Reporting Sex and Gender in Research (SAGER)

SAGER asks a simple question of every manuscript: did sex and gender matter here, and if so, can a reader see how you handled them? Here is how to answer it.

Explainer · 10 min read · Directive Publications Editorial Team

What the SAGER guidelines are

The Sex and Gender Equity in Research (SAGER) guidelines are a reporting guideline for one specific question: how a paper handles sex and gender. They were developed by the Gender Policy Committee of the European Association of Science Editors (EASE), which EASE set up in 2012, and published in May 2016 in Research Integrity and Peer Review as Heidari et al., "Sex and Gender Equity in Research: rationale for the SAGER guidelines and recommended use".

According to that paper, a panel of 13 experts from nine countries built the guidelines over three years. The work drew on a survey of 716 journal editors, scientists and others in publishing, a review of existing journal policies, and consultation with outside experts in sex and gender research.

Unlike CONSORT or STROBE, SAGER is not tied to one study design. It applies to research with humans, animals, and material derived from them, such as cells and tissues. It also covers fields whose results are applied to people, such as engineering and device testing. If you are unsure how SAGER fits with your design-specific guideline, our guide to choosing between CONSORT, PRISMA and STROBE covers the design side. SAGER sits alongside whichever of those you use.

Get the two terms right first

A general principle of SAGER is that authors should use the terms sex and gender carefully, so that one is not confused with the other. The paper's glossary defines them this way:

• Sex is a set of biological attributes in humans and animals, including chromosomes, gene expression, hormone function and reproductive or sexual anatomy. It is usually categorised as female or male, although the attributes vary.
• Gender is the socially constructed roles, behaviours and identities of female, male and gender-diverse people. The SAGER authors note that it is often wrongly treated as a binary.

In practice, this means your Methods should say which one you measured and how. SAGER asks authors to state whether sex was based on self-report, on examination of body characteristics, on genetic testing or on other means. A demographics table labelled "gender" that was filled from a hospital record's sex field is the kind of mismatch reviewers notice.

The ICMJE Recommendations (updated January 2026) say the same in their own words. They ask authors to use sex when reporting biological factors and gender for identity, psychosocial or cultural factors. Unless it is inappropriate, they ask authors to report the sex and/or gender of participants and the sex of animals or cells, and to describe the methods used to determine sex and gender.

What SAGER asks for, section by section

SAGER maps onto the structure of a paper. This table summarises the recommendations in the 2016 paper. For the full wording, go to the source.

Two details in the full text matter more than they look.

One-sex studies need a flag in the title. For a study of a non-sex-specific issue in which only one sex was used, SAGER says the title and abstract should include "in males" or "in females". A mouse study that used only males and does not say so in the title is the simplest SAGER failure to spot, and the simplest to fix.

Null results still get reported. SAGER asks for sex- and gender-based analyses to be reported regardless of a positive or negative outcome. Finding no difference is useful information for later meta-analyses, which is why SAGER also recommends making raw data available disaggregated by sex and gender for future pooling.

Animals, cells and devices

Senior authors in laboratory science sometimes assume SAGER is about clinical trials. It is not. The guidelines give explicit instructions for non-human work:

• Animal studies: use the term sex, and specify it in the title and abstract where only one sex was used.
• Primary cells and tissues: state the sex of the subjects or source donors for in vivo, in vitro and ex vivo work. In cell, molecular and biochemical experiments, state the origin and sex chromosome constitution of the cells, and the reasons if it is unknown.
• Immortalised cell lines: SAGER makes an exception for immortalised cell lines, which are highly transformed.
• Unknown sex: for embryonic cultures, mixed-culture lines or older experiments where sex was not recorded, SAGER recommends determining sex by chromosomal analysis. It says "mixed" or "unknown" should be used only when sex cannot be determined by any method.
• Devices and technology: explain whether the product will be used by all genders and whether it was tested with the user's gender in mind.

This is where a lab lead's habits matter most. Cell-line provenance and animal sex are easy to record at the bench and hard to reconstruct three years later when a reviewer asks.

