Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Aug 1;30(7):e70348. doi: 10.1002/ejp.70348

Gender, Gender Role Beliefs, and Evaluation of Pain in Others: Indirect and Moderated Pathways

Elizabeth J Richardson 1,✉, Melissa Shepherd 1, Zina Trost 2, Stephanie R Mallinas 3
PMCID: PMC13428360  PMID: 42541392

ABSTRACT

Background

Gender disparities in observer pain assessment are well‐documented, with women's pain often underestimated, yet the mechanisms underlying this bias remain unclear. This study examined how patient gender, gender role beliefs, and pain‐related attributions shape perceptions of pain.

Methods

One hundred thirty‐two community participants were randomised to view either a female or male patient with chronic pain performing a movement. Participants rated pain severity, interference, genuineness, and emotional attribution. Moderated serial mediation tested indirect pathways from patient gender to pain severity and interference and whether these effects varied by participants' gender role beliefs.

Results

Compared to the male patient, the female patient was rated as having less severe pain, less interference, more emotional/psychological contributors to the pain, and lower genuineness. Serial mediation indicated that patient gender predicted lower perceived pain interference through greater emotional/psychological attributions, which was associated with reduced perceived genuineness. For pain severity, indirect effects emerged through emotional/psychological attributions and genuineness, though serial mediation was not supported. Gender role beliefs did not moderate serial mediation but did moderate the links between patient gender and perceived genuineness, as well as the direct effect of gender on pain severity.

Conclusions

Findings extend evidence of gender bias in observer pain evaluation by showing that the female patient was judged as experiencing less interference and severity through pathways involving emotional/psychological attributions and reduced credibility. Results underscore the layered nature of gender bias in observer pain assessment and highlight the need for research with more diverse patients and clinical providers.

Significance Statement

This study advances understanding of gender bias in pain assessment by identifying pathways through which gender discrepancies may arise and how such pathways may be further conditioned by evaluators' gender role beliefs. These findings highlight the need to address both implicit gender stereotypes and attributional reasoning that may be associated with underestimating women's pain to promote more equitable care.

1. Introduction

Gender disparities in observer pain assessment and treatment are well‐documented, with women's pain often underestimated, rated less severe, and treated less aggressively than men's (Chen et al. 2008; Tait et al. 2009; Lord et al. 2009; Cleeland et al. 1994; Clerc Liaudat et al. 2018). When presenting with similar pain complaints, women are more likely to receive antidepressants or psychotherapy, while men receive physical therapy or radiological workup (Lord et al. 2009; Miller et al. 2018; Stålnacke et al. 2015). Clinic notes concerning pain also more often emphasise emotional aspects for women and biomedical diagnoses for men (Markowitz 2022). Importantly, both male and female providers prescribe pain medications less often and record pain scores less frequently for women compared to men (Guzikevits et al. 2024), suggesting gender biases in pain assessment reflect broader social stereotypes rather than the gender of the observer.

The subjective nature of pain may enable assessment through the beliefs and biases of the healthcare provider (Hadjistavropoulos and Craig 2002). For instance, women's pain is often minimised due to implicit beliefs framing it as psychologically driven, leading to perceptions that it is exaggerated or less genuine (Lloyd et al. 2020). Women with pain are more frequently judged as emotional and complaining, with their pain attributed to emotional/psychological causes more than men's (Samulowitz et al. 2018). These patterns reflect societal stereotypes of men as stoic and women as sensitive or emotional, stereotypes linked to underestimating women's pain (Zhang et al. 2021).

Gender role beliefs refer to expectations about appropriate behaviour for men and women (Kerr and Holden 1996), ranging from traditional views emphasising distinct gender roles to beliefs promoting equality (Beere et al. 1984; Kerr and Holden 1996). Traditional gender role beliefs endorse women in domestic roles and men as protectors and providers (Brown and Gladstone 2012; Kerr and Holden 1996). Health professionals with higher sexism and traditional gender role beliefs rated women's pain as less credible than men's (Prego‐Jimenez et al. 2022). Such gender stereotypes were also linked to assumptions that women exaggerate pain while men understate it (Paganini et al. 2023) and may explain why women with pain are more often recommended mental health treatments (Hirsh et al. 2014).

Observer pain assessments are influenced by both patient characteristics and evaluator appraisals that may reflect bias. When no visible pathology was present, patients were more likely referred to mental health services, a tendency mediated by perceived pain genuineness (Bernardes et al. 2013). Combined with prior findings (Lord et al. 2009; Miller et al. 2018; Stålnacke et al. 2015; Markowitz 2022), this suggests that pain judged as psychological/emotional in origin (mediator) may in turn be viewed as less credible (potential second mediator) and thus less impactful. Gender role beliefs may further shape the association between patient gender and provider beliefs.

Building on this foundation, this study examined whether the relationship between patient gender and two key pain outcomes, perceived pain severity and pain interference, was serially mediated by the observer's perceived emotional/psychological causes and pain genuineness. We hypothesised that the female patient would be perceived as having more emotional/psychological contributions to her pain and lower pain genuineness relative to the male patient. In turn, these perceptions were expected to be associated with lower ratings of pain severity and interference. We also hypothesised that higher traditional gender role beliefs would moderate these mediation effects, such that observers endorsing more traditional gender role beliefs would show stronger gender‐based differences in pain evaluations. This study focuses on gender rather than biological sex, emphasising perceptions of social identity. We primarily use ‘man’ and ‘woman’ for gender identity, but follow the American Psychological Association's (APA) guidelines (American Psychological Association) by using ‘male’ and ‘female’ as adjectives.

