Sex, the biological basis of female and male distinctions (see Term: Sex), is an important variable to consider when setting research priorities, developing hypotheses, formulating study designs.
In biomedical research, sex may need to be analyzed—in human and animal research subjects, and in organs, tissues, cells and their components (Tannenbaum et al., 2016; Clayton, 2016; IOM, 2012; Beery et al., 2011; Wizemann et al., 2001). In engineering, sex may need to be analyzed at the level of user physiology and biomechanics in both product and systems design (see Analyzing Standards and Reference Models).
Analyzing sex involves six steps:
- 1. Reporting the sex of research subjects or users. This is a prerequisite to sex analysis. Some granting agencies and peer-reviewed journals require reporting sex—for human, animal, and (where appropriate) organ, tissue and cell research (see Policy Recommendations). Reporting the sex of the research subject or considering the sex of the user/customer is important even in single-sex studies to allow meta-analysis, identify research gaps, and prevent over-generalizing findings beyond the sex studied. For example, in osteoporosis, early research used bone density reference models derived from healthy young white women to correlate bone density with fracture risk in older women, then applied this single-sex (female) model to men to estimate older men’s fracture risk. Later, researchers established a male reference population and developed diagnostics that account for sex, age, and other factors (see Case Study: Stem Cells).
- 2. Recognizing differences that exist within groups of females and males/women and men. Both biological and sociocultural factors differ substantially among individuals within each sex over their respective lives. They include profound changes associated with reproductive biology (such as occur at puberty and, in women, throughout the menstrual cycle, during pregnancy, and at menopause) and with aging. Take, for example, height. In the U.S., women are shorter than men on average, but about 3% of women are taller than the average man, and 6% of men are shorter than the average woman. The height difference between the average woman and man is less than the height difference between a 90th percentile woman and a 10th percentile woman, or the difference between 90th and 10th percentile men (see chart; see also Case Studies: Human Thorax Model: Pregnant Crash Test Dummies).
- 3. Collecting and reporting data on factors intersecting with sex in study subjects or users/consumers. Women and men (females and males) may differ by age, lifestyle (e.g. diet, physical activity, use of tobacco, alcohol, and other drugs, etc.), socioeconomic status, and other gendered behaviors and variables (see Analyzing Factors Intersecting with Sex and Gender). Efforts should be made to “match” female and male cohorts according to variables that might influence interpretation of study findings (see Designing Health & Biomedical Research). For example, in the development of prostheses for total knee arthroplasty, overlooking intersecting factors led to a focus on sex that did not improve patient outcomes (see Term: Overemphasizing Sex Differences as a Problem). Prosthesis designers observed statistically significant differences between women and men’s knee anatomy and produced a “gendered knee” which was marketed to women patients. Although biological sex does give rise to differences in knee anatomy, sex may not be the principal factor for prosthesis selection—in this case, height is a more important variable for matching patients to prostheses (see Case Study: De-Gendering the Knee).
- 4. Analyzing and reporting results by sex. Sex-specific analyses should be conducted and the findings reported. For example, women and men may require different airbag inflation energies, or different dosages of a drug to produce a given effect, because of differences in body size and composition. Adjusting the data for baseline differences and factors that intersect with sex is a crucial step in understanding the sex difference observed. For example, researchers who analyzed sex in studies of cardiovascular disease identified sex differences in arterial plaque formation: women tend to develop diffuse plaques, whereas men develop localized plaques (von Mering et al., 2004). This difference has ramifications for the design of stents (see Case Study: Heart Disease in Women).
- 5. Reporting null findings. Researchers should report when sex differences (main or interaction effects) are not detected in their analyses to reduce publication bias, an important consideration in meta-analyses (IOM, 2012). Where relevant, researchers should note when data regarding sex differences are statistically inconclusive, especially in the context of factors intersecting with sex. Statistical power may be limited in cases where it is difficult to recruit patients of one sex, for example.
- 6. Meta-Analysis. Good design and clear reporting may enable cross-study analysis. Combining data from multiple studies can increase statistical power, but it can also compound error, especially if factors intersecting with sex and gender are overlooked (Blauwet et al., 2007; Bailey, 2007).