Society for Endocrinology - a world-leading authority on hormones

Endocrinologist 151 Cover
Issue 151 Spring 2024

Endocrinologist > Spring 2024 > Features



Sex differences exist in blood glucose control and in diabetes. Despite this, female sex is under-represented in diabetes research, with the use of men, male animal models and male-derived tissues and cells favoured historically. Incorporating both sexes in experimental design, and using sex as a biological variable, holds the potential to uncover sex-specific mechanisms underlying diabetes pathogenesis. This is key for a shift towards more personalised diabetes treatment.


Even under physiologically healthy circumstances, blood glucose control is sexually dimorphic. For example, women show lower non-fasted blood glucose levels than men, but exhibit higher post-prandial blood glucose concentrations.1 Factors that contribute to the latter include differences in islet insulin secretory capacity and insulin sensitivity. Indeed, several studies have shown that female islets have a superior capacity to secrete insulin and that females are generally more insulin-sensitive.

Given this, it is perhaps unsurprising that sex differences exist in diabetes incidences. Type 1 diabetes is the only autoimmune disease where prevalence is higher in men. Type 2 diabetes is more common, and is diagnosed at a lower body mass index, in men compared with premenopausal women.2,3 Sex differences also exist in diabetes complications, with female sex associated with a higher risk of diabetes-related cardiovascular complications, whilst risk and severity of diabetic foot disease are higher in men.2

Another largely unexplored but critically important area is sex differences in efficacy of diabetes treatments. Metformin, which is used in the treatment of type 2 diabetes, may improve blood glucose levels to a greater extent in men, and is associated with more adverse effects in women.4,5

Sex plays an important role in diabetes pathogenesis, complications and response to therapy. Incorporation of sex as a biological variable from basic science to clinical studies will not only lead to the discovery of sex-specific mechanisms underlying diabetes pathogenesis, it also holds the potential to lead to more sex-tailored treatments, which may limit complications and prove more efficacious. Additionally, understanding the mechanisms by which females are more protected from diabetes may reveal novel therapeutic avenues.


The requirement stipulated by some funding agencies and journals to provide adequate justification for only using one sex in experimental design is a new development in the field of sex differences in diabetes. This should facilitate an increase in the incorporation of (in particular) female animal models into experiments in the future. However, a significant proportion of manuscripts still use only one sex, or fail to report which sex was used (as many as ~40%, according to a recent study published in bioRχiv, which looked at sex incorporation in experimental design from manuscripts published in Diabetes).6 Moreover, whilst some papers report the use of both sexes, sex is not necessarily always used as a biological variable, which limits potential insight into sex-specific mechanisms.

The most accessible change to drive a greater incorporation of both sexes into diabetes research is more accurate reporting. For in vitro studies, the sex of the animal or individual from which cell lines and tissues were derived, and for in vivo studies, the sex of the participants, should be explicitly stated in the methods section of any manuscript. Additionally, when only one sex is used, sufficient justification should be given for the exclusion of the other sex.

Another simple step to improve the incorporation of both sexes into experimental design is to collect data from male and female experimental units separately, rather than pooling this data. This allows for sex to be used as an independent variable in downstream statistical analysis.


Incorporation of sex as a variable in statistical analysis can only be achieved when both sexes are used. This is often problematic with in vivo and ex vivo diabetes research, since female models are often resistant to the desired phenotype: diabetes development. This includes induced diabetes models, such as high-fat-fed and streptozotocin-treated mice, as well as spontaneous models, such as the Akita and KINGS mice.7

Studies have found that diabetes induction protocols can be altered to promote diabetes in female models (for example, by increasing streptozotocin dosage). In addition, interventions (such as high-fat feeding) can be used with spontaneous diabetes models to induce diabetes in females. It is therefore likely that the incorporation of both sexes in diabetes research will require the use of different protocols to achieve diabetes.


Incorporation of both sexes is also critical for in vitro research using cell lines. Since the sex chromosomes will differ between cell lines derived from male and female tissue, it is likely that this may influence the response to a given experimental intervention. Indeed, in the field of cancer research, the sex of the cell line used has been found to directly impact upon in vitro experiments.8

The sex of the INS-1 cell line, a commonly used cell line in diabetes research, is male. Going forward, an enhanced awareness of the potential impact of sex on cell lines and inclusion of cell line sex in publications is necessary. Furthermore, in the development of new immortalised cell lines for diabetes research, scientists should endeavour to develop lines to represent each sex, so sex can be incorporated as a biological variable.


Both men and women stand to benefit medically from an improved inclusion of both sexes and incorporation of sex as a biological variable in experimental design, both at preclinical and clinical levels. Here, we have described simple steps, such as clearly stating sex in method sections, that can be taken to kick-start sex incorporation. However, bolder steps, including the development of different sex cell lines, will inevitably be necessary to ensure that both sexes are equally represented in diabetes research.

Diabetes UK-funded Post-Doctoral Researcher

Reader in Integrative Physiology and Diabetes, School of Cardiovascular and Metabolic Medicine and Sciences, Diabetes and Obesity Theme, King’s College London


  1. Mauvais-Jarvis F 2018 Physiology & Behavior
  2. Kautzky-Willer A et al. 2023 Diabetologia
  3. Tatti P & Pavandeep S 2022 Diabetology
  4. de Vries ST et al. 2020 Drug Safety
  5. Schütt M et al. 2015 Experimental & Clinical Endocrinology & Diabetes
  6. Cherian C et al. 2023 bioRχiv
  7. Daniels Gatward LF et al. 2021 Diabetic Medicine
  8. Kim S et al. 2019 Oncology Letters

This Issue:

Spring 2024

Spring 2024

The Endocrinologist


Summer 2024

Summer 2024