The Male Default Prevails in Biomedical Research: Sex Inclusion in Nature (2025)

Despite Nature's 2025 policy requiring sex inclusion reporting, a review of 513 articles reveals that the "male default" persists due to inconsistent experimental design, unbalanced sex ratios, and a lack of meaningful sex-based analysis, indicating that transparency-focused reporting summaries alone are insufficient to ensure rigorous sex inclusion in biomedical research.

The Gist

Imagine the world of medical research as a massive, high-stakes cooking competition. The goal is to create a "universal recipe" (a drug or treatment) that works for everyone. For decades, the judges (scientists and journals) have been told, "Hey, you need to taste-test your dish on both men and women because our palates are different!"

But a new study published in Nature (one of the most famous "cookbooks" in science) in 2025 reveals a shocking truth: The chefs are mostly still cooking for men, even when they claim to be cooking for everyone.

Here is the breakdown of what the paper found, using simple analogies:

1. The "Checklist" Illusion

The journal Nature introduced a rule: "Before you publish your recipe, you must fill out a checklist saying whether you used male or female ingredients."

• The Reality: The chefs filled out the checklists, but they were playing a game of "minimal compliance."
• The Analogy: It's like a restaurant owner telling a health inspector, "Yes, I have a fire extinguisher!" (checking the box), but then hiding it in a locked closet and never actually using it.
• The Data: About 63% of the recipes claimed to use "both sexes." But when the researchers looked closer, they found that in many of these cases:
  • They only used one sex for the actual cooking, even if they said they had both.
  • They used 100 men and 1 woman (a 100:1 ratio), which isn't really "both."
  • They didn't tell you how many of each they used, so you couldn't verify anything.

2. The "Default Setting" Problem

The study found that when scientists did pick just one sex to study, they almost always picked men.

• The Analogy: Imagine a video game where the character "Male" is the default skin that works for every map. If you want to play on a "Female" map, you have to unlock it as a special, rare level.
• The Finding:
  • Female-only studies were mostly about things that only happen to women (like ovarian cancer). This makes sense.
  • Male-only studies were about things that happen to everyone (like heart disease, Alzheimer's, or liver issues).
  • The Result: We are learning how diseases affect men, and assuming those results apply to women. But biology isn't a "one-size-fits-all" t-shirt; it's more like a suit that needs tailoring.

3. The "Noise" Excuse

Why do scientists keep picking men? The paper found they often use the excuse that women are "too complicated."

• The Analogy: Imagine a scientist saying, "I'm only testing this car on a straight, flat road because testing it on a bumpy, winding road (women's hormonal cycles) would make the data messy and hard to understand."
• The Problem: By avoiding the "bumpy road," they miss out on how the car actually performs in real life. They are prioritizing a "clean" experiment over a "real" one.

4. The Dangerous Consequence: The "Masked" Drug

The paper gives a scary example involving a real Alzheimer's drug called Lecanemab.

• The Story: When the drug was first tested, they mixed men and women together in one big group. The result looked okay: "It slows memory loss by 27%!"
• The Twist: Years later, someone looked at the data separately.
  • Men: The drug worked amazingly well (43% improvement).
  • Women: The drug barely worked (12% improvement).
• The Lesson: Because they didn't separate the data, the drug looked like a "success" for everyone, but it was actually a failure for half the population. If they had analyzed the sexes separately from the start, they might have realized the drug needed to be dosed differently for women.

5. Why This Matters

The authors argue that just asking scientists to "fill out a form" isn't enough.

• The Metaphor: It's like asking a driver to check a box saying "I have a seatbelt," but not actually requiring them to wear it.
• The Solution: The journal needs to stop accepting "checklist" answers. They need to demand that scientists not only include both sexes but also compare them to see if the results are different.

The Bottom Line

For too long, science has treated "Male" as the standard human and "Female" as a special exception. This paper shows that even in the most prestigious journals, this habit is still alive and well. Until journals stop accepting "we didn't check because it's too hard" as an excuse, we risk creating medicines that work great for men but fail (or even hurt) women.

In short: We can't build a bridge that only holds up for one type of weight. We need to test the bridge for everyone, or the whole thing might collapse.

Technical Summary

1. Problem Statement

Despite over 30 years of recognition that sex is a critical biological variable (SABV) influencing disease prevalence, presentation, and therapeutic response, its integration into biomedical research remains inconsistent. While major funding agencies (e.g., NIH, CIHR) and guidelines (e.g., ARRIVE, SAGER) have mandated the inclusion of sex in study design, and journals like Nature have introduced reporting checklists (Life Science Reporting Summary), there is a lack of empirical evidence regarding the effectiveness of these policies.

The core problem is that transparency-focused reporting requirements may be insufficient to drive meaningful changes in study design and analysis. There is a persistent "male-default" model where male subjects are treated as the generalizable norm, while female subjects are excluded or treated as a source of "noise," leading to a lack of sex-disaggregated data and missed opportunities to discover sex-specific biological mechanisms.