2. Method

2.1. Participants

140 individuals were recruited from CloudResearch's Connect. Connect is an online participant recruitment platform that incorporates data quality controls intended to reduce to prevent fraudulent responses and improve reliability (Hartman et al. 2023). This sample size aligns with recommended ranges for mediation analyses, as simulation studies suggest this is within the range of participants sufficient for detecting medium indirect effects (Fritz and Mackinnon 2007). Participants from the US and those 18 or older were recruited. Additionally, the platform ensured an even number of men and women participated. All participants provided informed consent for the study as required by the IRB.

2.2. Measures

2.2.1. Gender Role Beliefs

Gender role beliefs were assessed with the 10‐item Gender Role Beliefs Scale (GRBS) (Brown and Gladstone 2012). Questions assess beliefs about men's and women's roles in the workforce and household (e.g., ‘Women should be concerned with their duties of childbearing and house‐tending, rather than with the desires for professional and business careers.’) Additionally, the scale assessed beliefs about chivalry and the need to protect women (e.g., ‘Men should continue to show courtesies to women such as holding open the door or helping them on with their coats.’) Participants responded on a 7‐point Likert scale, 1 = strongly agree and 7 = strongly disagree. Items were reverse‐scored and summed for a total score that ranged between 7 and 70, with higher scores indicating greater traditional gender role beliefs. The measure had good reliability (α = 0.83) in the present sample. The GRBS assesses general beliefs about men's and women's roles in society rather than beliefs about men's and women's pain.

2.2.2. Observer Pain Assessment

Participants assessed the patient's pain severity using the 0–10 numeric rating scale in which 0 = no pain and 10 = worst pain imaginable (Jensen and Karoly 2011). Pain interference (‘The patient's pain interferes with important activities in their life.’), pain genuineness (‘The patient's pain was genuine.’), and emotional/psychological factors (‘The patient's pain was caused by emotional or psychological factors.’) contributing to the patient's pain were rated using a Likert scale with 1 = strongly disagree and 7 = strongly agree.

2.3. Stimuli

We selected one video of a man and one video of a woman from a larger video database of individuals with chronic low back pain who were recorded performing standardised functional tasks. The man and the woman had self‐reported their gender and had conventional Western gender presentation cues (e.g., hairstyle and clothing) to ensure that patient gender was readily identifiable to observers. The people in the videos had experienced persistent low back pain for at least 3 months, with pain occurring on at least half of the days over the past 6 months and no history of low back surgery in the previous year. Details of individuals in this database and video‐recorded functional tasks can be found elsewhere (Kissi et al. 2024). Videos were full length views of patients from the front while they sat down in a chair from a standing position, then standing back up from the chair, a behavioural task that has been shown to elicit pain and pain expression (De Ruddere et al. 2013, 2014) and has successfully been used in prior work examining observer bias in pain (Miller et al. 2018). Trained coders evaluated videotaped movements and coded the occurrence of pain behaviours per second using the Pain Behaviour Coding Manual (Sullivan et al. 2000) and the Facial Action Coding System (Ekman et al. 2002) with scoring procedures detailed elsewhere (Kissi et al. 2024). Both patients in the videos were White and matched on scored level of pain expression. In addition to pain behaviour, a previous unpublished pilot sample of 100 healthy laypersons was used to rate the perceived age, weight category (normal‐weight, overweight, obese), and attractiveness of each patient in the video database, using the procedures outlined by Miller et al. (2018). These ratings were used to match the selected patient videos used for the current study on age, weight, and perceived attractiveness. Relative to the full pool of candidate patient videos, the selected male and female patients differed by 0.15 standard deviations or less on ratings for these characteristics.

2.4. Procedure

Using simple randomisation, participants read a brief vignette about a patient. In addition to the patients in the videos being male‐presenting and female‐presenting, the vignette disclosed a gendered title and pronouns that indicated the target's gender identity. The vignette read: ‘Mr./Mrs. Smith reports significant low back pain from a job‐related injury 18 months ago, but the pain has continued. He/She reports that the pain interferes with work as well as daily activities, such as shopping, cooking, or completing household chores.’ While an initial physical injury was noted, no additional biomedical diagnostic information was provided, allowing participants to interpret the persistence of the pain and its causes. Following the vignette, participants watched the video of the patient whose gender matched that of the vignette that was assigned. Participants then provided assessments about the patient's pain (pain severity, pain interference, pain genuineness, emotional/psychological causes) and completed the GRBS. Embedded in the questionnaire were two attention check items; one item asked the gender of the patient, and the other item asked participants to select 3 for the answer. Participants also responded to demographic questions, including their racial/ethnic identity (‘What is your race? (check all that apply) White, Black or African American, Native American or Alaska Native, Asian American or Pacific Islander, other)’ and gender identity (‘What is your gender? male, female, transgender male, transgender female, gender variant/nonconforming, other)’ consistent with the APA's guidelines (American Psychological Association 2020).

2.5. Data Analysis

Data were analysed using SPSS, version 30.0 (IBM, Chicago, IL). Pearson correlations were used to evaluate simple linear relationships between age of participant evaluators, GRBS scores, and observer pain assessment measures, and independent samples t‐tests were used to determine relationships between patient gender and observer pain assessment measures. Given the small number of individuals who identified as Hispanic or of another racialised identity, we collapsed across racial‐ethnic minority groups (Hispanic, Black/African American, Asian American, Indigenous American/Alaska Native, Other) and used a dichotomous variable of racial‐ethnic minority/non‐minority as a covariate in the analyses.

