IJE Advance Access published online on April 17, 2007
International Journal of Epidemiology, doi:10.1093/ije/dym022
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Decreased risk of death from coronary heart disease amongst men with higher ‘femininity’ scores: a general population cohort study
MRC Social and Public Health Sciences Unit, 4 Lilybank Gardens, Glasgow G12 8RZ, Scotland, UK.
* Corresponding author: Kate Hunt, MRC Social and Public Health Sciences Unit, 4 Lilybank Gardens, Glasgow G12 8RZ, Scotland, UK. E-mail: k.hunt{at}msoc.mrc.gla.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsCommentAcknowledgementsReferences |
|---|
Context At all ages men have higher rates of coronary heart disease (CHD) than women, although similar proportions of men and women eventually die of CHD. Gender differences in CHD incidence and mortality are often explained in relation to biological (hormonal) and behavioural risk factors (e.g. smoking), but psychological factors and broader social constructions of gender are rarely considered.
Objective To examine the relationship between measures of gender role orientation at baseline in 1988 and mortality from CHD over 17 years (to June 2005).
Design Prospective cohort study linked to national mortality reporting.
Setting Socially varied mainly urban area centred on city of Glasgow in West Central Scotland, UK.
Participants In total, 1551 participants (704 men and 847 women) aged 55 years took part in detailed interviews with nurses trained in survey methods in 1988. These included a wide range of measures of physical development and functioning, self reported health and health behaviour, personal and social circumstances and a measure of gender role orientation (yielding scores for ‘masculinity’ and ‘femininity’).
Main outcome measures Mortality from CHD up to June 2005 (88 CHD deaths in men; 41 CHD deaths in women).
Results After adjusting for smoking, binge drinking, body mass index, systolic blood pressure, household income and psychological well-being, higher ‘femininity’ scores in men were associated with a lower risk of CHD death (hazards ratio per unit increase in ‘femininity’ score 0.65, 95% CIs 0.48–0.87, P = 0.004). No such relationship was observed amongst women. ‘Masculinity’ scores were unrelated to CHD mortality in either men or women.
Conclusions These results suggest that social constructions of gender influence the risk of ill health, here death from CHD. Men who are less able to identify themselves with characteristics identified as ‘feminine’ or expressive (who have a more limited stereotypically masculine self-image) may be at increased risk of coronary disease. Further research on the link between social constructions of gender and health is needed.
Keywords Gender, coronary heart disease, mortality, femininity, masculinity, gender role orientation, Bem sex role inventory
Accepted 30 January 2007
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsCommentAcknowledgementsReferences |
|---|
Gender differences in coronary heart disease (CHD) rates have been described as one of the most striking features’ of cardiovascular mortality in the 20th century,1 ‘enigmatic’,2 and ‘one of the most interesting of all epidemiological questions’.3 Although the ratio of male to female deaths has fluctuated to some degree over time, in the UK1 and elsewhere4 and CHD is increasingly recognized as a major cause of morbidity and mortality in women, men are still disadvantaged in experiencing or dying from CHD around 10 years earlier than women.5 Despite advances in understanding CHD, and the recognition of major risk factors (notably cigarette smoking, diabetes, hyperlipidaemia and hypertension which are thought to be present in up to 80–90% of patients with CHD),5,6 questions remain about why some people develop CHD and others do not, not least because many of the risk factors are present in people who have not developed CHD.7 According to the Committee on Understanding the Biology of Sex and Gender Differences, ‘none (of these questions) is more compelling than the differences between the sexes’8 (p. 159). It is often assumed4 that endogenous oestrogens provide a degree of cardioprotection to women prior to the menopause,8 although, as Barrett-Connor has remarked, ‘proof of this hypothesis has been surprisingly elusive’.3 A protective effect of oestrogen cannot explain gender differences alone given the rapidly changing trends in CHD mortality by gender.4 Further explanations for the patterns of CHD by gender have included other biological explanations (e.g. sex-linked inheritance, or central obesity), and classic risk factors such as smoking, alcohol consumption and HDL cholesterol. However, whilst differences in common CHD risk factors between men and women are clearly important, they do not appear to explain fully the differences.3,9
In recent years more attention has focused on whether masculinity, or particular practices of masculinity,10 is detrimental to men's health.11–16 It has been suggested that a sizeable portion of men's excess mortality over women is linked to masculine identity, men's roles and gendered patterns of socialization.9 In relation to CHD, Helgeson suggests that traditional masculinity may both place men at greater risk for heart disease and influence their adjustment to heart disease, invoking men's supposed ‘unwillingness to rely on others for assistance and inability to express emotions’ (p. 69). However, she notes the shortage of empirical research in this area.9
In general, physiological and behavioural risk factors for CHD have been more widely researched than psychosocial factors, although there has been interest in depression and anxiety, psychosocial conditions (particularly in the formal labour market), social support and ‘type A’ or ‘hostile’ personalities.17 Helgeson has remarked on the failure to recognize the role that ‘traditional masculinity plays in the development of (Type A) behaviour’ (p. 755).18 Hemingway and Marmot17 proposed that psychosocial factors could be linked to CHD in various ways: indirectly via health behavioural risk factors (such as smoking, alcohol and dietary consumption and physical activity); directly via acute or chronic physiological changes; and via uptake of medical care.
