The Effect of Gender Ideal Adherence on Carer Burden and Help-Seeking in Male Carers of People With Multiple Sclerosis

Among chronic diseases, multiple sclerosis (MS) is the leading cause of disability in young and middle-aged adults and markedly affects day-to-day life.¹  This imposes a heavy burden on families of people with MS who must adapt to the impact of the disease on daily living and future plans.² 

Two factors make the study of informal carers of people with MS, especially spouses or partners, unique. First, they are commonly placed in caregiving roles at disease onset at a time when couples have to manage the competing demands of relationships, children, and careers. Second is the striking 3:1 female-to-male gender ratio in MS, a unique phenomenon among chronic neurological conditions, making it more common to find males in caring roles.³ 

There are well-established links between being an informal carer and physical and mental health.⁴  A meta-analysis of 24 studies that included 2097 carers revealed that being an informal carer of a person with MS adversely affects social involvement, financial state, and physical and psychological health.⁵  These effects can be summarized into the concept of carer burden.

Carer burden refers to the allostatic (comprehensive, physical, and emotional) load an informal carer is under due to their role.^6-8  The stress process model from Pearlin et al⁹  informs much of the research into levels of carer burden, how informal carers cope, and the impact caring has on physical and mental health.

Research on carers for people with neurological conditions suggests that female informal carers exhibit higher levels of burden and poorer quality of life and rely more on emotional rather than practical coping styles in comparison with male informal carers (MICs). However, this characterization has been criticized, as MICs are largely underrepresented in the literature.^10-13 

From the few studies to date examining gender differences among caregivers of people with MS, findings have been equivocal. Data from some studies suggest that caregiver gender has no impact on distress.^14,15  Although Buchanan et al report greater burden among male carers,¹⁶  Knight et al found that wives exhibited higher burden scores than husbands.¹⁷  Indeed, results from a large-scale study of 1333 carers, of whom two-thirds were male, found that female carers of people with MS reported greater levels of stress and burden and higher levels of medication use for psychological concerns. Male carers reported more physical concerns, with the authors hypothesizing that males somatize their psychological distress.¹⁸ 

MICs appear to have poorer outcomes in accessing both formal and informal social support (help-seeking) than their female counterparts.¹⁹  A sample of 106 informal carers of people with MS reported a significant difference in need for social support, with women identifying a greater need,²⁰  possibly attributable to previous negative experiences in requesting support and insufficient information regarding service availability. MICs also reported guilt and ambivalence about requesting help, which may have been caused by their expressed commitment to the duty of caring.¹⁹ 

Drawing from sociological literature, gender socialization theory may offer useful perspectives for explaining this ineffective coping strategy. Research on gender theory has developed our understanding of culturally specific masculine and feminine ideals. These concepts are now widely recognized and could have clinical implications when applied to assist in understanding the behavior and attitudes of MICs.

Traditional masculinity still falls within certain stereotyped characteristics,²¹  namely: Do not be feminine, be respected for achievements, do not show weakness, and seek adventure and risk. It has been argued that this ideology of masculinity produces “predictable and unfortunate results,”²²  such as an emphasis on dominance, extreme self-reliance, and restricted emotionality. If operating within contemporary lifestyles, these “ideal” masculine traits may produce conflict in heterosexual couples as women work full-time and continue to hold the burden of care for children and housework.²³  Yet simultaneously, men are under pressure to conform to modern values of nurturing children, expressing their emotions, sharing housework, and curbing aggression and risk-taking. The tension between traditional masculine ideals and modern, progressive values gives rise to what researchers in the field have termed gender role conflict (GRC).

GRC can be conceptualized as a form of stress, akin to cognitive dissonance, where there is a tension among one's behavior, expectations (from oneself and others) about how one should behave, and a fear of not “being a man.”²⁴  To mitigate this stress, it is argued that men with higher levels of GRC adopt maladaptive coping styles.²⁵  Carver and Connor-Smith characterize this as disengagement-focused coping that includes avoidant behaviors and attitudes,²⁶  such as substance use,²⁷  denial of distress, and resistance to help-seeking.²⁸  In addition to enforcing these avoidant strategies, GRC appears to negatively affect well-being. A review of studies by O’Neil revealed positive correlations between GRC and both anxiety and depression.²⁹ 

However, not all men adhere so closely to these stereotyped masculine ideals, or they appear to embrace them in some contexts and reject them in others.³⁰  This suggests some men display flexibility toward their gender construct.

