, and reinforced within real-life organizational settings.

This article reframes the female data deficit as a behavioral problem. If the healthcare industry wants consistent inclusion to move from lofty aspiration to routine practice, we need to understand the specific behavioral barriers at play and design interventions that work with—rather than against—how teams actually make decisions under pressure.

Critical Reframing: the system is behaving as designed

Clinical development teams are not irrational or indifferent to inclusion. They are operating in an environment shaped by time pressure, regulatory uncertainty, cost constraints, and accountability for delivery. Within that context, behavior is directed towards optimizing predictability and the probability of approval. Historical precedents, accepted templates, and “what worked last time” become powerful anchors.

In behavioral terms, inclusion is rarely the default option. And where inclusion is framed as additional work, additional risk, or additional cost, good intentions alone are rarely enough to change behavior. Sustained change requires altering the decision environment itself.

Key behavioral barriers and why they persist

1. Defaults and status quo bias: repeating what feels safe

The barrier
Protocol templates, statistical plans, and development playbooks often treat gender‑specific design and analysis as optional. Under time and resource pressure, teams default to familiar approaches. Status quo bias and reliance on precedent make deviation feel risky, even when guidance supports change.

What this looks like in practice

• Male‑dominant early‑phase cohorts
• Sex‑specific endpoints added late or labeled “exploratory”
• Subgroup analyses underpowered and dropped first when timelines tighten.

Actionable interventions

• Make gender‑specific design a default, not an add‑on, in protocol templates
• Require justification to opt out, reviewed alongside other core design decisions
• Use structured checklists at protocol sign‑off to prompt explicit consideration of sex differences.

The EAST framework helps policymakers to design more effective interventions. By mapping our interventions against the four EAST components (Easy, Attractive, Social, and Timely), we can align our suggestions with the ways that people naturally make decisions.

EAST lens: Make it Easy and Attractive to do the right thing by redesigning defaults so inclusion requires less effort than omission.

2. Risk perception and loss aversion: overestimating the cost of inclusion

The barrier
Risks associated with including women—particularly in early-phase studies or pregnancy‑related contexts—feel immediate, concrete, and attributable. By contrast, the risks of exclusion (incomplete evidence, post‑marketing safety issues, labeling changes) are delayed, diffuse, and often owned by someone else later in the lifecycle (see below).

What this looks like in practice

• Conservative exclusion criteria framed as “risk management”
• Pregnancy and lactation treated as blanket exclusions
• Decisions driven by fear of delay rather than quality of evidence.

Actionable interventions

• Reframe risk discussions to compare inclusion risks with the known costs of evidence gaps
• Use regulatory pre‑mortems (e.g., “If this product required a post‑approval label change, what would we wish we had done differently?”)
• Introduce staged inclusion pathways with clear governance rather than binary include/exclude decisions.

EAST lens: Make the long‑term risks of exclusion less acceptable from a Social standpoint and reduce ambiguity by providing clear, managed pathways for inclusion.

3. Incentives and feedback: optimizing for speed over evidence quality

The barrier
Clinical teams are commonly rewarded for staying on time, on budget, and aligned with precedent. High‑quality gender‑specific evidence, by contrast, is rarely visible in performance metrics or rewarded explicitly. Where regulatory feedback on inclusion is delayed or inconsistent, learning is slow.

What this looks like in practice

• Short‑term delivery metrics outweigh longer‑term evidence considerations
• Inclusion framed as a threat to timelines rather than a marker of excellence
• Repeated reliance on acceptable minimums.

Actionable interventions

• Embed evidence‑quality indicators (including gender‑specific data) into internal development metrics
• Provide consistent regulatory and ethics committee feedback when inclusion expectations are not met
• Highlight and reward “positive deviants”: programs that planned inclusively and avoided downstream corrections.

EAST lens: Make inclusion more Attractive by aligning incentives and feedback with evidence quality, not just speed. Subvert the expectation that guidelines mean delays by emphasizing the Timely downstream benefit of inclusive data.

4. Organizational dynamics and diffused accountability: when no one owns the problem

The barrier
Responsibility for inclusion is spread across sponsors, CROs, ethics committees, investigators, and regulators. When ownership is unclear, action slows. Each actor may support inclusion in principle while assuming that others will drive it in practice.

What this looks like in practice

• Inclusion discussed but not operationalized
• Assumptions that regulators or ethics committees will “catch” issues later
• Incremental drift back to familiar designs.

Actionable interventions

• Assign explicit ownership for gender‑specific design decisions at program level
• Build inclusion checkpoints into governance and decision gates, not just ethics review
• Use cross‑functional forums to surface and resolve inclusion trade‑offs early.

EAST lens: Make accountability Social and explicit, so inclusion becomes part of collective norms rather than an individual burden.

Why Information Alone Does Not Change Practice

Education and guidance are necessary but insufficient. Most teams already know what “good” looks like. Behavior is shaped by defaults, incentives, perceived consequences, and norms. Without changing these structural features, behavior predictably reverts under pressure.

This is why repeated publication of guidance, without operational reinforcement, delivers diminishing returns.

From Diagnosis to Implementation

Behavioral science frameworks, such as EAST, help bridge the gap between intention and impact. They also help teams design interventions that are more likely to work in real environments.

Crucially, behavior change does not rely solely on education. Many of the most effective interventions are discrete changes to systems, processes, and decision architectures that make inclusive choices easier, safer, and more rewarding.

What Teams Can Do Differently, Starting Now

• Redesign protocol templates and governance so inclusion is the default
• Make the risks of exclusion visible and discussable early
• Align incentives and feedback with evidence quality
• Clarify ownership for inclusion decisions
• Use behavioral frameworks consistently to diagnose and address barriers.

If behavior is the barrier, behavioral solutions are required. Sustainable progress will come not from asking teams to try harder, but from redesigning systems so that generating high‑quality, inclusive evidence becomes the natural and expected way of working.