One additional psychological measure that has received little attention when studying gender differences in health19 is gender role orientation (GRO). The originator of a widely used measure of GRO, Sandra Bem,20,21 stimulated debate on androgyny (principally within psychology). Bem argued that ‘psychologically "androgynous"’ individuals (those who were able to be ‘both instrumental and expressive, both assertive and yielding, both masculine and feminine’,22 (p. 634) emphasis in original) would have a ‘sex role adaptability (that) enables them to engage in situationally effective behaviour without regard for its stereotype as masculine or feminine’ (p. 634). She raises the possibility that ‘the androgynous individual will some day come to define a new and more human standard of psychological health’. Although she did not extend this to other aspects of health or health-related behaviour, it is interesting to test empirically whether such relationships exist.
Helgeson asserts that masculinity is positively associated with psychological health, but negatively with health behaviour or physical health.9 Others have hypothesized that protective health behaviours will be positively associated with both masculinity (instrumentality) and femininity (expressiveness).23 However, studies which have examined health behaviours in relation to Bem's ‘masculinity’ and ‘femininity’ scores do not consistently support this hypothesis. In a general population sample in the UK, measures of ‘masculinity’ and ‘femininity’ were positively associated with current smoking amongst women born in the early 1950s (but not amongst women born in the early 1930s); for men, ‘femininity’ scores were positively associated with smoking in men born in the early 1930s but not amongst men 20 years younger, but ‘masculinity’ was not related to smoking at either age.24 Amongst white-collar employees in the UK, masculinity but not femininity scores were positively associated with smoking in both men and women in two large organizations, a bank and a university, and heavy drinking was positively associated with masculinity but not femininity in both organizations.25 A study comparing adults in Moscow and Toronto reported that masculinity was associated with lower alcohol use among men and higher use among women in Moscow, but that femininity was associated with higher alcohol use among women in Toronto.26 We do not find the inconsistency of these results surprising; rather they illustrate the central importance of the broader (social, historical, political and cultural) context 27 and the complexity of the ways in which behaviours are linked to gender, as illustrated by historical research on smoking.28–30
Although measures of GRO have been widely used, examining the relationship between ‘masculinity’ and ‘femininity’ scores and a whole host of other outcomes related directly or indirectly to health (e.g. depression and anxiety, eating disorders, alcohol dependence, sexuality and sexual orientation, body image, health risk taking), we have been able identify only one study (a population-based cohort study of cardiovascular disease amongst residents of Rancho Bernardo initially surveyed in 1972–74)31 which has examined the relationship between these and CHD mortality. These unpublished data, presented by Barrett-Connor (Fig. 2),3 suggest that ‘masculine’, ‘feminine’ and ‘androgynous’ traits were not related to fatal CHD, but this preliminary analysis was based on just 24 deaths. Here we investigate in a community-based sample whether measures of GRO are related to CHD mortality in men and women.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsCommentAcknowledgementsReferences |
|---|
Participants are from the West of Scotland Twenty-07 Study, a longitudinal study of social patterning of health in three age cohorts. Here we investigate mortality in the oldest cohort who were born in the early 1930s, and were aged around 55 years when first studied in 1988. Respondents were sampled from residents in the Central Clydeside Conurbation, a socially varied but mainly urban area centred on the city of Glasgow in the west of Scotland, UK. The initial sample size for the 1930s cohort was 1551. Baseline data collection comprised two detailed structured interviews, the second conducted by registered nurses trained in interview techniques, usually in the respondent's home, and included a wide range of measures of self-reported health and health behaviour, physical development and functioning and personal and social circumstances. Interviews were supplemented by a self-completion questionnaire, mailed out in advance and collected by the nurse at the time of the interview, which included several standardized, well-validated measures.