Understanding gender role flexibility (GRF) originates from work by Bem and her concept of androgyny, which challenges bipolar concepts of gender because individuals can express both masculine and feminine behaviors depending on the situational appropriateness of the behavior.³¹  Bem et al claimed this state of androgyny is psychologically more adaptive as it permits greater psychological adjustment.^31,32 

GRF has been proposed as a cognitive asset that fosters successful coping with the changing conditions of daily life. Martin et al suggest that GRF may reflect a thinking process that enables androgynous individuals to detect situational characteristics, facilitating the adoption of diverse coping strategies to handle different situational demands.³³  This idea of coping flexibility fits with Lazarus and Folkman's model of adaptation goodness of fit, which is being able to adjust one's coping resources to meet situational demands.³⁴ 

The current study was designed to investigate the relative influence of gender constructs on MIC attitudes toward help-seeking and carer burden in contrast to other variables known to contribute to carer burden. It was hypothesized that attitudes toward help-seeking and carer burden are affected by gender role constructs, specifically that GRC and GRF contribute to significant variance in attitudes of help-seeking and carer burden in comparison with other known predictors, including demographic variables and level of current social support.

We used a cross-sectional descriptive design, gathering data from both paper-based and online surveys, to examine the relative influence of GRC and GRF on MIC attitudes toward help-seeking and perceived social support. This was in relation to dependent variables known to predict greater social support, attitudes, and behavior such as demographic factors and carer burden. Prior to performing regression analyses, Pearson correlation coefficient 2-tailed tests were computed between all independent variables, with the critical a level set at 0.01 to control for type 1 error. Additionally, primary component analysis was conducted on demographic factors to reduce the number of independent variables in regression models. Variance in help-seeking behavior and current support from GRC and GRF were analyzed through a multiple regression analysis.

The study sample consisted of an opportunistic sample of adult MICs of people with MS. Participants were recruited through specialist MS nurses in the United Kingdom, the Channel Islands, and the Isle of Man and via MS-related Facebook groups, X (formerly known as Twitter), and MS-related online forums and websites. Paid Facebook and X advertisements were used in a limited and focused way to increase recruitment and allow access to a wider pool of international participants. Participation was on a volunteer basis. The current study was approved by the University of Leicester Ethics Subcommittee for Psychology, National Health Service Health Research Authority, Guernsey Ethics Committee, Jersey Ethics Committee, and Isle of Man Research Ethics Committee.

Demographic details elicited were age, time spent caring, time in relationship, hours spent working, and number and ages of children. Additionally, number of hours of professional care received for participants’ own physical and psychological needs, as well as support from health professionals, was reported. These represent secondary measures of help-seeking and social support, differentiating between support for the MIC and the patient.

The Gender Role Conflict Scale-Short Form (GRCS-SF)³⁵  assessed gender role conflict. It consists of 16 items across 4 subscales: Restricted Emotionality (RE); Success, Power, and Competition (SPC); Restrictive Affectionate Behavior Between Men (RABBM); and Conflicts Between Work and Family Relations (CBWFR). Responses were measured on a 6-point Likert scale (1 =“strongly disagree” to 6 = “strongly agree”), with a possible score range of 16 to 96 overall and 4 to 24 for each subscale; higher scores indicate greater conflict (a=.77 for the RE and CBWFR; a=.78 for the RABBM; a=.80 for the SPC).³⁵ 

The brief version of the Zarit Burden Interview (B-ZBI)^6,36  measured caregiver strain, one of the primary dependent variables in the regression analyses. It consists of 12 items addressing physical health, psychological well-being, finances, and social life and is measured on a 5-point Likert scale (0 = “never” to 4 = “nearly always”), with a possible score range of 0 to 48; higher scores indicate higher perceived burden (a = .92).³⁷  The B-ZBI demonstrates good correlations with the long ZBI (0.92-0.97).³⁶ 

The 8-item modified Medical Outcomes Study Social Support Survey (mMOS-SS)³⁸  assessed social support. Two subscales (emotional and instrumental [tangible] social support), composed of 4 items each, maintain the theoretical structure of the MOS-SS and identify potentially modifiable social support deficits (a =.93).³⁸  Possible scores range from 4 to 20 for each subscale.

