Gender differences in health: a Canadian study of the psychosocial, structural and behavioural determinants of health

doi:10.1016/j.socscimed.2003.09.008

Social Science & Medicine

Volume 58, Issue 12, June 2004, Pages 2585-2600

https://doi.org/10.1016/j.socscimed.2003.09.008 Get rights and content

Abstract

Gender-based inequalities in health have been frequently documented. This paper examines the extent to which these inequalities reflect the different social experiences and conditions of men's and women's lives. We address four specific questions. Are there gender differences in mental and physical health? What is the relative importance of the structural, behavioural and psychosocial determinants of health? Are the gender differences in health attributable to the differing structural (socio-economic, age, social support, family arrangement) context in which women and men live, and to their differential exposure to lifestyle (smoking, drinking, exercise, diet) and psychosocial (critical life events, stress, psychological resources) factors? Are gender differences in health also attributable to gender differences in vulnerability to these structural, behavioural and psychosocial determinants of health?

Multivariate analyses of Canadian National Population Health Survey data show gender differences in health (measured by self-rated health, functional health, chronic illness and distress). Social structural and psychosocial determinants of health are generally more important for women and behavioural determinants are generally more important for men. Gender differences in exposure to these forces contribute to inequalities in health between men and women, however, statistically significant inequalities remain after controlling for exposure. Gender-based health inequalities are further explained by differential vulnerabilities to social forces between men and women. Our findings suggest the value of models that include a wide range of health and health-determinant variables, and affirm the importance of looking more closely at gender differences in health.

Introduction

This paper builds on the contribution of our own work to the understanding of the social structural and behavioural determinants of health (Denton & Walters, 1999; McDonough, 2000; McDonough & Walters, 2001; Walters, McDonough, & Strohschein, 2002) and the psychosocial determinants of health (McDonough, Walters, & Stroschein, 2002). It adds to this literature by assessing the relative contribution to the health of these three categories and by evaluating two explanations of gender differences, namely that men and women are differentially exposed to these determinants of health or that they are differentially vulnerable to these determinants. Further, to gain a comprehensive understanding of the role that such factors play in determining health, we use multiple indicators of both physical and mental health. In this section, we provide an overview of the theoretical and empirical literature on the social explanations of inequalities in health between men and women.

Section snippets

Gender inequalities in health

While women generally experience poorer health than men, the pattern of gender differences in health is varied (Arber & Cooper, 1999; Macintyre, Hunt, & Sweeting, 1996; also see the special issue of Social Science & Medicine edited by Hunt & Annandale, 1999). Women have lower rates of mortality but, paradoxically, report higher levels of depression, psychiatric disorders, distress, and a variety of chronic illnesses than men (Baum & Grunberg, 1991; McDonough & Walters, 2001; Verbrugge, 1985).

Social explanations of inequalities in health

The roots of health inequalities are many, interconnected and complex. In addition to health inequalities being linked to genetic and biological differences, social variables have been identified as a source of health inequalities, which are generally grouped into three categories: social structural, behavioural and psychosocial factors.

In terms of structures of social inequality, research has documented a relationship between health inequalities and socio-economic inequalities in income,

Social explanations of inequalities in health: a gender divide

Since gender is a measure of both biological/genetic and social differences, it is likely that the health inequalities between men and women discussed earlier reflect both sex-related biological and social factors, and the interplay between them (Bird & Rieker, 1999; Verbrugge, 1989). In terms of social factors, researchers pose two general hypotheses to account for gender-based inequalities in health. The differential exposure hypothesis suggests that women report higher levels of health

Research question

We have argued that levels of health are determined by social structures of inequality, differences in health-related behaviours and psychosocial factors including stressful life events, chronic stressors and psychological resources. But the picture is much more complex than that. Life style and psychosocial factors are rooted in the social structures of inequality that define people's lives, such as gender.

