JPP Student Journal Club Commentary: When Adults Evaluate a Child’s Pain: Considering Study Methodology, Gender Bias and Parental Experience

Accurate assessment of pediatric pain is a necessary part of providing healthcare, ensuring the comfort and adequate management of pain for our youngest patients. Though it can appear deceptively simple, conducting a pain assessment is exquisitely complex, taking into account patient self-report, caregiver and provider perceptions, situational context, developmental stage, disease and medical factors, behavioral expressions (facial expressions, posture, and verbalization), and cognitive biases present in any decision-making (Croskerry, 2003). One area of bias that has been investigated in the adult literature is the assumptions, expectations, or beliefs related to an individual’s health on account of their gender (Risberg, Johansson, & Hamberg, 2009). Though uncomfortable to acknowledge in a time when many efforts promote gender equity, gender biases continue to be widespread in our lives, work, education, and healthcare. Pain medicine is no exception: research shows women’s pain is taken less seriously, women receive less effective pain medications, and mind-body dualism is applied in gendered ways to our understanding of pain, with women’s pain more often attributed to psychological factors (Samulowitz, Gremyr, Eriksson, & Hensing, 2018). Previous research has suggested that gender biases in children’s healthcare are present globally (Khera, Jain, Lodha, & Ramakrishnan, 2014), though little attention has been paid to the potential of gender to influence pediatric pain specifically (Boerner et al., 2018). In the current issue of the Journal of Pediatric Psychology, (Earp et al., 2019) provided an extension of a study by Cohen, Cobb, & Martin (2014), showing adult participants rated the pain of an ambiguously gendered child as being more intense when the child was described as a boy, compared to a girl. Earp and colleagues replicated this finding with a larger, more diverse sample of adults than the original study. There are several unique aspects of the methodology of this study that warrant comment. First, the authors are to be commended for their use of a replication and extension design, and the pre-registration of this project through Open Science Framework (https://osf.io), where they have made the protocol, survey materials, raw data, and analysis syntax available and all changes are tracked. Such open science practices critical for transparency in reporting, encouraging replication and extension, increasing the ease with which researchers can conduct meta-analyses, and enhancing scientific rigor by encouraging data analysis by outside sources. The transparency around methodology afforded by such an approach (particularly for tightly controlled experimental studies such as that by Earp and colleagues) substantially improves the readers’ ability to understand and critically appraise the procedure and results. The second unique methodological approach is the use of Amazon Mechanical Turk (MTurk), an online crowdsourcing platform where adult workers are paid a small amount to participate in tasks, such as research studies. As the authors describe, MTurk has been increasingly used as a form of recruitment in social science research, though its use is relatively uncommon in pediatric research. There are some benefits that the authors enumerated, such as a more diverse and attentive sample than undergraduate student studies. However, there are also some concerns about methods such as MTurk, including the generalizability and reliability of findings from participants who participate in multiple consecutive research studies, and the ethics of fair compensation for MTurk workers (Buhrmester, Talaifar, & Gosling, 2018; Hamby & Taylor, 2016). In their investigation of gender bias, a central finding by Earp and colleagues was that controlling for participant’s endorsement of explicit gender stereotypes eliminated the effect. The authors’ explanation for this phenomenon was that the stereotyped belief about girls expressing more pain than boys (e.g., girls are more emotive and boys are more stoic) accounted for the gender-based discrepancy in pain ratings, as the overt display of pain exhibited by the child may be interpreted as being above a certain threshold for the boy (i.e., “he must be in a lot of pain to show that much pain behavior”), whereas it would be interpreted as more typical and less noteworthy for a girl. This is a striking finding, given that all participants were observing identical videos of a real child, and the only difference was the name and pronoun used to describe the child. It is notable that the child in the video was actually female, which may complicate interpretation of the results of this judgment study. Pilot testing revealed that participants were likely to correctly identify the child as female. Societal