The association between masculinity scores at ages 18-20 and mortality due to alcohol, violent death and CVD: a prospective study based on the Swedish 1969 conscript cohort
Anna Månsdotter, Andreas Lundin, Daniel Falkstedt, Tomas Hemmingsson
Abstract
Objective: To analyse the relationship between masculinity scores aimed at identifying the effective soldier among men in their late adolescence and subsequent mortality patterns.
Methods: The study is based on a cohort of 49 321 young Swedish males born 1949/51 conscripted for compulsory military training in 1969/70. The independent variable consists of a five scale measure of masculinity from the conscription considering leisure interests (Strong Vocational Interest Blank, 1936) and occupational ambitions (Minnesota Multiphase Personality Inventory, 1943). The actual ranking was perfomed by a psychologist based on information from both questionnaires and interviews. The dependent data on mortality was collected from national registers during 1973-2004 and consists of all cause mortality, alcohol related death, suicide, other violent deaths, and cardiovascular disease (CVD) mortality.
Results: The general picture is that the risk (crude odds ratio) for all cause mortality, and for deaths due to alcohol drinking, suicide, violent death, and CVD is higher among men ranked less masculine than ordinary in terms of leisure and occupational interest. Concurrently, being ranked more masculine than ordinary was protective for all mortality causes. After adjusting for childhood socioeconomic position, blood pressure, lifestyles (alcohol drinking, smoking, and BMI), mental disorder and low intelligence, and earlier contacts with the police and social services, the relationship between masculinity ranking in late adolescence and future death risks was much decreased. Yet, the least masculine had a remaining statistically significant increased risk for overall mortality of 50% compared to the medium masculine men.
Conclusion: The ranking of masculinity among men from the conscription examination is predictive for subsequent all cause mortality and several gendered death causes; mainly in the sense that the lowest ranked (n=5.4%) and the moderately low ranked (n=21.5%) have higher death risks than the medium ranked (n=39.4%). This opposes our initial hypothesis and also the common understanding that masculinity would represent a risk factor for premature death. One probable explanation is that risk preference, which is normally considered vital for demonstrating masculinity and which links to excess mortality, was not considered in the utilised masculinity measure.
Keywords: masculinity, mortality, alcohol, violent death, CVD
Anna Månsdotter, Andreas Lundin, Daniel Falkstedt, Tomas Hemmingsson
Abstract
Objective: To analyse the relationship between masculinity scores aimed at identifying the effective soldier among men in their late adolescence and subsequent mortality patterns.
Methods: The study is based on a cohort of 49 321 young Swedish males born 1949/51 conscripted for compulsory military training in 1969/70. The independent variable consists of a five scale measure of masculinity from the conscription considering leisure interests (Strong Vocational Interest Blank, 1936) and occupational ambitions (Minnesota Multiphase Personality Inventory, 1943). The actual ranking was perfomed by a psychologist based on information from both questionnaires and interviews. The dependent data on mortality was collected from national registers during 1973-2004 and consists of all cause mortality, alcohol related death, suicide, other violent deaths, and cardiovascular disease (CVD) mortality.
Results: The general picture is that the risk (crude odds ratio) for all cause mortality, and for deaths due to alcohol drinking, suicide, violent death, and CVD is higher among men ranked less masculine than ordinary in terms of leisure and occupational interest. Concurrently, being ranked more masculine than ordinary was protective for all mortality causes. After adjusting for childhood socioeconomic position, blood pressure, lifestyles (alcohol drinking, smoking, and BMI), mental disorder and low intelligence, and earlier contacts with the police and social services, the relationship between masculinity ranking in late adolescence and future death risks was much decreased. Yet, the least masculine had a remaining statistically significant increased risk for overall mortality of 50% compared to the medium masculine men.
Conclusion: The ranking of masculinity among men from the conscription examination is predictive for subsequent all cause mortality and several gendered death causes; mainly in the sense that the lowest ranked (n=5.4%) and the moderately low ranked (n=21.5%) have higher death risks than the medium ranked (n=39.4%). This opposes our initial hypothesis and also the common understanding that masculinity would represent a risk factor for premature death. One probable explanation is that risk preference, which is normally considered vital for demonstrating masculinity and which links to excess mortality, was not considered in the utilised masculinity measure.
Keywords: masculinity, mortality, alcohol, violent death, CVD
