Sex bias in alcohol research: A 20-year comparative study

Highlights

• The percentage of studies using both sexes subjects was significantly higher in 2015–2019 than in 2000–2014.

• Clinical studies showed a drastically higher percentage of both-sexes studies than basic studies utilizing other mammalians.

• Reasons for the use of single-sex subjects mainly included insufficient subject numbers and misconceptions of females.

• The variations in inclusion and analysis of the sexes across different biomedical disciplines for alcohol research were wide.

• Implementation of NIH SABV policy increased ratio of NIH-funded papers with sex-based analysis.

Abstract

We aimed to assess the sex-inclusive and sex-based analysis bias in alcohol research for the past 20 years. Data were abstracted from 2988 original research articles published from 2000 through 2019 in 51 representative journals across 9 biomedical disciplines. An analysis in 5-year intervals revealed that the percentage of studies using participants of both sexes was significantly higher between 2015 and 2019 than between 2000 and 2014. When stratified, clinical studies showed a higher percentage of both-sex studies compared to basic studies using animals. The reasons for the use of single-sex cohorts mainly included insufficient participant numbers and misconceptions surrounding the hormonal variability of females. Implementation of the NIH SABV policy promoted the ratio of NIH-funded papers with sex-based analyses. In conclusion, sex bias in alcohol-related biomedical studies has improved over the past 20 years, particularly after the implementation of the SABV policy. Although clinical studies increasingly included sex-based analysis, basic studies were biased towards the use of males.

Introduction

Excessive alcohol consumption, such as binge drinking, heavy drinking, and any drinking by pregnant women, is associated with a variety of short-term and long-term health risks. The development of acute and chronic liver diseases, neuronal system dysfunction, heart disease, mental health problems, and social problems can be found in most heavy drinkers (Friedmann, 2013). With 3 million deaths per year, alcohol is currently one of the leading causes of preventable death worldwide (Witkiewitz et al., 2019). Alcohol use and drinking patterns differ significantly between men and women, and most alcohol-related diseases have sex differences. Women are more susceptible than men to the toxic effects of alcohol on the liver for any given dose of alcohol (Szabo, 2018), whereas men have a greater tendency toward alcoholism than women (Kanny et al., 2018). However, a greater incidence of alcohol-related neurological damage is observed in women than in men (Mancinelli et al., 2009), and stress-induced drinking initiation, maintenance, and relapse have become evident in women in recent years (Peltier et al., 2019). Additionally, women are more vulnerable to alcohol-related cardiovascular conditions than are men (Agabio et al., 2016). Alcohol consumption also generates injury to the reproductive systems of both men and women, with distinct mechanisms. Menstrual irregularity due to disrupted hormone levels is found in women with a chronic or binge drinking history (Carroll et al., 2015). Increased endogenous estradiol and testosterone levels are associated with the incidence of female-related diseases, such as infertility, spontaneous abortion, and cancers (e.g., breast and ovarian) (Dam et al., 2016, Erol et al., 2019). In men, acute and chronic consumption of alcohol suppresses testosterone synthesis via elevation of cortisol, and ethanol also causes oxidative stress and lower semen quality (Jensen et al., 2014). Thus, excessive alcohol consumption is considered to be a potent contributor to male infertility (Ricci et al., 2017). The mechanisms responsible for sex-related differences after alcohol consumption are complex, and may imply the existence of genetic, hormonal, emotional, socioeconomic, and even political differences between men and women (Erol and Karpyak, 2015, Salvatore et al., 2017).

Although the sex differences in alcohol consumption and related disorders are well known, the consideration of sex in medical research has historically been inadequate, as it is widespread practice to use male participants in basic and preclinical research. In several clinical trials, only male patients were recruited to avoid variability (Beery and Zucker, 2011, Sugimoto et al., 2019). One reason for this sex bias in animal research is that sex hormone levels of females exhibit fluctuations with the estrous cycle (Kubota et al., 2016), which may influence alcohol metabolism and tissue vulnerability (Dazzi et al., 2007). From the early 2000 s, there have been numerous calls to equally include both male and female participants in biomedical studies (Beery and Zucker, 2011, Mogil and Chanda, 2005, Holdcroft, 2007, Kim et al., 2010). In 2016, the US National Institutes of Health (NIH) implemented a “sex as a biological variable” (SABV) policy requiring inclusion of both sexes in any funded research unless authors could justify exclusion (Clayton and Collins, 2014, Clayton, 2018). However, the percentage of studies that use only 1 sex or avoid analyses of sex differences remains high in many biomedical disciplines, possibly because inclusion of both sexes is time-consuming, costly, and includes more data variability (which, at least for preclinical alcohol research, has been shown to be untrue) (Priddy et al., 2017); thus, authors are disincentivized to study sex differences (Woitowich and Woodruff, 2019, Arnegard et al., 2020). Sex bias analysis has recently been reported in biology (Woitowich et al., 2020), neuroscience (Mamlouk et al., 2020), surgery (Mansukhani et al., 2016), dermatology (Kong et al., 2016), microbiology/immunology (Potluri et al., 2017), and cardiovascular pharmacology (Scott et al., 2018). Alcohol consumption causes evident sex-related differences, yet no study has analyzed the progress of sex-inclusive research practices. Here we present a 20-year comparative study (2000–2004, 2005–2009, 2010–2014, and 2015–2019) to assess sex-inclusive research practices in 2988 papers published in 51 scholarly journals across 9 biomedical disciplines. Clinical and basic studies were further stratified to analyze the differences between human trials and animal studies.