In the UK, the National Health Service Central Registers provide a reliable and virtually complete facility, which initiates an automatic report of the date and cause of any deaths occurring in ‘flagged’ cohorts. This is a standard approach for ascertaining mortality in the UK. This analysis reports mortality follow-up to June 2005. For all deaths, a copy of the death certificate was available. For our primary analyses, those with codes 410 to 414.9 in ICD9 and I20 to I25.9 in ICD10 for the main cause of death were coded as CHD deaths. In supplementary analyses, we also coded any mention of CHD on the death certificate as a CHD death, although we expected this more inclusive outcome to reveal a weaker relationship with risk factors for CHD.
Measurement of ‘masculinity’, ‘femininity’ and androgyny
‘Masculinity’ and ‘femininity’ scores were measured using the Short Form of the Bem Sex Role Inventory (BSRI), a widely used self-complete measure of GRO.20,24,32 The BSRI is premised on the assumption that masculinity and femininity are both conceptually and empirically independent. It relies on an individual's endorsement (on a scale from 1—’never or almost never true’ to 7—’always or almost always true’) of a series of qualities, which had been judged in the USA to be culturally characteristic of either males or females. This measure has been validated in Scotland.33 The ‘masculinity’ and ‘femininity’ scores are each the mean of ratings of 10 items (‘masculinity’—defend my own beliefs, independent, assertive, strong personality, forceful, have leadership abilities, willing to take risks, dominant, willing to take a stand, aggressive; ‘femininity’—affectionate, sympathetic, sensitive to the needs of others, understanding, compassionate, eager to soothe hurt feelings, warm, tender, loves children, gentle). As an inspection of missing data showed that cases with any missing data mostly either missed all or only one or two items, masculinity and femininity scores were calculated for subjects who answered eight or more of the 10 items. ‘Masculinity’ and ‘femininity’ scores were available for 1415 (91.2%) and 1417 (91.4%) of the 1551 participants, respectively.
The ‘masculinity’ and ‘femininity’ scores were generally normally distributed.19 The internal reliability of both scales was high (Cronbach's alpha: masculinity scale 0.84; femininity scale 0.87). As expected, correlations between the two scores are low (r = 0.07).
There has been considerable discussion about how best to score the BSRI and to assess whether people are ‘androgynous’ or ‘sex-typed’ on the basis of their BSRI scores.32 Bem recommended a ‘median-split’ method yielding four distinct groups depending on whether individuals scored above (high) or below (low) the median on the ‘masculinity’ and ‘femininity’ scales: ‘androgynous’ [high ‘femininity’ score (F), high ‘masculinity’ score (M)]; ‘masculine’ (high M, low F), ‘feminine’ (high F, low M) or ‘undifferentiated’ (low M, low F). However, we do not favour this method since it loses much of the information on variation within the population, and is prone to misclassification with many people's scores differing little from the median.32 Hence, our primary analyses use the masculinity and femininity scores as continuous variables, testing for an interaction between the two scales as an indicator of ‘androgyny’. However, we have chosen also to include results from the median-split four-category method for comparability with the only other (unpublished) data on GRO and CHD mortality (the analysis of the Rancho Bernardo cohort described above).3 Here, as one method advocated by Bem,20 we have calculated the median value for our population by weighting the sample to give equal numbers of men and women. This is another disadvantage to using the median split method since the measure is effectively standardized to our study population rather than to some generic, generalizable group. We could have used cut-offs recommended in the manual for the BSRI, but these are based on a non-representative sample of US college students collected in the early 1970s. This seems an inappropriate reference group given age and cultural differences and historical changes in gender roles. Scores greater than or equal to the median score are classified as ‘high’, all others as ‘low’. We use the ‘androgynous’ group as the comparator group when assessing the relationship with CHD mortality since Bem consistently theorized (albeit with respect to psychological health) that this is the group which should be in the best health, for both men and women.
Covariates
Of the major established risk factors for CHD, we included smoking, alcohol use, blood pressure, body mass index (BMI) and socio-economic status as covariates. Respondents were categorized as never smokers, ex-smokers and current smokers using self-reports of smoking history. We use a measure of binge drinking, rather than total consumption of alcohol, as an indicator of alcohol consumption as this is most consistently shown to be related to mortality.34,35 We have defined binge drinking as consumption of 10+ units on a single occasion in the last week in men, and 7+ units in women.36,37 Blood pressure was measured in the sitting position, in the right arm after a five-minute rest, using a Hawksley random zero sphygmomanometer and an adult size cuff (with inflatable bladder 35 x 12 cm). Two readings of systolic and diastolic phase 5 were recorded to the nearest even number; the average of the two systolic measurements is used in this analysis. BMI was calculated as weight (kg)/height (m)2. Height was measured to the nearest centimetre using Nivotoise stadiometers, respondents standing without shoes, heels against the wall and head in the Frankfort plane. Weight (clothed) was measured to the nearest 0.5 kg by portable electronic scales calibrated at the local trading standards office. Household income was used as an indicator of socio-economic status. No measures of hyperlipidaemia are available because blood samples were not taken. No adjustment was made for age because of the very limited age range of participants, nor for ethnicity as the sample was predominantly white, reflecting the relative lack of ethnic diversity in this part of the UK.