2. Methodology

• Study Design: A cross-sectional analysis of original research articles published in the journal Nature in 2025.
• Sample: 513 original research articles involving:
  • Human participants.
  • Vertebrate animals.
  • Cell lines.
• Data Collection:
  • Articles were screened for mentions of sex/gender in the main text and the Life Science Reporting Summary.
  • Search terms included: "female," "male," "women," "men," "sex," "gender."
• Coding Criteria:
  • Inclusion: Whether both sexes were included, or if single-sex designs were used.
  • Reporting: Whether sample sizes were reported by sex.
  • Analysis: Whether sex was treated as a discovery variable (analyzed as an independent variable) versus a covariate or pooled.
  • Justifications: Reviewers extracted author-provided justifications for excluding one sex or failing to analyze by sex.
  • Context: Whether single-sex studies focused on sex-specific conditions (e.g., ovarian cancer) or general conditions (e.g., aging, neurodegeneration).

3. Key Results

The study reveals a significant gap between policy intent and actual research practice:

• Reporting Gaps: While 82.1% of articles reported sex, 17.9% did not report it in either the manuscript or the summary. Notably, over 30% of studies failed to disclose sex in the main manuscript itself.
• Nominal Inclusion: Of the 62.6% of articles that claimed to include both sexes:
  • 45.5% did not report sample sizes by sex, making it impossible to verify balance or inclusion across experiments.
  • 38.3% exhibited flawed inclusion practices, such as:
    • Reverting to single-sex experiments within the same paper.
    • Alternating between male-only and female-only experiments.
    • Using highly skewed sex ratios (≥2:1).
    • Treating sex solely as a covariate rather than an independent variable.
• Lack of Discovery Analysis: Only 7.4% of all articles (and 11.8% of those including both sexes) explicitly analyzed sex as a discovery variable (i.e., testing for sex differences).
• The Male Default in Single-Sex Studies:
  • Single-sex studies accounted for 19.5% of all articles.
  • Female-only studies (46% of single-sex) were almost exclusively focused on sex-specific conditions (e.g., ovarian/endometrial disorders).
  • Male-only studies (54% of single-sex) predominantly focused on conditions affecting both sexes (e.g., aging, liver disease, neuroscience), with only 11.1% addressing male-specific conditions (e.g., prostate disease).
• Author Justifications: Common rationales for excluding females included:
  • Assumed biological equivalence between sexes.
  • Framing female hormonal cycles as a source of "noise" or variability.
  • Methodological constraints (sample size).
  • Claims that sex differences were "beyond the scope" or "irrelevant."

4. Key Contributions

• Empirical Evidence of Policy Failure: The study provides concrete data showing that the Nature Life Science Reporting Summary, while increasing transparency, has not successfully eliminated the male-default bias or ensured rigorous sex-disaggregated analysis.
• Definition of "Nominal Inclusion": The authors identify a critical category of research where both sexes are technically "included" but the study design (skewed ratios, alternating sexes, lack of disaggregated data) renders the inclusion scientifically meaningless.
• Systemic Bias Identification: The paper highlights that the exclusion of females is not random; it is systematically applied to general conditions, whereas males are excluded only for male-specific conditions. This reinforces the conceptual norm that male biology is the "universal" standard.
• Editorial Policy Critique: The study contrasts Nature with other Nature Portfolio journals (e.g., Nature Medicine, Nature Communications) that adopted stricter, SAGER-aligned policies in 2022. The findings suggest that Nature's lack of similar enforcement allows the male-default model to persist.

5. Significance and Implications

• Scientific Rigor and Reproducibility: The failure to analyze sex as a discovery variable obscures clinically meaningful differences. The authors cite the Lecanemab Alzheimer's therapy case as a prime example: a pooled analysis showed a 27% overall benefit, but sex-disaggregated analysis later revealed a 43% benefit for males and only 12% for females. Ignoring sex differences risks ineffective or harmful treatments for half the population.
• Editorial Responsibility: As a flagship journal, Nature sets global research norms. The study argues that checklists alone are insufficient. To advance rigor, journals must:
  • Require sex-disaggregated data presentation.
  • Enforce stricter standards for justifying single-sex designs.
  • Mandate sex as an independent variable in analysis, not just a covariate.
• Future Directions: Without stronger editorial expectations and enforcement, reporting summaries risk becoming "administrative checkboxes" that entrench the male-default model rather than advancing the generalizability and translational impact of biomedical research.

Conclusion: The paper concludes that while transparency is a necessary first step, it is not a sufficient solution. The biomedical community must move beyond reporting that sex was considered to demanding how sex was integrated into the experimental design and statistical analysis to ensure research is valid for all humans.