The checklist editors use

The original paper presented SAGER as a table, and editors later found that a checklist would make compliance easier to monitor. In 2022, editors from the Lancet group published a SAGER checklist in European Science Editing (Van Epps et al.). It follows the structure of the CONSORT checklist and comes in two versions: one for studies with human participants, and one for other studies such as applied science and cell biology.

The human-participant version adds items that go beyond the original SAGER table. These are worth knowing because they reflect what at least some editors now ask for:

• A sex or gender breakdown of the study population in the abstract
• Separate rows in Table 1 for male, female and other sex or gender categories if collected
• Sex- and gender-based analyses and, for clinical trials, adverse event data disaggregated by sex or gender: in the main paper if pre-specified, otherwise in the appendix
• An explicit scientific rationale for contraception requirements and for excluding people who are pregnant or lactating

The authors reported that, at the time of writing, Lancet journals were using the checklist after initial peer review and sending sex and gender requests in the decision letter. They noted that, ideally, it would be used before peer review. If you would rather not meet those requests at revision stage, run the checklist before you submit.

Where SAGER sits in journal and funder policy

SAGER is a guideline, not a law, and how strongly it is applied depends on who is asking.

• ICMJE encourages authors to refer to SAGER, alongside asking for correct use of the terms and for the influence of sex and/or gender to be discussed where appropriate.
• Publishers: Elsevier reports that in 2023 it, Cell Press and The Lancet incorporated SAGER into their information for authors. Springer Nature publishes a SAGER editorial policy that encourages authors to follow the guidelines.
• WHO: in December 2023, a comment in The Lancet announced WHO's adoption of SAGER together with the GATHER statement for health estimates.
• Funders: some funders set their own expectations at the design stage. For example, NIH notice NOT-OD-15-102 introduced its policy on sex as a biological variable for applications from January 2016. Funder policies change, so check the current version that applies to your grant.
• Indexes: SAGER is listed in the EQUATOR Network library of reporting guidelines, and EASE lists volunteer translations into several languages.

The practical point is that the target journal's instructions for authors decide what you must do. SAGER tells you how to do it well.

Mistakes that draw reviewer comments

These problems come straight from what SAGER asks authors to avoid or explain:

1. Using sex and gender as synonyms. Pick the right term for what you measured and say how you measured it.
2. Leaving a one-sex study unflagged. If only one sex was studied for a condition that affects both, the title and abstract should say so, and the Methods should justify it.
3. Post hoc subgroup claims. SAGER advises against post hoc gender-based analysis when the study design cannot support meaningful conclusions. If you present an exploratory split, label it as exploratory and avoid causal language. Our guide to reporting p-values and confidence intervals covers how to present it.
4. Silently dropping a null sex analysis. If you ran it, report it.
5. Treating covariation as causation. SAGER notes that covariation between biology and behaviour is not evidence of physiological causation. Consider social, cultural and situational explanations too.
6. Saying nothing when no analysis was done. A missing analysis is acceptable if you explain why and discuss its consequences. Our guide to writing a limitations section reviewers trust shows how to frame it without undermining the paper.

A pre-submission check for senior authors

Before the manuscript goes to the journal, check:

1. Sex and gender are used correctly throughout, including in tables and figure legends
2. Methods state how sex and/or gender were determined
3. One-sex studies are flagged in the title and abstract, and the exclusion is justified
4. Animal sex, and the sex of donors of primary cells or tissues, are stated
5. Table 1 shows every sex or gender category collected
6. Sex-disaggregated results are reported where appropriate, including null findings
7. Any exploratory sex or gender analysis is labelled as exploratory
8. The Discussion covers generalisability across sexes and genders, or explains why no analysis was done
9. The target journal's own requirements on sex and gender have been checked

The cheapest place to get SAGER right is the protocol, not the manuscript. If sex and gender are written into the design, the reporting largely takes care of itself. Directive Publications sets out what it asks of authors on its for authors page. Read it alongside your target journal's own guidance before you submit.