Although psychological well-being is often not included as a covariate in studies of CHD, we have included it here because it has been shown to be related both to CHD mortality17 and to masculinity scores, including in the Twenty-07 Study.19,27 Here we have measured psychological well-being using the 12-item version of the General Health Questionnaire (GHQ), a well-validated and commonly used measure in community-based studies.38 Scores on each item range from 0 to 3, Likert scores (range 0–36) are used.
Statistical analysis
Frequencies or descriptive statistics for the predictor variables and covariates, and descriptive statistics by gender of the masculinity and femininity scores were obtained. T-tests were used to test for gender differences in the mean masculinity and femininity scores. Univariate statistics were used to investigate relationships between the masculinity and femininity scores and the covariates. A series of Cox regression models was run separately for males and females to assess the relationship between CHD mortality and (i) ‘masculinity’ and ‘femininity’ scores, (ii) the BSRI fourfold classification and (ii) each of the covariates. Initially, univariate relationships were assessed. A Cox regression model testing for an interaction between the ‘masculinity’ and ‘femininity’ scores was also run. Finally, models were run including both the ‘masculinity’ and ‘femininity’ scores together (or the BSRI ‘median-split’ 4-fold classification), adjusted for smoking, binge drinking, blood pressure, BMI, occupational social class and GHQ score. On theoretical grounds, the ‘androgynous’ group was taken as the reference category in the ‘median-split’ analyses (as Bem argued that they should have the best psychological health, and we have extrapolated this to wider health).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsCommentAcknowledgementsReferences |
|---|
Over a quarter (28.3%, 439 deaths) of the cohort died from all causes during follow-up (from baseline interview in 1988 to 12 June 2005). As expected CHD was the most prevalent cause of death and CHD mortality was higher amongst men; 12.5% of men (88 deaths) and 4.8% of women (41 deaths) had died with CHD as the major cause of death.
As expected, men had higher scores on the ‘masculinity’ scale than women (mean 4.50 vs 4.08, P < 0.001) and lower scores on the ‘femininity’ scale (mean 5.30 vs 5.63, P < 0.001) (Table 1). More men than women were classified as ‘masculine’ (35% vs 14%), and fewer as ‘feminine’ (13% vs 36%), but similar proportions of men and women were classified as ‘androgynous’ (24% vs 22%) and ‘undifferentiated’ (29% vs 28%). Around a quarter of the sample (27% men, 26% women) were more socio-economically advantaged (head of household in a professional or intermediate job), although the reported household income was lower for women than men. More women than men were never smokers (41% vs 23%) and fewer were current smokers (40% vs 51%). Binge drinking in the previous week was much higher in men (33%) than women (5%). Mean GHQ scores were marginally higher in women (11.94, 5.35 SD) than men (10.98, 4.86 SD), whereas the reverse was apparent for systolic blood pressure with slightly elevated values in men [138.02 (21.26 SD) vs 135.99 (20.06 SD)]. Mean BMI did not differ between the sexes. Univariate analyses (data not shown) revealed expected relationships between covariates and CHD death amongst men: a significantly increased hazard ratio (HR) for CHD death associated with being a current smoker (HR = 1.80, 95% CIs 1.33–2.44, P < 0.000) and having poorer psychological well-being (HR per unit increase in GHQ score = 1.06, 95% CIs 1.02–1.11, P = 0.002). A trend towards higher risk with increasing systolic blood pressure (HR per unit increase = 1.01, 95% CIs 1.00–1.02, P = 0.08) was apparent. Perhaps owing to the lower number of deaths, none of the univariate risk ratios were statistically significant for women although there was some evidence of a relationship between current smoking and CHD mortality (HR = 1.48, 95% CIs 0.98–2.22, P = 0.06).
|
Table 2a shows the relationship between the continuous ‘masculinity’ and ‘femininity’ scores for CHD death and all covariates separately for men and women. Similar results are presented in Table 2b for the median-split 4-fold classification.
