Background & Aims
Women are more likely than men to experience a wide variety of pain-related conditions1. Despite the clear sex differences in these disorders the etiology of these disparities is not fully understood. Some have attributed these disparities to the social dimension of pain across genders due to differences in psychosocial variables such as willingness to report pain, gender-specific stressors, and gender roles1,2. Although individual social factors have been investigated for their role in determining sex differences in pain, previous work has not taken the step to explore the overarching role that conformity to gender norms plays in determining these outcomes. In the following study, we derive a weighted gender index measuring each participant’s conformity to gender norms based on a wide array of psychological and social features. We then use this gender index to assess gender’s effect on the prevalence of chronic pain.
Methods
We are conducting research within the All of Us Research Program, which includes a diverse group of 409,420 Americans, reflecting a wide range of socioeconomic and ethnic backgrounds. We determined the odds ratios for various conditions based on sex using Fisher’s exact test within this cohort. Subsequently, we developed a gender index via logistic regression, taking into account 72 psychosocial factors to estimate the probability of each participant being female or male. This probability, ranging from 0 to 1, served as a gender index to assess each person’s adherence to the cohort’s gender norms. After stratifying by sex, we computed the odds ratios to see how increases in femininity, as measured by the gender index, correlated with the likelihood of developing each condition.
Results
Women showed higher odds of having a variety of pain-related conditions including Headaches (OR=1.82, p=0), Migraines (OR=2.78, p=0), Irritable Bowel Syndrome (OR=2.28, p=1.57×10-138), abdominal pain (OR=1.65, p=1.2×10-14), and Chronic Fatigue Syndrome (OR=1.78,p=2.84×10-43). Our logistic regression model was able to differentiate the sexes with moderate accuracy in a held-out testing set (AUC=0.74). Increased femininity was associated with increased odds of having female-biased pain-related conditions in both men and women. Values ranged from an OR of 0.06 (female)/0.14 (male) for irritable bowel syndrome to 0.22 (female)/ 0.23 (male) for migraines and 0.14(female)/0.26(male) for abdominal pain (all p values<0.0001). Gout, a pain condition with a higher prevalence among men (OR=0.22, p=4.12x10-312), showed an association with masculinity across sexes OR of 0.15 (female)/ 0.16 (male), p<0.0001.
Conclusions
Chronic pain is the number one cause of disability worldwide1 with women disproportionately experiencing this burden. Understanding the mechanisms underlying this disparity will be essential to making informed treatment and public policy recommendations. In our research, we devised a weighted gender index that was able to discriminate between the sexes and showed associations with key female-biased pain disorders. These findings impact how we understand the epidemiological differences in pain prevalence between the sexes and can help guide health policy.
References
1. Fowler SL, Rasinski HM, Geers AL, Helfer SG, France CR. Concept priming and pain: an experimental approach to understanding gender roles in sex-related pain differences. Journal of Behavioral Medicine. 2011;34(2):139–147.
2. Wuest J, Merritt-Gray M, Ford-Gilboe M, Lent B, Varcoe C, Campbell JC. Chronic Pain in Women Survivors of Intimate Partner Violence. The Journal of Pain. 2008;9(11):1049–1057.