Precision Medicine Can’t Deliver Without Representation.

Rethinking Representation in Clinical Trials

The Illusion of Precision

Precision medicine promises to individualize treatment.

But when the underlying data excludes entire populations, we don’t just fall short of that promise—we reinforce old disparities under a new name.

If we build personalized medicine on incomplete datasets, we don’t just risk inefficacy. We risk institutionalizing inequity—under the banner of clinical precision.

Why this matters to Latinos in STEMM:

As more Latinos enter research, medicine, and biotech leadership, we have an opportunity—and a responsibility—to shape how science is done. Representation in clinical trials isn’t just about who gets included as patients. It’s about who asks the questions, leads the studies, and ensures equity is built into the evidence from the start.

The Data Gap in Plain Sight: Incomplete Evidence, Incomplete Care

Clinical trials remain the gold standard of evidence in medicine. But the populations they’re based on rarely reflect the full diversity of those most impacted by chronic disease.

Let’s take a look:

⚠️ Underrepresentation Is the Norm

Latinos = 19% of the U.S. population 👉 Only 5–6% of clinical trial participants【1】
Black Americans = 13% of the U.S. population 👉 Just 8% of trial participants【1】
In oncology, Black participation can be as low as 4%—even for cancers with higher mortality in Black patients【2】

These aren’t just statistical gaps. They’re blind spots that limit our understanding of safety, dosing, and effectiveness for millions of Americans.

Same Drug, Different Outcomes

Pharmacogenomic differences can significantly affect how individuals respond to medications—altering metabolism, efficacy, and risk of side effects. But when clinical trials don’t include diverse populations, these differences may go undetected until a drug reaches the broader market.

A few well-documented examples:

lopidogrel (Plavix): Up to 30% of Latinos and Asians carry CYP2C19 loss-of-function alleles, reducing the drug’s antiplatelet effect【3】
Codeine: CYP2D6 variation can cause poor pain control—or dangerous toxicity—in patients with certain metabolizer profiles【4】
Warfarin: Variants in CYP2C9 and VKORC1 affect dosing needs. Some polymorphisms are more common in Hispanic populations, increasing bleeding risk if not accounted for【9, 10】
Irinotecan: Reduced UGT1A1 activity (especially 28/28 genotype) increases risk of toxicity; prevalence differs across ethnic groups【11】
Rosuvastatin: Asian populations show higher blood concentrations, prompting lower recommended starting doses【12】
Albuterol: Among Latino children with asthma, Puerto Ricans and Mexicans show distinct responses to albuterol, highlighting ethnicity-specific treatment variability【13】

These are just the known cases. The concern is what we may be missing.

Without adequate representation in trials, we risk overlooking unknown genetic or metabolic differences that affect how underrepresented populations respond to new treatments.

The consequences often don’t appear until post-market surveillance—when the stakes are much higher and the opportunity to adjust is limited.

Precision medicine can’t fulfill its promise if we build it on incomplete data.

Well-Studied Diseases, Poorly Represented Populations

Consider three conditions with high disparities and low representation:

Cardiovascular Disease: Black Americans have the highest hypertension rates in the world—yet remain underrepresented in trials for ACE inhibitors and SGLT2 inhibitors【5】
Diabetes: Latinos are 60% more likely to be diagnosed【6】 but are routinely underrepresented in trials that inform treatment
Lupus: 1 in 250 young Black women will develop it. Yet fewer than 15% of participants in lupus trials are Black. Latina representation is even lower【7】

This pattern isn’t accidental—it reflects decades of decisions about who gets studied, whose outcomes are prioritized, and which communities are left waiting.

The Efficiency–Equity Tradeoff No One Wants to Talk About

This isn’t about blame—it’s about confronting structural realities in trial strategy that too often go unspoken.

Pharmaceutical companies are under immense pressure to move fast. Clinical trials are expensive, timelines are tight, and every delay increases costs in an industry where development already exceeds $2 billion per drug.

To meet milestones and ensure high-quality data, many sponsors turn to elite academic centers in the U.S. or conduct trials in Eastern Europe, or other global regions with streamlined recruitment and infrastructure.

When we speed up research by skipping the hardest-to-reach patients, we lock inequity into the system—at scale.

This accelerates innovation—and that’s a good thing. But it also means that communities of color in the U.S. are often bypassed because they’re seen as harder to enroll, slower to consent, or even riskier for data variability.

In the race for speed, cost savings and statistical power, we are systematically excluding the very populations we need more evidence on.

When Race Isn’t Even Reported

Even when studies do include more diverse participants, race and ethnicity data are often omitted:

A JAMA review found fewer than 40% of major randomized trials reported race/ethnicity at all【8】
This makes it nearly impossible to assess outcomes by population—or hold anyone accountable

Transparency isn’t just about ethics. It’s about reproducibility and rigor.

Fear Isn’t the Problem—It’s the Symptom

There’s a common narrative that communities of color don’t trust science.

This is partially true, but what we call “mistrust” is often memory—of harm, exclusion, and being treated as an afterthought in discovery.

Latinos and Black Americans may opt out of trials not because they don’t care, but because:

They've been historically excluded from innovation—except when exploited by it and research organizations and pharma companies have made little effort to help educate those communities on the importance of participation.
They rarely see representation among investigators or outreach staff
Study materials often lack linguistic or cultural competence, creating real barriers to informed consent

This isn’t disinterest. It’s self-protection. Unless we acknowledge that, minorityrecruitment efforts will continue to fall flat.

Why It Matters to Pharma, Biotech, and Policymakers

This isn’t just a DEI issue. It’s a matter of market performance, regulatory success, and scientific validity.

When clinical trials exclude the populations most impacted by disease, here’s what’s at stake:

🧿 Regulatory Scrutiny

The FDA has taken steps to strengthen diversity requirements in clinical trials. Under the Food and Drug Omnibus Reform Act of 2022 (FDORA), most phase 3 trials must now submit Diversity Action Plans. While draft guidance was withdrawn in early 2025, recent Senate hearings suggest momentum to formalize enforcement is still building.

📉 Market Access and Uptake

Therapies that don’t reflect the communities they aim to serve face slower adoption, especially in populations with justified mistrust of the system.

💰 Financial Risk

Post-market safety signals, label revisions, and re-education campaigns are costly. Inclusive trial design can reduce this risk from the start.

📊 Data Credibility

Incomplete trial data undermines external validity. Payers and providers are demanding real-world evidence that actually reflects real-world patients.

🌍 Ethical Responsibility

At the core of science and medicine is the principle of justice. Including diverse populations isn’t just strategic—it’s the right thing to do.

What Needs to Change—Now

Here’s how sponsors, funders, and leaders can start closing the gap:

Invest in Infrastructure Partner with bilingual clinics, FQHCs, and trusted local providers who already serve diverse communities.
Design for Inclusion Up Front Use epidemiologic data to define trial populations—not as a footnote, but as a starting point.
Diversify Research Leadership Representation among PIs increases participation. Require diversity in your investigator network, not just your brochure.
Fund Culturally Relevant Education Community-based research navigators and translated consent materials aren’t extras—they’re essentials.
Measure and Report Transparently Publish inclusion data in primary results. If it matters, make it visible.

What’s at Stake

Science that doesn’t reflect everyone doesn’t serve everyone.

The next frontier in precision medicine is representation.

If we want to build a future where personalized medicine delivers on its promise, we need clinical research that starts with inclusion—not as an afterthought, but as a foundation.

If you're funding innovation, designing trials, or leading research—this is your moment to act. Representation is not a side project. It is the future of effective, ethical science.