|
|
In men, higher ‘femininity’ scores were associated with a lower risk of CHD death: the unadjusted HR for each unit increase in femininity score was 0.68 (95% CIs 0.50–0.91, P = 0.011) (Table 3). This was little changed following adjustment for other risk factors (HR per unit in femininity score 0.65, 95% CIs 0.48–0.87, P = 0.004). No relationship was observed between ‘femininity’ and CHD in women. There was no relationship between ‘masculinity’ score and CHD mortality in either men or women (unadjusted HRs 0.99, 95% CIs 0.76–1.30, P = 0.950) for men, 0.90, 95% CIs 0.62–1.33, P = 0.608 for women). These ratios altered little when adjusting for other potential risk factors (smoking, binge drinking, BMI, socio-economic status and psychological well-being). There was no evidence of interaction and between ‘masculinity’ and ‘femininity’ score for either men or women.
|
Various sensitivity analyses were performed. First, as 13.6% of men and 22.2% of women had missing data on the binge-drinking variable, the unadjusted and adjusted models were run without binge drinking. Unadjusted and adjusted HRs were very similar to those reported in Table 3 (adjusted HRs: men, masculinity 1.14 (0.89–1.45), femininity 0.69 (0.53–0.90); women, masculinity 0.89 (0.63–1.27), femininity 1.25 (0.78–2.02). Secondly, all models were rerun with any mention of CHD on the death certificate as the outcome. In general, there was a small attenuation in the HRs for men (adjusted risk ratios, masculinity 1.04 (0.81–1.32), femininity 0.75 (0.57–0.98)) and minor discrepancies for women (adjusted HRs, masculinity 1.15 (0.83–1.59), femininity 1.21 (0.78–1.88)).
For the purposes of comparability with results from the Rancho Bernardo cohort, Table 4 reports analyses using the median split method. In comparison with men who were classified as ‘androgynous’ (i.e. had ‘high’ masculinity and ‘high’ femininity scores), those who were classified as ‘masculine’ (i.e. had ‘high’ ‘masculinity’ and ‘low ‘‘femininity’ scores) had a higher risk of CHD mortality (unadjusted HR = 1.58, 95% CIs 1.08–2.31, P = 0.018). After adjustment for other risk factors, the HR increased somewhat (Adjusted HR = 1.69, 95% CIs 1.15–2.47, P = 0.007). However, the men classified as ‘feminine’ (i.e. high ‘femininity’ and low ‘masculinity’) and ‘undifferentiated’ (i.e. low ‘femininity’ and low ‘masculinity’) did not differ from the ‘androgynous’ men in CHD risk. No differences in CHD risk were seen between the groups amongst women.
|
|
Comment |
|---|
|
TopAbstractIntroductionMethodsResultsCommentAcknowledgementsReferences |
|---|
Although there has been increasing attention in recent years to men's health,39–41,11,12,15 some focusing specifically on men's experience of heart disease,9,18, 42–44 empirical research linking experiences of masculinities to health is limited. It is ironic that, although there has been a predominance of male subjects in research on heart disease (until recently at least),45 earlier research on heart disease was gender neutral, failing to recognize the ways in which gendered expectations and experiences are socially constructed and may contribute to men's increased risk of heart disease.
Our results showed a relationship between lower levels of ‘femininity’, or expressive traits, and increased CHD mortality in men but not women, in a cohort in which we saw the expected relationship between key traditional risk factors46 and CHD mortality. Scores on a ‘masculinity’, or instrumental dimension did not relate to CHD risk in either men or women. These results, if replicated in other studies, raise a number of questions about why higher levels of stereotypically feminine or expressive traits are associated with lower CHD risks in men. In this analysis, we have been best able to address the first of the pathways proposed by Hemingway and Marmot (that psychosocial factors are linked to CHD indirectly via health behavioural risk factors). However, whilst we have adjusted for two important health behavioural risk factors (smoking and binge drinking), we were not able to control for other such factors (e.g. diet, general orientation towards health maintenance, etc). Similarly, we do not have the measures in this study to address any direct pathways via acute or chronic physiological changes; studies with more biological measures are required to investigate such pathways. Similarly, a different design would be required to investigate how instrumentality and expressiveness relate to the uptake of medical care, although it is perhaps relevant to note that a study of college students in the US reported the ‘unexpected’ finding that men with high ‘femininity’ scores (as measured by the BSRI) were more concerned about health risks than other men or women.47