The Willingness to Seek Help Questionnaire (WSHQ)³⁹  assessed participants’ willingness to seek professional help, another key dependent variable in the regression analyses. This 25-item questionnaire is rated on a 4-point Likert scale (0 = “Do not identify with the statement at all” to 3 = “Identify completely with statement”), with a possible score range of 0 to 75; higher scores indicate greater willingness to seek help (a =.85).³⁹ 

The short-form Personality Attributes Questionnaire (PAQ)⁴⁰  assessed how people perceive themselves regarding traditional masculine/instrumental and feminine/expressive traits. Responses are recorded on a 5-point Likert scale for 8 items on each subscale, with a possible score range of 0 to 32 for each subscale. Factor analysis supports the validity of the masculine/instrumental and feminine/expressive subscales^41,42  across socioeconomic and age groups.⁴⁰ 

Specialist MS nurses identified possible participants and invited them to participate in the study, providing them with an information sheet and paper version of the survey, which was returned to the researcher via post. All participants were able to access the survey in an online version powered by SurveyMonkey (Symphony Technology Group) via clicking a link advertised on various internet platforms. All data were anonymized.

An a priori power calculation was undertaken using IBM's SPSS, version 25, to determine the number of participants required to undertake the proposed hierarchical regression. The power calculation was made to detect at least a medium effect at P = .05. This produced a minimum sample size of N = 62. A hierarchical regression was undertaken, as previous literature has shown significant relationships among social support factors, demographics, and carer burden. Hierarchical regression analysis allowed for an examination of the relationships between carer burden and known predictors alongside an exploration of the links between gender constructs, burden, and MIC help-seeking.

An international sample of 68 participants was recruited (see TABLE S1 for participants’ countries of residence). Fifty-nine completed the survey online, and 9 completed it on paper. Participants were aged 18 to 87 years (mean [SD], 52.47 years [13.32]). TABLE 1 summarizes the participants’ characteristics. Although there was variation in the amount of hours participants worked, the majority were not working full-time (< 40 hours per week). Respondents reported low levels of professional and informal psychological support for themselves, with a mean (SD) of 2.4 minutes per week (0.16). The majority of respondents reported no psychological support.

TABLE 2 summarizes the means from the self-report questionnaire measures. B-ZBI scores indicated a mean (SD) burden score of 17.72 (9.46). mMOS-SS scores indicated a mean social support score of 21.61 (7.90). The mean level of perceived help-seeking was 39.74 (7.37). Participants indicated a mean GRC score of 50.43 (12.27); of the GRC subscales, the highest scores came from RE (mean, 13.19 [4.62]) and CBWFR (mean, 13.46 [4.62]). In terms of gender flexibility, responses on the PAQ indicated similar scores on masculine traits (instrumentality; mean, 19.79 [5.36]) and feminine traits (expressivity; mean, 20.54 [5.37]).

Primary component analysis (PCA) resulted in 3 factors: relationship variables, work and children variables, and support-related variables. PCA was also conducted on gender variables but did not provide significantly fewer factors than the component subscales of GRC and the PAQ.

Data were screened for outliers using both Mahalanobis and Cook distances. Analysis indicated the need to remove 3 participants’ data. Tests of normality and homoscedasticity were met. As 2 separate regression analyses were undertaken, a Bonferroni correction was set at P <.025.

TABLE S2 displays the results of a hierarchical multiple regression analysis conducted to test the hypothesis that gender constructs are strongly associated with help-seeking attitudes. The variable mMOS-SS was entered into the first step, as this may be considered a predictor of help-seeking. Perception of social support was strongly linked to attitudes toward help-seeking (β = .23; P = .017). The addition of gender constructs did not significantly increase the regression's explanatory power; however, when the variables PAQ-Expressivity and GRCS-RABBM were added, the model still retained significance (F [3, 59] = 3.335; P = .025). Therefore, the hypothesis that gender constructs are strongly associated with attitudes toward help-seeking was not supported. This suggests that additional factors may be affecting MIC willingness to seek support.