To test for mediation effects, the general steps to establish mediation outlined by Baron and Kenny (1986) were followed, but the PROCESS macro add‐on for SPSS by Hayes (2018) was used to statistically test for mediation. Using PROCESS, mediation was determined by obtaining a 95% confidence interval based on a bootstrapping procedure with 10,000 resamples using a seed of 12,895. In mediation analyses, the bootstrapping method determines whether the indirect effect in a hypothesised mediation model is significantly different from zero and has become an established alternative to normal‐theory tests, such as the Sobel test, that assume normality of the indirect effects (Hayes 2018). If the general steps to mediation are fulfilled (e.g., the predictor is significantly associated with the outcome, the predictor is significantly associated with the proposed mediator(s), the proposed mediator is significantly associated with the outcome in the presence of the predictor, but not vice versa) and the bootstrapped confidence interval testing the significance of the indirect effect is significant, then mediation is established. Dividing the indirect effect by the bootstrapped SE provided the t‐statistic associated with coefficients of indirect paths in the model. The PROCESS macro was also used to test gender role beliefs as a moderator of (Hayes 2018) the mediation effects. When comparing slopes for conditional effects, values for the moderator are defined at the 16th, 50th, and 84th percentiles, which approximate the mean, and ±1 SD (Hayes 2018). The moderating effect on the predictor to the mediators was tested. An index of moderated mediation was used to test the significance of the moderated mediation, that is, the difference of the indirect effects across levels of need for cognition (Hayes 2018) with significant effects determined by bootstrapped confidence intervals.

3. Results

132 participants passed both checks and were thus retained for use in the analyses. Means and standard deviations of demographic, predictor, and outcome variables for the participants are shown in Table 1. There were 91 participants who were White (68.9%), 21 who were Black or African American (15.9%), 11 who were Asian American (8.3%), two who were Native American/Alaska Native (1.5%), one who was Hispanic (0.8%), four who identified as multiracial (3.0%), and two who did not identify their race or ethnic background (1.5%). There were 66 men (50.0%) and 66 women (50.0%) and no participants identifying as transgender or gender non‐conforming. Twenty‐five (18.9%) indicated a history of being treated for chronic pain in the past. Pain severity ratings were moderate overall (M = 4.07, SD = 2.01; Table 1) and were not concentrated at the lower bound of the scale with only one participant (0.8%) rating the patient's pain as 0, indicating nearly all observers perceived the patient as experiencing at least some degree of pain. There were no significant differences between this group and those without a chronic pain history on how they estimated pain severity (t(130) = −0.156, p = 0.876), pain interference (t(129) = 0.374, p = 0.729), emotional/psychological contributions to the pain (t(130) = 0.097, p = 0.923), and pain genuineness (t(130) = −1.209, p = 0.229), nor was there a difference in GRBS scores (t(129) = 0.075, p = 0.941). There was even randomisation across gender of the participant and the patient, such that 33 women and 33 men were randomised to evaluate the female and male patient, respectively.

TABLE 1.

Demographic characteristics and means and standard deviations of variables of interest.

Demographics n %
Men/Women 66/66 50.0/50.0
Racial Minority/NHW 39/91 29.5/68.9
n Range M SD
Age 132 18–73 34.96 9.34
Variables of Interest
GRBS total score 131 10–67 29.06 13.27
Pain severity 132 0–10 4.70 2.01
Pain interference 131 1–7 5.10 1.49
Emotional cause of pain 132 1–7 2.93 1.61
Pain genuineness 132 2–7 5.27 1.26
Open in a new tab

Abbreviations: GRBS, Gender Role Beliefs Scale; NWH, non‐Hispanic white.

Zero‐order correlations for continuously measured variables are included in Table 2. A positive correlation was found between GRBS scores and the degree to which the patient's pain was perceived to be due to emotional or psychological factors, such that greater traditional gender role beliefs about women were related to a greater tendency to attribute pain to psychological and emotional causes. Participant age was not associated with GRBS scores (r = 0.016, p = 0.859), suggesting that endorsement of traditional gender role beliefs in this sample was not related to participant age. Perceived pain genuineness was positively associated with pain interference and pain severity, while negatively associated with viewing the patient's pain to be in part due to emotional or psychological factors. Table 3 shows the mean differences in pain ratings made by participants between the male and female patients. Overall, participants judged the female patient to have less severe pain, less interference from that pain with activities, higher contributions of emotional or psychological factors contributing to the pain, and less pain genuineness relative to the male patient.

TABLE 2.

Zero order correlations for continuous study variables.

Gender role beliefs total Perceived pain severity Perceived pain interference Emotional psychological cause Perceived pain genuineness
1. Participant age 0.016 −0.043 −0.013 −0.077 −0.009
4. Gender Role Beliefs — −0.013 −0.158 0.228** −0.147
6. Pain severity — 0.603** 0.031 0.598**
7. Pain interference — −0.134 0.753**
8. Emotional cause of pain — −0.225**
Open in a new tab

Note: *p < 0.05; **p < 0.01.

TABLE 3.

Mean differences between pain ratings for male and female patients.