As far as we are aware, only one other study has examined whether ‘masculinity’ and ‘femininity’ are related to CHD mortality.3 Barrett-Connor reported a preliminary analysis of CHD deaths in men and women in the Rancho Bernardo cohort in the USA, and found no association between ‘masculinity’, ‘femininity’ and ‘androgenicity’ (also measured using the BSRI using the median-split method) after 6 years of follow-up. However, this analysis was based on a small number of deaths (14 in men and 10 in women). When using the ‘median-split’ method in our analyses, the increased risk associated with low ‘femininity’ scores was confined to men who had high ‘masculinity’ scores (i.e. the ‘masculine’ men). The ‘feminine’ men (high F, low M) and the ‘undifferentiated’ men (low F, low M) did not differ from the androgynous men (high M, high F). Thus, the two methods of analysing the relationship between GRO and CHD mortality are only partially consistent. Both suggest that GRO is not associated with CHD risk in women. The ‘continuous’ method suggests that higher femininity scores are protective in men (irrespective of masculinity scores since the interaction term was not significant). However, the ‘median-split’ method suggests that lower ‘femininity’ scores were only detrimental when men also had high ‘masculinity’ scores, possibly equating with a more ‘extreme’ version of masculinity. Helgeson has reported a relationship between ‘negative or extreme masculinity’ (also referred to as ‘unmitigated agency’) and heart attack severity amongst US men (not mediated by type A behaviour, poor health practices or impaired social networks)18 and suggests that there is ‘emerging evidence that aspects of masculine identity and male socialization are related to men's risk for coronary heart disease’ (p. 68).9
Our study has a number of strengths. It is a community-based general population sample suggesting a high degree of generalizability of our findings. We have: accurate, independent reports of deaths; validated and widely used measures of GRO and psychological well-being; and directly measured height, weight and blood pressure. The study also has limitations. First, because we are comparing men and women of the same age, we have relatively few CHD deaths in women and therefore reduced statistical power for detecting relationships between any risk factors and CHD mortality. Secondly, we do not have measurements of some important CHD risk factors such as HDL cholesterol. Thirdly, had our sample size been large enough to test more robustly for interactions between the ‘masculinity’ and ‘femininity’ scores there may not have been discrepancies between the two methods.
As we have noted elsewhere,48 theorists of masculinity10 have stressed the complex and multifaceted nature of masculinity, or masculinities. Here we have used an instrument which taps into some dimensions of stereotypical masculinity and femininity, but clearly has limitations.32 The BSRI allows people to rate themselves against ‘some’ aspects contributing to common cultural notions of hegemonic (or dominant) forms of masculinity (which might be health-damaging, such as being willing to take risks and being aggressive) and femininity, but is inevitably limited. Hoffman and Borders argue that there is now a challenge to think of new models for conceptualizing masculinity and femininity ‘in non-stereotypical ways’, and that the BSRI may ‘best be used to examine one's instrumentality and expressiveness, providing that the importance attached to one's classification is minimized and providing that instrumentality and expressiveness are described as human traits rather than as synonyms for masculinity and femininity’ (p. 53).32
Our results present some empirical support for theories that social constructions of gender are important for health, and suggest that this is an important area to study further. We believe that it is important to explore ways in which societal constructions of gender may explain sex differences in longevity, and in specific diseases like CHD which contribute substantially to premature mortality and to inequalities in health by gender (and other social characteristics). We hope that these results, which suggest that men who are less able to identify themselves with characteristics that have been identified as being ‘feminine’ or expressive (and thus perhaps subscribe to a more limited and stereotypically ‘masculine’ range of behaviours) are at an increased risk of dying of CHD, will stimulate discussion and further research into the links between the social construction of gender and health.
| KEY MESSAGES What is already known:
What this article adds:
|
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMethodsResultsCommentAcknowledgementsReferences |
|---|
The West of Scotland Twenty-07 Study is funded by the UK Medical Research Council and the data were originally collected by the MRC Social and Public Health Sciences Unit. We are grateful to all of the participants in the Study, and to the survey staff and research nurses who carried it out and to Sally Macintyre and the anonymous referees for comments on an earlier draft. The data are employed here with the permission of the Twenty-07 Steering Group. KH, HL and CE are all funded by the MRC. David Batty is a Wellcome Trust Fellow.
Conflicts of interest: None declared.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsCommentAcknowledgementsReferences |
|---|
1 Charlton J, Murphy M, Khaw K, Ebrahim S, Davey Smith G. (1997) Cardiovascular diseases. In Charlton J and Murphy M (Eds.). The Health of Adult Britain 1841–994.(The Stationery Office, London) pp. 60–81.
2 Lawlor D, Ebrahim S, Whincup P, et al. (2004) Sex differences in body fat distribution and carotid intima media thickness: cross sectional survey using data from the British regional heart study. J Epidemiol Comm Health 58:700–4.