TABLE S3 displays the results of a hierarchical multiple regression analysis conducted to test the hypothesis that gender constructs explain significant variance in carer burden, as measured by the B-ZBI. To be parsimonious with predictor variables, hours worked and number of dependent children were combined into the factor work-children, which was selected as a known predictor of burden in addition to the GRCS and PAQ subscales. Regression analysis results indicate that the final model explained 36.5% of the variance in burden (R² = 0.365; F [1, 44] = 3.033; P = .013). The effect size was large (F² = 0.57), although the adjusted R² was less (0.244), suggesting possible overfit in the final model. The predictors GRCS-RE, GRCS-CBWFR, and PAQ-Expressivity were significantly associated with carer burden (P < .05); however, only GRCS-RE was significant (P < .025) at the level set for the Bonferroni correction (β = 1.12; P = .002). Unexpectedly, the work-children factor was not significantly linked to burden. Therefore, the hypothesis that gender constructs significantly explain variance in carer burden was supported. This finding highlights a potential target for interventions aiming to attenuate burden among MICs.

The stress process model from Pearlin et al highlights how carers may ameliorate the effects of caring through coping styles and engaging social supports.⁹  Evidence indicates that men are reluctant to seek help, and it was hypothesized that the need to live up to cultural masculine ideals may be a barrier to help-seeking. Although gender constructs were not significantly associated with attitudes toward help-seeking, there is evidence for their importance in MIC reluctance to seek support for themselves.

GRC was not significantly associated with help-seeking in the current study, contradicting prior research.²⁹  MICs demonstrated a lower willingness to seek help for themselves when compared with the general population. Segal et al compared younger and older men on the WSHQ (means of 51.3 and 50.3, respectively), and both groups demonstrated more willingness to seek psychological help than the current sample (mean, 38.1).⁴³  Thus, MICs may have poorer attitudes toward help-seeking than the general population, in spite of their increased risk of developing mental health problems. Examination of the self-reported hours of personal psychological and informal support in the current sample indicated substantial differences from the general population. The average number of hours per week of psychological support that participants received was negligible. In addition to concerns about ability to access appropriate psychological support, there are other possible barriers to MICs engaging in psychological selfcare, including informal support.

Pederson and Vogel theorized that self-stigma mediates the relationship between GRC and attitudes toward self-help seeking.⁴⁴  Self-stigma is conceptualized as the negative effect that help-seeking has on one's self-concept; thus, avoiding stigma is prioritized above experiencing emotional suffering.⁴⁵  Within the context of caring, appearance of coping may be a mediating variable; MICs need to maintain an appearance of coping to avoid emotionally burdening the individual with MS and other family members. This has been clinically observed and reported in qualitative literature.⁴⁶  MICs are thus potentially cutting themselves off from informal support sources, especially their partners, through the gender-related belief that they need an appearance of coping and stoicism as well as the need to protect others from emotional distress.

MIC scores on the SPC subscale of the GRCS were weakly correlated with levels of support for the individual with MS. This may suggest that MICs who perceive difficulty in fulfilling masculine ideals of success are more likely to seek professional care for their partner. This finding supports the characterization of male carers who struggle with the gendered role of caregiving and seek help to offload these tasks.⁴⁷  As Hanlon argues,⁴⁸  men tend to approach caregiving in a task-oriented manner, considering the instrumental and practical aspects that can be outsourced. This finding can be understood by recognizing the context of this study and the tools used. Commonly used to measure attitudes toward help-seeking for oneself, the WSHQ may also be interpreted as seeking help for others in the context of caring.