Male patient Female patient df t p Cohen's d
M (SD) M (SD)
Emotional factors 2.515 (1.428) 3.348 (1.678) 130 −3.073 < 0.001 −0.535
Pain genuineness 5.712 (0.890) 4.833 (1.410) 130 4.282 0.003 0.745
Pain interference 5.530 (1.153) 4.662 (1.670) 129 3.469 < 0.001 0.606
Pain severity 5.318 (1.647) 4.076 (2.165) 130 3.711 < 0.001 0.646
Open in a new tab

3.1. Serial Mediation of Patient Gender and Pain Interference

A moderated serial mediation model controlling for participant demographics (participant age, gender, and racial/ethnic identity) and perceived pain severity of the patient was used to explore the mediating role of attributing emotional/psychological contributors to the patient's pain and perceived pain genuineness on the relationship between patient gender and pain interference. Hayes' (2018) SPSS PROCESS Macro (model 6) was first used to obtain direct and indirect relationships in the serial mediation models without the moderator (gender role beliefs) and the path coefficients from this model are shown in Figure 1. The total effect of the model was not significant, b = −0.3283, t = −1.4981, p = 0.1367, and likely reflects that direct and indirect effects include opposite directions. The direct effect of gender of the patient on perceived pain interference was also not significant (b = −0.0169, t = −0.0897, p = 0.9287), suggesting full mediation. As shown in Table 4a, perceived emotional/psychological factors contributing to pain and perceived pain genuineness serially mediated the relationship between patient gender and perceived pain interference. Compared to participants assessing the male patient, participants assessing the female patient perceived pain to be more likely due to emotional/psychological factors, which in turn was associated with perceiving the pain as less genuine, which ultimately was associated with lower perceived pain interference.

FIGURE 1.

FIGURE 1

Open in a new tab

Path coefficients and associated standard errors for serial mediation models. Values associated with the model predicting pain interference in black, above path lines, and values associated with the model predicting pain severity in grey, underneath path lines. EF = perceived emotional or psychological factors contributing to pain; PG = perceived pain genuineness. *p < 0.05, **p < 0.01.

TABLE 4a.

Total, direct, and indirect effects of the serial mediation model for pain interference and pain severity.

Total effect Direct effect Path Indirect effect t 95% CI
Lower Upper
−0.3283 (p = 0.1367) −0.0169 (p = 0.9287) Gender → EF → PI 0.0074 0.117 −0.1157 0.1419
Gender → PG → PI −0.2300 −1.760 −0.5055 0.0200
Gender → EF → PG → PI −0.0888 −1.820 −0.2015 −0.0108
−1.2122 (p = 0.0005) −0.5654 (p = 0.0640) Gender → EF → PS 0.1890 1.800 0.0220 0.4239
Gender → PG → PS −0.7437 −3.329 −1.1964 −0.3307
Gender → EF → PG → PS −0.0920 −1.294 −0.2522 0.0274
Open in a new tab

3.2. Serial Mediation of Patient Gender and Pain Severity

Hayes' (2018) SPSS PROCESS Macro (model 6) was also used to obtain direct and indirect relationships in the serial mediation of the relationship between patient gender and perceived pain severity of the patient, controlling for participant demographics. Path coefficients from this model are also shown in Figure 1. The total effect of this model was significant, b = −1.2122, t = −3.5660, p = 0.0005. The direct effect of patient gender on perceived pain severity was not statistically significant (b = −0.5654, t = −1.8692, p = 0.0640). While the proposed mediators were not found to serially mediate the relationship between patient gender and perceived pain severity, there were two significant indirect paths that emerged (Table 4a). Belief that emotional and psychological factors contributed to the patient's pain mediated the relationship between patient gender and perceived pain severity for the patient. Similarly, perceived pain genuineness mediated the relationship between patient gender and participants' estimation of the patient's pain severity.

3.3. Gender Role Beliefs as a Moderator of Indirect and Direct Effects

Gender role beliefs were examined as a conditioning variable (moderator) to the direct and indirect effects using Model 85 in SPSS PROCESS Macro (Hayes 2018). Gender role beliefs did not moderate the serial mediation effect observed in the relationship between patient gender and perceived pain interference of the patient, nor in the significant indirect path effects observed in the model of the relationship between patient gender and perceived pain severity (Table 4b). However, it significantly moderated the indirect path from patient gender to pain interference through perceived pain genuineness (ΔR 2 = 0.0236, F(1,119) = 5.2186, p = 0.0241), such that increasing participants' levels of traditional gender role beliefs was associated with perceiving the female patient's pain as less genuine, while the perceived genuineness of the male patient's pain remained relatively stable across levels of gender role beliefs (Figure 2). Johnson‐Neyman analysis revealed that this effect was significant when GRBS exceeded a score of 31.42 (scale range 10–67).

TABLE 4b.

Conditional mediation effects using gender role beliefs as a moderator.

Index of moderated mediation (LLCI, ULCI) Path Levels of GRBS b SE 95% CI
Lower Upper
0.0000 (−0.0043, 0.0036) Gender → EF → PI Low 0.0097 0.0689 −0.1231 0.1693
Mid 0.0095 0.0634 −0.1108 0.1496
High 0.0092 0.0666 −0.1240 0.1547
−0.0217 (−0.0435, −0.0041) Gender → PG → PI Low 0.0702 0.1761 −0.2647 0.4322
Mid −0.2123 0.1381 −0.5053 0.0507
High −0.5026 0.2074 −0.9598 −0.1463
0.0002 (−0.0052, 0.0059) Gender → EF → PG → PI Low −0.0933 0.0636 −0.2440 0.0033
Mid −0.0913 0.0494 −0.2045 −0.0114
High −0.0892 0.0589 −0.2247 0.0035
0.0001 (−0.0130, 0.0103) Gender → EF → PS Low 0.1765 0.1279 −0.0038 0.4783
Mid 0.1782 0.1006 0.0144 0.4000
High 0.1800 0.1226 −0.0271 0.4418
−0.0106 (−0.0498, 0.0253) Gender → PG → PS Low −0.6381 0.3195 −1.2879 −0.0287
Mid −0.7775 0.2348 −1.2534 −0.3321
High −0.9169 0.3682 −1.6885 −0.2427
−0.0001 (−0.0060, 0.0065) Gender → EF → PG → PS Low −0.0743 0.0785 −0.2580 0.0487
Mid −0.0750 0.0699 −0.2356 0.0456
High −0.0757 0.0809 −0.2690 0.0525
Open in a new tab

Abbreviations: EF, emotional or psychological factors contributing to patient's pain; GRBS, gender role beliefs scale; PG, perceived pain genuineness; PI, perceived pain interference.