3 Barrett-Connor E. (1997) Sex differences in coronary heart disease. Why are women so superior? The 1995 Ancel Keys Lecture. Circulation 95:252–64.
4 Lawlor DA, Ebrahim S, Davey Smith G. (2001) Sex matters: secular and geographical trends in sex differences in coronary heart disease mortality. Br Med J 323:541–45.
5 Khot U, Khot M, Bajzer C, Sapp S, Ohman E, Brener S, et al. (2003) Prevalence of conventional risk factors in patients with coronary heart disease. JAMA 290:898–904.
6 Greenland P, Knoll M, Stamler J, Neaton J, Dyer A, Garside D, et al. (2003) Major risk factors as antecedents of fatal and nonfatal coronary heart disease events. JAMA 290:891–97.
7 McConnachie A, Hunt K, Emslie C, Hart C, Watt G. (2001) "Unwarranted survivals" and "anomalous deaths" from coronary heart disease: prospective survey of general population. Br Med J 323:1487–91.
8 Exploring the biological contributions to human health. In Wizemann TM and Pardue M-L (Eds.). Does sex matter? (2001) (National Academy Press, Washington DC).
9 Helgeson VS. (1995) Masculinity, men's roles, and coronary heart disease. In Sabo D and Gordon D (Eds.). Men's health and illness Gender, power and the body.(Sage, California) pp. 68–104.
10 Connell R. (1995) Masculinities.(Polity Press, Cambridge).
11 Watson J. (2000) Male bodies: Health, Culture and Identity.(Open University Press, Buckingham).
12 Courtenay W. (2000) Constructions of masculinity and their influence on men's well-being: a theory of gender and health. Soc Sci Med 50:1385–401.[CrossRef][Web of Science][Medline]
13 O’Brien R, Hunt K, Hart G. (2005) Men's accounts of masculinity and help-seeking: ‘It's caveman stuff, but that is to a certain extent how guys still operate’. Soc Sci Med 61:503–16.[CrossRef][Web of Science][Medline]
14 Banks I. (2001) No man's land: Men, illness and the NHS. BMJ 323:1058–60.
15 Robertson S. (2003) Men managing health. Men's Health J 2:111–13.
16 Stanistreet D, Bambra C, Scott-Samuel A. (2005) Is patriarchy the source of men's higher mortality? J Epidemiol Comm Health 59:873–76.
17 Hemingway H and Marmot M. (1999) Evidence based cardiology: psychosocial factors in the aetiology and prognosis of coronary heart disease: systematic review of prognostic cohort studies. Br Med J 318:1460–67.
18 Helgeson VS. (1990) The role of masculinity in a prognostic predictor of heart attack severity. Sex Roles 22:755–74.[CrossRef][Web of Science]
19 Annandale E and Hunt K. (1990) Masculinity, femininity and sex: an exploration of their relative contribution to explaining gender differences in health. Sociol Health Ill 12:24–46.[CrossRef]
20 Bem SL. (1981.) Bem Sex-Role Inventory. Professional Manual. Palo Alto.(Consulting Psychologists Press Inc, California).
21 Bem S. (1974) The measurement of psychological androgyny. J Cons Clin Psychol 42:155–62.[CrossRef][Web of Science][Medline]
22 Bem SL. (1975) Sex role adaptability: one consequence of psychological androgyny. J Pers Soc Psychol 31:634–43.[CrossRef][Web of Science]
23 Shifren K and Bauserman RL. (1996) The relationship between instrumental and expressive traits, health behaviours, and perceived physical health. Sex Roles. 34:841–64.[CrossRef][Web of Science]
24 Hunt K, Hannah MK, West P. (2004) Contextualising smoking: masculinity, femininity and class differences in smoking in men and women from three generation in the west of Scotland. Health Edu Res Theory Prac 19:239–49.
25 Emslie C, Hunt K, Macintyre S. (2002) How similar are the smoking and drinking habits of men and women in non-manual jobs? Eur J Pub Health 12:22–28.
26 van Gundy K, Schieman S, Kelley MS, Rebellon CJ. (2005) Gender role orientations and alcohol use among Moscow and Toronto adults. Soc Sci Med 61:2317–30.[CrossRef][Web of Science][Medline]
27 Hunt K. (2002) A generation apart? An examination of changes in gender-related experiences and health in women in early and late mid-life. Soc Sci Med 54:663–76.[CrossRef][Web of Science][Medline]
28 Hilton M. (1995) "Tabs", "Fags" and the "Boy Labour Problem" in late Victorian and Edwardian Britain. J Soc Hist 28:586–607.