GRF, as measured by the PAQ-Expressivity subscale, was not significantly associated with attitudes toward help-seeking. However, the expressivity subscale was weakly correlated to help-seeking attitudes, with similar findings reported in the general male population.⁴⁹  One may conceptualize the expressivity subscale as an indication of the ability to use skills and qualities that are traditionally considered feminine.⁵⁰  Cheng posits that this is a form of gender flexibility⁵⁰  and supports the knowing how³⁴  and goodness of fit⁵¹  hypotheses. These hypotheses refer to the ability to adapt one's behavior to situational needs. In this study, this may be considered as being behaviorally flexible and not tied to hegemonic gender roles. Therefore, it may be beneficial to consider gender flexibility as a component of coping flexibility.⁵²  Furthermore, one can conceptualize gender constructs as having a balance of effects on help-seeking, or hindering, and facilitating factors as described in Brown et al's model of carer coping (TABLE S4).⁵³ 

Gender constructs were significantly associated with burden and explained the variance in burden when compared with the previously observed factor of work and family. The CBWFR subscale of the GRCS was significantly associated with burden scores. This finding can be conceptualized as internalized gender conflict stresses adding to carer burden. MIC perceptions of caring demands are significantly influenced by the degree to which MICs feel they are not embodying hegemonic masculine ideals.

In the current study findings, the RE and CBWFR subscales significantly explained variance in burden. However, Baker et al found that the RABBM subscale significantly determined variance in strain scores,⁵⁴  implying that men who struggle with affectionate behavior between men report increased burden. This has implications for avoiding emotional support from male friends and family, as previously suggested in relation to men's reluctance to engage in psychological therapy with male therapists.⁵⁵  The difference in findings between these studies may be explained by the younger sample participants in the current study in contrast to the older sample participants in the study by Baker et al⁵⁴  who were no longer in paid employment, so the balance of career and family might have been less pertinent. In the context of this study, the RE subscale is best understood as another component in the need to give an appearance of coping to preclude emotionally burdening others.

The effect of GRC increasing carer burden is balanced by GRF (PAQ-Expressivity subscale), further supporting the argument that a flexible coping predisposition reduces burden. Therefore, one may consider a modified stress process model wherein burden is influenced by gender role constructs and social support, which can affect help-seeking attitudes.

MICs of people with MS can be considered an at-risk group who are vulnerable to developing poor mental health. Both GRC and GRF significantly explain variance in carer burden. Although regression analysis did not indicate that gender constructs are directly associated with attitudes toward help-seeking, other measures provided evidence that MICs are generally unlikely to access personal supports and that this may be influenced by gender factors. These findings build upon previous carer coping research, but more importantly, they provide a nuanced view of the factors that affect MIC coping and provide insights into how to tailor services to engage this population.

A substantial proportion of participants (42%) were recruited via the internet (Facebook and X). Although this method has been criticized for producing unrepresentative samples, empirical data and contemporary prevailing opinion suggest social media outreach can provide a better and more diverse representation of the community.^56-58  Additionally, the sex of the person with MS was not identified. To date, there have been no studies examining this, so it is unknown what effect this may have on a couple's ability to cope.

Further research on developing understandings of appearance of coping is required, as this study's findings, qualitative research, and clinical anecdotal evidence identify appearance of coping as a potential barrier to seeking help and dealing with carer burden. Investigation is required to assess whether educational and therapeutic interventions can substantially affect attitudes toward gender roles and change help-seeking behavior, as previously shown in male helpseeking literature.^59,60 

An avenue of intervention would be for MICs to develop greater understanding of the constraints of traditional masculinity and the effects of adherence upon them and their relationships.^61,62  In this way, men can move toward a caring masculinity⁶³  and be able to embrace and recognize the rewards of caring.

If gender flexibility is reconceptualized as a coping flexibility, as advocated by Cheng et al,⁵²  it may provide additional interventions. A critical component of coping flexibility is acceptance, a skill used when external factors of problems are uncontrollable. This appears consistent with previous findings on couples’ adjustment to coping with MS,⁶⁴  and thus, acceptance-based therapeutic interventions, such as acceptance and commitment therapy, may assist in developing unique interventions for MICs.

Gratitude to the specialist multiple sclerosis (MS) nurses of the United Kingdom, Isle of Man, Jersey, Guernsey, and the MS Society of New Zealand for recruitment of participants. Thanks to Cyleste Fisher for the preparation of this manuscript.