norms of hegemonic masculinity dictate that it is generally more acceptable for girls to behave/dress in masculine ways (e.g., “tomboys”), than it is for a boy to behave/dress in a feminine manner (McCreary, 1994). This begs the question of whether the findings may have differed if the ambiguously gendered child in the video had actually been a boy. One could speculate that perhaps the ambiguous nature of the child’s gender buffered against the pervasiveness of gendered stereotypes. Further research is needed to determine whether a more masculine-appearing child identified as being male would have their pain assessed in a similar manner; or whether observers may be more likely to discredit this behavior as being exaggerated in the context of expectations of male stoicism. To further move the field forward, similar judgment study frameworks could be used, modifying the effect of gender-related factors (e.g., how typically “masculine” or “feminine” the child appears) and the intersection of gender bias with other potentially relevant sources of bias [e.g., child race, observer gender and medical education, level of diagnostic ambiguity of the patient’s pain (Bernardes, Costa, & Carvalho, 2013)] to determine their impact on pain assessment, and whether such biases are modifiable through training and awareness (Parker, Larkin, & Cockburn, 2018). An important potential source of additional variation in this study was that subset of participants were parents themselves. While these individuals were not assessing their own child’s pain, research has demonstrated that there are neural changes as a result of becoming a parent that alter many of the cognitive processes that are involved in pain assessment, such as overall cognitive ability, empathy, and emotion regulation (Barha & Galea, 2017; Hoekzema et al., 2017; Kim et al., 2014; Kuo, Carp, Light, & Grewen, 2012). Parents are also involved in socializing gendered behaviors in their children, with mothers and fathers playing different roles in these processes (e.g., Garside & Klimes-Dougan, 2002). Further investigation into whether parental experience of adult raters moderates gendered findings in pain assessments may help consolidate the findings of this study with other seemingly discrepant results [e.g., Moon et al. (2008)]. Another important future direction of research outlined by the authors is to understand how these findings apply in clinical settings. Importantly, this direction should include youth whose gender expressions or identity does not match their assigned sex at birth. While there is a burgeoning literature on the health of gender-diverse youth (Rider, Mcmorris, Gower, Coleman, & Eisenberg, 2018), as well as Disorders of Sex Development where the youth’s condition has potential implications for gender identity (Meyer-Bahlburg, Dolezal, Baker, Ehrhardt, & New, 2006; Sandberg, Pasterski, & Callens, 2017), there is little overlap of these fields with pain research despite the increasing need to address the health disparities of these populations. As the authors point out, because children lack full autonomy and must rely on adults to meet their pain control needs, it is imperative for future research to untangle differential biasing factors influencing various third party pain assessors, such as parents and other caregivers. Considering the way that individual beliefs, biases, and experiences interact with larger systemic issues as they relate to gendered healthcare is likely a complex undertaking, but studies such as that by Earp and colleagues offer us the insight that pediatrics is not immune to this issue, suggesting that we cannot ignore this problem any longer. K.E. Boerner is supported by a Mental Health Fellowship funded by the BC Children’s Hospital Foundation and is a trainee member of Pain in Child Health: A CIHR Strategic Training Initiative. Conflicts of interest: K.E. Boerner and A.K. Dhariwal have received funding to investigate the unique role of parents in understanding and mitigating child symptoms (current shared funding from the BC Children’s Hospital Research Institute and the International Centre for Excellence in Emotionally Focused Therapy). K.E. Boerner has previously held funding related to sex/gender issues in pediatric pain and the role of parents in pediatric pain (project funding from the Canadian Pain Society, American Psychological Association, Dalhousie University Department of Psychiatry, and the Nova Scotia Health Research Foundation, as well as travel awards and poster prizes for this work, and trainee awards from the Canadian Institutes of Health Research Institute for Gender and Health and Institute of Neurosciences, Mental Health and Addiction). K.E. Boerner has also received travel funding and a paper award from the International Association for the Study of Pain in support of her work in the area of sex, gender, and pain.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.