29 Hilton M. (1996) Constructing Tobacco: perspectives on consumer culture in Britain 1850-1950. PhD thesis.(Lancaster University, Lancaster).
30 Elliot R. (2001) ‘Destructive but sweet’: cigarette smoking among women 1890-1990. PhD thesis [PhD].(Glasgow University, Glasgow).
31 Criqui MH, Barrett-Connor E, Austin M. (1978) Differences between respondents and non-respondents in a population-based cardiovascular disease study. Am J Epidemiol 108:367–72.
32 Hoffman RM and Borders LD. (2001) Twenty-five years after the Bem Sex Role Inventory: a reassessment and new issues regarding classification variability. Meas Eval Couns Dev 34:39–55.
33 Stroebele SA. (1992) A validation of the Bem Sex Role Inventory (BSRI) in the West of Scotland. Perception of sex stereotypes. MPH thesis [MPH Thesis].University of Glasgow.
34 Kauhanen J, Kaplan GA, Goldberg D, Cohen RD, Lakka TA, Salonen JT. (1997) Frequent hangovers and cardiovascular mortality in middle-aged men. Epidemiology 8:310–14.[CrossRef][Web of Science][Medline]
35 Kauhanen J, Kaplan GA, Goldberg D, Salonen JT. (1997) Beer binging and mortality: results from the Kuopio ischaemic heart disease risk factor study, a prospective population based study. Br Med J 315:846–51.
36 Moore L, Smith C, Catford J. (1994) Binge drinking: prevalence, patterns and policy. Health Educ Res 9:497–505.
37 Royal College of Psychiatrists. (2001) Alcohol – can the NHS afford it?(Royal College of Physicians, London).
38 Goldberg D. (1978) Manual of the General Health Questionnaire.(NFER Publishing Company, Windsor).
39 Sabo D and Gordon DF. (1995) Rethinking men's health and illness. The relevance of gender studies. In Sabo D and Gordon D (Eds.). Men's Health and Illness: Gender Power and the Body.(Sage Publications, Newbury Park, California) pp. 1–21.
40 Men's Health and Illness: Gender, Power and the Body. (1995) (Sage PublicationsIn Sabo D and Gordon DF (Eds.). , Thousand Oakes, CA).
41 Sabo D. (2000) "Men's health studies: Origins and trends.". J Am Coll Health 49:133–43.[Web of Science][Medline]
42 White A. (1999) ’I feel a fraud’: Men and their experiences of acute admission following chest pain. Nurs Crit Care 467–73.
43 White A and Johnson M. (2000) Men making sense of their chest pain - niggles, doubts and denials. J Clin Nurs 9:534–41.[CrossRef][Web of Science][Medline]
44 O’Brien R, Hart G, Hunt K. "Standing out from the herd2": (Re) negotiating masculinity in relation to men's experience of illness. Int J Men's Health. (in press).
45 Emslie C. (2005) Women, men and coronary heart disease: a review of the qualitative literature. J Adv Nurs 51:382–95.[CrossRef][Web of Science][Medline]
46 Yusuf OS, Reddy S, Ounpuu S, Anand S. (2001) Global burden of cardiovascular diseases: Part 1: general considerations, the epidemiologic transition, risk factors, and impact of urbanization. Circulation 104:2746–53.
47 Kaplan MS and Marks G. (1995) Appraisal of health risks: the roles of masculinity, femininity and sex. Sociol Health Ill 17:206–21.[CrossRef]
48 Hunt K, Sweeting H, Keoghan M, Platt S. (2006) Sex, gender role orientation, gender role attitudes and suicidal thoughts in three generations. A general population study. Soc Psych Psych Epid 41:641–47.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. Benzeval, G. Der, A. Ellaway, K. Hunt, H. Sweeting, P. West, and S. Macintyre Cohort Profile: West of Scotland Twenty-07 Study: Health in the Community Int. J. Epidemiol., October 1, 2009; 38(5): 1215 - 1223. [Full Text] [PDF]
|
|
|
|
 |
|
 G. D. Batty, H. Lewars, C. Emslie, C. R. Gale, and K. Hunt Internationally recognized guidelines for 'sensible' alcohol consumption: is exceeding them actually detrimental to health and social circumstances? Evidence from a population-based cohort study J. Public Health Med., September 1, 2009; 31(3): 360 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Mansdotter, A Lundin, D Falkstedt, and T Hemmingsson The association between masculinity rank and mortality patterns: a prospective study based on the Swedish 1969 conscript cohort J Epidemiol Community Health, May 1, 2009; 63(5): 408 - 413. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||