FIGURE 2.

FIGURE 2

Open in a new tab

Gender role beliefs as a moderator of the indirect relationship between patient gender and pain interference through perceived pain genuineness.

Gender role beliefs did not moderate the indirect pathways from patient gender to perceived pain severity via perceived emotional/psychological factors contributing to the patient's pain and/or perceived genuineness. It did however moderate the direct path between patient gender and pain severity (ΔR 2 = 0.0400, F(1,120) = 8.8699, p = 0.0035), such that holding greater traditional gender role beliefs was associated with increasing perceived pain severity for the female patient but not for the male patient (Figure 3). The Johnson‐Neyman analysis revealed that when GRBS scores were < 28.5, the effect of gender is significant. Specifically at low levels of traditional gender role beliefs, the female patient was perceived as experiencing less severe pain. With higher GRBS scores (~42), this effect reverses direction and the female patient is perceived as experiencing more severe pain but is not significant until GRBS scores are > 62.

FIGURE 3.

FIGURE 3

Open in a new tab

Gender role beliefs as a moderator of the direct relationship between patient gender and pain severity.

4. Discussion

This study tested whether patient gender affects observers' pain judgements, with effects mediated by attributions to emotional/psychological causes and pain genuineness, and moderated by the perceiver's traditional gender role beliefs. Results generally supported the hypothesis that women's pain is judged differently than men's, consistent with prior findings of underestimation (Samulowitz et al. 2018; Chen et al. 2008; Patrick‐Smith and Bull 2024), and further suggest that this bias may operate through appraisal mechanisms in which women's pain is more likely to be attributed to emotional or psychological factors and perceived as less genuine. These findings support evidence of gender bias in pain evaluation (Lord et al. 2009; Samulowitz et al. 2018; Markowitz 2022) and extend it by identifying a potential pathway through which bias may operate. Using a serial mediation model, the female patient was rated as experiencing less pain interference, an effect transmitted indirectly through greater attribution to emotional/psychological causes and lower perceived genuineness. Lower genuineness was, in turn, linked to reduced interference ratings. This pathway remained significant after controlling for perceived pain severity and observer demographics, suggesting that gender influences pain interference judgements indirectly through appraisals of cause and authenticity. Thus, gender bias in pain evaluation may be more than overt dismissal and comprise various other cognitive appraisals by the observer, of which they may or may not be aware or are entirely automatic.

Although gender role beliefs did not moderate the serial indirect effect, it significantly moderated the indirect path from patient gender to pain interference via perceived pain genuineness. Observers with more traditional gender role beliefs rated the female patient's pain as less genuine, while perceptions of the male patient's pain remained stable regardless of gender role beliefs. This suggests that traditional gender role beliefs may strengthen the bias against female patients by discounting pain credibility, which in turn is associated with lower perceived interference, consistent with prior research (Samulowitz et al. 2018; Prego‐Jimenez et al. 2022; Hoffmann and Tarzian 2001).

Serial mediation did not emerge for perceived pain severity. Rather, two indirect paths were observed, highlighting a contrast with the pain interference model. For the female patient, lower perceived pain genuineness was associated with less severity. Similarly, gender effects on perceived pain severity were transmitted through judgements of emotional/psychological contributions to the pain. Unlike perceived pain interference, perceived pain severity appears linked to either credibility or etiological appraisals rather than both. Although the present study was not designed to examine cognitive mechanisms underlying pain judgements, the different mediation patterns observed for pain severity and pain interference raise the possibility that observers rely on potentially distinct appraisal processes when evaluating the sensory experience versus the extent to which pain affects functioning. Future research that assesses how observers integrate information regarding pain credibility, emotional attributions, and functional consequences when forming judgements about others' pain could help elucidate these appraisal processes.

Gender role beliefs only moderated the direct path between patient gender and perceived pain severity, but contrary to expectations, the direction was reversed. Participants with less traditional gender role beliefs rated the female patient as experiencing less pain. This aligns with Zhang et al. (2021), who found women were perceived as having less pain than men even when controlling for gender stereotypes of the observers and suggests that biases may stem more from entrenched pain‐specific stereotypes rather than general gender role beliefs. For example, vocalisations and grimacing are often seen as typical expressions of pain for women, whereas stoicism or suppression of expression is more typically associated with men (Keogh and Boerner 2020). Although patient videos were matched on several factors, including pain expression, observers may have relied on gendered expectations of how pain should appear. If a patient's observed behaviour did not align with these expectations, such as less pain expression than expected for a woman, observers may have inferred lower pain severity. However, stronger traditional gender role beliefs were associated with higher estimated pain severity for the female patient. This may reflect broader gender stereotypes, such as the belief that women are more fragile and less able to endure physical hardship (Brown and Gladstone 2012; Kerr and Holden 1996). Broader gender assumptions, when present, may override pain‐specific gender expectations, leading observers who hold more traditional gender beliefs to infer that women experience more severe pain regardless of visible pain expression. Notably, gender role beliefs moderated only the direct effect of patient gender on perceived severity, suggesting traditional beliefs influence immediate, surface‐level judgements of severity but may also shape more nuanced appraisals, such as assessments of pain cause or legitimacy, when evaluating its interference with life activities.

An alternative explanation is that traditional gender role beliefs include stereotypes of women's fragility, leading observers to perceive fragile individuals as experiencing greater pain. Such beliefs are closely linked to benevolent sexism, an ideology encompassing beliefs that women are weak, vulnerable, and in need of protection (Bareket and Fiske 2023; Glick and Fiske 2001). Observers with more traditional gender role beliefs may therefore perceive women as fragile and in greater pain, whereas those with less traditional beliefs may resist stereotypes of feminine fragility, perceiving women as experiencing less pain. Indeed, research shows that benevolent sexism is linked to higher perceived femininity of White women, which in turn was associated with inferring women to experience more pain (Brown‐Iannuzzi et al. 2022). Relatedly, women high in benevolent sexism viewed menstruation as impairing women's abilities (e.g., work performance) and believed menstruating women should avoid certain tasks (e.g., carrying heavy items), reflecting paternalistic attitudes that women need special protections (Marván et al. 2014). These findings support the idea that benevolent sexism, through stereotypes of women's fragility, contributes to perceiving women as experiencing greater pain, potentially explaining why participants with more traditional gender role beliefs rated women's pain as higher in the present study.

4.1. Implications

Despite increased attention to equity and implicit bias in healthcare, our results suggest that credibility assessments may represent one process through which gender‐related differences in pain judgements can emerge. Discrepancies in estimating pain were observed even among participants with more egalitarian gender role beliefs, highlighting the potential importance of pain‐specific gender attitudes and the need for further examining how credibility judgements influence pain assessment in clinical contexts. Prior work suggests that using counter‐stereotypical examples or encouraging reflection on credibility heuristics may help mitigate these effects. For instance, Ruben and Stosic (2024) found that feedback on patients' actual self‐reported pain intensity reduced bias more effectively than raising general awareness of bias. Future research should examine whether encouraging providers to actively reflect on assumptions and compare them with patients' reports can resolve gender differences in pain assessments in clinical settings.

4.2. Limitations and Future Directions

Several limitations should be noted. First, the cross‐sectional design means that mediation results reflect indirect associative effects, not causal relationships. Second, the sample consisted of general community participants without specific recruitment of healthcare professionals, which may limit generalisability to clinical settings. Future studies should examine whether similar moderated mediation patterns emerge among clinicians who routinely assess and manage pain. Third, vignette‐ and video‐based designs reduce ecological validity and may not fully capture the complexity of real‐world clinical encounters, highlighting the need for contemporaneous, in‐person studies of patient‐provider interactions. Relatedly, the present study focused on observer perceptions rather than accuracy relative to the patient's self report. Future work should therefore examine the degree to which observer judgements correspond with patients' own reports of pain. Fourth, self‐report measures of gender role beliefs may not fully capture implicit biases that influence real‐time judgements about pain. Incorporating measures of implicit bias as well as pain‐specific gender stereotypes may better elucidate factors contributing to pain judgements.

Although patient race was held constant to focus on gender effects, this does not imply that race is unimportant. Prior work suggests that race and gender can interact, with women of colour often perceived as experiencing less pain than White women or men of colour (Ruben and Stosic 2024). More broadly, pain judgements are likely shaped by multiple intersecting social identities and patient characteristics. While the selected video stimuli were matched on age, body size, attractiveness, and observable pain behaviours, future research should more directly examine how combinations of gender, race/ethnicity, age, body size, and other social identities influence perceptions of pain. In addition, the present study focused on binary gender categories. Both the patient stimuli and participant measures were structured around male–female comparisons, which do not capture the experiences of individuals with diverse gender identities. Expanding this work to include nonbinary and transgender individuals would provide greater understanding of how gender influences pain evaluation. The findings should also be interpreted within the sociocultural context in which the study was conducted. Gender role beliefs and expectations surrounding pain expression are shaped by cultural norms, many of which reflect Western perspectives on gender and emotional expression. As a result, the present findings may not generalise to cultures with different gender norms. Future research should examine how attributional pathways in pain judgements operate across cultural contexts.

Finally, another contextual factor that may influence observer judgements is the presence or clarity of biomedical evidence. In the present vignette, the patient's pain was described as originating from a job‐related injury, but no additional diagnostic information was provided. Observers may therefore have relied more heavily on assumptions about the causes of pain, which may interact with gender stereotypes. It remains unclear whether the availability of clearer biomedical evidence would attenuate gender differences in pain judgements.

5. Conclusions

This study extends prior research on gender bias in pain evaluation. Patient gender was associated with perceived pain interference indirectly through attributions to emotional or psychological causes and perceptions of pain genuineness. Similarly, patient gender was linked to perceived pain severity through two distinct pathways: through perceived emotional/psychological contributors to pain and through perceived pain genuineness. More traditional gender role beliefs were associated with greater attribution of a female patient's pain to emotional or psychological causes and moderated the relationship between patient gender and both pain genuineness and pain severity. These findings suggest a complex, layered nature of gender bias in pain assessment. Because the data are cross‐sectional and based on a community sample, future research with clinicians and with designs that allow causal inference is needed to determine whether these patterns of association influence pain assessment and treatment decisions in clinical contexts.

Author Contributions

This study was designed by E.J.R., M.S., and Z.T. Data collection was performed by E.J.R. and M.S., and data were analysed by E.J.R. All authors (E.J.R., M.S., Z.T., and S.R.M.) critically examined the results and participated in preparation and editing of the manuscript. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. American Psychological Association . 2020. Publication Manual of the American Psychological Association. 7th ed. American Psychological Association. [Google Scholar]
  2. Bareket, O. , and Fiske S. T.. 2023. “A Systematic Review of the Ambivalent Sexism Literature: Hostile Sexism Protects Men's Power; Benevolent Sexism Guards Traditional Gender Roles.” Psychological Bulletin 149, no. 11–12: 637–698. 10.1037/bul0000400. [DOI] [PubMed] [Google Scholar]
  3. Baron, R. M. , and Kenny D. A.. 1986. “The Moderator‐Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations.” Journal of Personality and Social Psychology 51: 1173–1182. [DOI] [PubMed] [Google Scholar]
  4. Beere, C. A. , King D. W., Beere D. B., and King L. A.. 1984. “The Sex‐Role Egalitarianism Scale: A Measure of Attitudes Toward Equality Between the Sexes.” Sex Roles 10, no. 7/8: 563–576. 10.1007/BF00287265. [DOI] [Google Scholar]
  5. Bernardes, S. F. , Costa M., and Carvalho H.. 2013. “Engendering Pain Management Practices: The Role of Physician Sex on Chronic Low‐Back Pain Assessment and Treatment Prescriptions.” Journal of Pain 14, no. 9: 931–940. 10.1016/j.jpain.2013.03.004. [DOI] [PubMed] [Google Scholar]
  6. Brown, M. J. , and Gladstone N.. 2012. “Development of a Short Version of the Gender Role Beliefs Scale.” International Journal of Psychology and Behavioral Sciences 2, no. 5: 154–158. 10.5923/j.ijpbs.20120205.05. [DOI] [Google Scholar]
  7. Brown‐Iannuzzi, J. L. , Cooley E., Cipolli W., and Mehta S.. 2022. “Race, Ambivalent Sexism, and Perceptions of Situations When Police Shoot Black Women.” Social Psychological and Personality Science 13, no. 1: 127–138. 10.1177/1948550620987659. [DOI] [Google Scholar]
  8. Chen, E. H. , Shofer F. S., Dean A. J., et al. 2008. “Gender Disparity in Analgesic Treatment of Emergency Department Patients With Acute Abdominal Pain.” Academic Emergency Medicine 15, no. 5: 414–418. [DOI] [PubMed] [Google Scholar]
  9. Cleeland, C. S. , Gonin R., Hatfield A. K., et al. 1994. “Pain and Its Treatment in Outpatients With Metastatic Cancer.” New England Journal of Medicine 330, no. 9: 592–596. 10.1056/NEJM199403033300902. [DOI] [PubMed] [Google Scholar]
  10. Clerc Liaudat, C. , Vaucher P., De Francesco T., et al. 2018. “Sex/Gender bias in the Management of Chest Pain in Ambulatory Care.” Women's Health 14: 1745506518805641. 10.1177/1745506518805641. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. De Ruddere, L. , Goubert L., Stevens M., de C Williams A. C., and Crombez G.. 2013. “Discounting Pain in the Absence of Medical Evidence Is Explained by Negative Evaluation of the Patient.” Pain 154, no. 5: 669–676. 10.1016/j.pain.2012.12.018. [DOI] [PubMed] [Google Scholar]
  12. De Ruddere, L. , Goubert L., Stevens M. A., Deveugele M., Craig K. D., and Crombez G.. 2014. “Health Care Professionals' Reactions to Patient Pain: Impact of Knowledge About Medical Evidence and Psychosocial Influences.” Journal of Pain 15, no. 3: 262–270. 10.1016/j.jpain.2013.11.002. [DOI] [PubMed] [Google Scholar]
  13. Ekman, P. , Friesen W. V., and Hager J. C.. 2002. Facial Action Coding System. Manual and Investigator's Guide. Research Nexus. [Google Scholar]
  14. Fritz, M. S. , and Mackinnon D. P.. 2007. “Required Sample Size to Detect the Mediated Effect.” Psychological Science 18, no. 3: 233–239. 10.1111/j.1467-9280.2007.01882.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Glick, P. , and Fiske S. T.. 2001. “An Ambivalent Alliance: Hostile and Benevolent Sexism as Complementary Justifications for Gender Inequality.” American Psychologist 56, no. 2: 109–118. 10.1037/0003-066X.56.2.109. [DOI] [PubMed] [Google Scholar]
  16. Guzikevits, M. , Gordon‐Hecker T., Rekhtman D., et al. 2024. “Sex Bias in Pain Management Decisions.” Proceedings of the National Academy of Sciences of the United States of America 121, no. 33: e2401331121. 10.1073/pnas.2401331121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hadjistavropoulos, T. , and Craig K. D.. 2002. “A Theoretical Framework for Understanding Self‐Report and Observational Measures of Pain: A Communications Model.” Behaviour Research and Therapy 40, no. 5: 551–570. 10.1016/s0005-7967(01)00072-9. [DOI] [PubMed] [Google Scholar]
  18. Hartman, R. , Moss A. J., Jaffe S. N., Rosenzeig C., Robinson J., and Litman L.. 2023. “Introducing Connect by CloudResearch: Advancing Online Participant Recruitment in the Digital Age.” PsyArXiV. 10.31234/osf.io/ksgyr. [DOI]
  19. Hayes, A. 2018. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression‐Based Approach. 2nd ed. Guilford. [Google Scholar]
  20. Hirsh, A. T. , Hollingshead N. A., Matthias M. S., Bair M. J., and Kroenke K.. 2014. “The Influence of Patient Sex, Provider Sex, and Sexist Attitudes on Pain Treatment Decisions.” Journal of Pain 15, no. 5: 551–559. [DOI] [PubMed] [Google Scholar]
  21. Hoffmann, D. E. , and Tarzian A. J.. 2001. “The Girl Who Cried Pain: A bias Against Women in the Treatment of Pain.” Journal of Law, Medicine & Ethics 29, no. 1: 13–27. 10.1111/j.1748-720x.2001.tb00037.x. [DOI] [PubMed] [Google Scholar]
  22. Jensen, M. P. , and Karoly P.. 2011. “Self‐Report Scales and Procedures for Assessing Pain in Adults.” In Handbook of Pain Assessment, edited by Turk D. C. and Melzack R., 3rd ed., 15–34. Guilford. [Google Scholar]
  23. Keogh, E. , and Boerner K. E.. 2020. “Exploring the Relationship Between Male Norm Beliefs, Pain‐Related Beliefs and Behaviours: An Online Questionnaire Study.” European Journal of Pain 24, no. 2: 423–434. 10.1002/ejp.1499. [DOI] [PubMed] [Google Scholar]
  24. Kerr, P. S. , and Holden R. R.. 1996. “Development of the Gender Role Beliefs Scale (GRBS).” Journal of Social Behavior and Personality 11, no. 5: 3–16. [Google Scholar]
  25. Kissi, A. , Vorensky M., Sturgeon J. A., et al. 2024. “Racial Differences in Movement‐Related Appraisals and Pain Behaviors Among Adults With Chronic Low Back Pain.” Journal of Pain 25, no. 5: 104438. 10.1016/j.jpain.2023.11.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lloyd, E. P. , Paganini G. A., and ten Brinke L.. 2020. “Gender Stereotypes Explain Disparities in Pain Care and Inform Equitable Policies.” Policy Insights From the Behavioral and Brain Sciences 7, no. 2: 198–204. 10.1177/2372732220942894. [DOI] [Google Scholar]
  27. Lord, B. , Cui J., and Kelly A. M.. 2009. “The Impact of Patient Sex on Paramedic Pain Management in the Prehospital Setting.” American Journal of Emergency Medicine 27, no. 5: 525–529. 10.1016/j.ajem.2008.04.003. [DOI] [PubMed] [Google Scholar]
  28. Markowitz, D. M. 2022. “Gender and Ethnicity bias in Medicine: A Text Analysis of 1.8 Million Critical Care Records.” PNAS Nexus 1, no. 4: pgac157. 10.1093/pnasnexus/pgac157. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Marván, M. L. , Vázquez‐Toboada R., and Chrisler J. C.. 2014. “Ambivalent Sexism, Attitudes Towards Menstruation and Menstrual Cycle‐Related Symptoms.” International Journal of Psychology 49, no. 4: 280–287. 10.1002/ijop.12028. [DOI] [PubMed] [Google Scholar]
  30. Miller, M. M. , Allison A., Trost Z., et al. 2018. “Differential Effect of Patient Weight on Pain‐Related Judgements About Male and Female Chronic Low Back Pain Patients.” Journal of Pain 19, no. 1: 57–66. 10.1016/j.jpain.2017.09.001. [DOI] [PubMed] [Google Scholar]
  31. Paganini, G. A. , Summers K. M., Ten Brinke L., and Lloyd E. P.. 2023. “Women Exaggerate, Men Downplay: Gendered Endorsement of Emotional Dramatization Stereotypes Contributes to Gender bias in Pain Expectations.” Journal of Experimental Social Psychology 109: 104520. 10.1016/j.jesp.2023.104520. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Patrick‐Smith, M. , and Bull S.. 2024. “Medical Student Perceptions of Gender and Pain: A Systematic Review of the Literature.” BMC Medicine 22, no. 1: 434. 10.1186/s12916-024-03660-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Prego‐Jimenez, S. , Pereda‐Pereda E., Perez‐Tejada J., Aliri J., Goñi‐Balentziaga O., and Labaka A.. 2022. “The Impact of Sexism and Gender Stereotypes on the Legitimization of Women's Low Back Pain.” Pain Management Nursing 23, no. 5: 591–595. 10.1016/j.pmn.2022.03.008. [DOI] [PubMed] [Google Scholar]
  34. Ruben, M. A. , and Stosic M. D.. 2024. “Documenting Race and Gender Biases in Pain Assessment and a Novel Intervention Designed to Reduce Biases.” Journal of Pain 25, no. 9: 104550. 10.1016/j.jpain.2024.104550. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Samulowitz, A. , Gremyr I., Eriksson E., and Hensing G.. 2018. “‘Brave Men’ and ‘Emotional Women’: A Theory‐Guided Literature Review on Gender bias in Health Care and Gendered Norms Towards Patients With Chronic Pain.” Pain Research & Management 2018: 6358624. 10.1155/2018/6358624. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Stålnacke, B. M. , Haukenes I., Lehti A., Wiklund A. F., Wiklund M., and Hammarström A.. 2015. “Is There a Gender bias in Recommendations for Further Rehabilitation in Primary Care of Patients With Chronic Pain After an Interdisciplinary Team Assessment?” Journal of Rehabilitation Medicine 47, no. 4: 365–371. 10.2340/16501977-1936. [DOI] [PubMed] [Google Scholar]
  37. Sullivan, M. J. L. , Tripp D. A., and Santor D.. 2000. “Gender Differences in Pain and Pain Behavior: The Role of Catastrophizing.” Cognitive Therapy and Research 24: 121–134. 10.1023/A:1005459110063. [DOI] [Google Scholar]
  38. Tait, R. C. , Chibnall J. T., and Kalauokalani D.. 2009. “Provider Judgments of Patients in Pain: Seeking Symptom Certainty.” Pain Medicine 10, no. 1: 11–34. 10.1111/j.1526-4637.2008.00527.x. [DOI] [PubMed] [Google Scholar]
  39. Zhang, L. , Losin E. A. R., Ashar Y. K., Koban L., and Wager T. D.. 2021. “Gender Biases in Estimation of Others' Pain.” Journal of Pain 22, no. 9: 1048–1059. 10.1016/j.jpain.2021.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from European Journal of Pain (London, England) are provided here courtesy of Wiley

RESOURCES