In the previous section, we presented various findings from the literature to argue that

Data

Data from the Canadian National Population Health Survey (NPHS) are used for this analysis. Based on a multi-stage stratified probability sampling design developed by Statistics Canada, the NPHS collects information from private household residents on health and illness, use of health services, determinants of health, and demographic and economic characteristics of individuals. While the NPHS produces data for both cross-sectional and longitudinal purposes, this study is based on the 1994–1995

Methods of analysis

Gender differences in health are determined using Multiple Classification Analysis (MCA) techniques (Table 2). To assess the relative contribution of the structural, behavioural and psychosocial determinants, block-regression coefficients (i.e., R²) are used to determine the relative importance of social structural, behavioural and psychosocial determinants of health (Table 3). The differential exposure hypothesis is formally tested using MCA techniques (Table 2). Multiple linear/logit

Limitations

Certain limitations with the data used here must be acknowledged. First, the NPHS household data file does not cover persons residing in institutions, most of whom are older women. In addition, the gender bias in mortality (i.e., men compared to women at middle ages are more likely to suffer from life-threatening chronic health conditions such as diabetes and heart disease, and therefore have a higher probability of being deceased by older age) may produce a healthier population of elderly men.

Results

Table 1 presents for males and females separately the means and percentages for the social structural, behavioural and psychosocial determinants of health. Table 2 presents the means and percentages for the four measures of health. Previous research has documented gender differences in health and those results are confirmed here. In particular, women report slightly lower levels of self-rated health, functional health, and higher levels of distress than men. They also report more chronic health

Discussion

Health is a multi-dimensional concept and studies of the determinants of health are difficult to compare given the various measures used by those working in the field. We have been comprehensive in the measures chosen for analysis by including self-rated health, perhaps the most frequently used measure of health in the social sciences; distress, a common measure of mental health; HUI, a new functional measure of health which provides an overall health score to summarize health capacity on eight

References (53)

S. Arber
Comparing inequalities in women's and men's healthBritain in the 1990s
Social Science & Medicine
(1997)
S. Arber et al.
Gender differences in health in later lifeThe new paradox?
Social Science & Medicine
(1999)
C. Bird et al.
Gender mattersan integrated model for understanding men's and women's health
Social Science & Medicine
(1999)
M. Denton et al.
Gender differences in structural and behavioural determinants of healthan analysis of the social production of health
Social Science & Medicine
(1999)
K. Hunt et al.
Relocating gender and morbidityExamining men's and women's health in contemporary Western societies. Introduction to Special Issue on gender and health
Social Science & Medicine
(1999)
M. Kivimaki et al.
Sense of coherence and healthEvidence from two cross-lagged longitudinal samples
Social Science & Medicine
(2000)
P.M. Lantz et al.
Socioeconomic disparities in health change in a longitudinal study of US adultsThe role of health-risks behaviours
Social Science & Medicine
(2001)
M. Marmot et al.
Social inequalities in healthNext questions and converging evidence
Social Science & Medicine
(1997)
S. Macintyre
The Black Report and beyond what are the issues?
Social Science & Medicine
(1997)
S. Macintyre et al.
Gender differences in healthAre things really as simple as they seem?
Social Science & Medicine
(1996)

S. Matthews et al.

Social inequalities in healthAre there gender differences?

Social Science & Medicine

(1999)

P. McDonough et al.

Gender and healthReassessing patterns and explanations

Social Science & Medicine

(2001)

P. McDonough et al.

Chronic stress and the social patterning of women's health in Canada

Social Science & Medicine

(2002)

D. Shye et al.

Gender differences in the relationship between social network support and mortalityA longitudinal study

Social Science & Medicine

(1995)

D.G. Uitenbroek et al.

Health lifestyle behaviour and socio-demographic characteristics. A study of Varna, Glasgow and Edinburgh

Social Science & Medicine

(1996)

V. Walters et al.

The influence of work, household structure, and social, personal and material resources on gender differences in healthAn analysis of the 1994 Canadian national population health survey

Social Science & Medicine

(2002)

A. Antonovsky

Unravelling the mystery of healthHow people manage stress and stay well

(1987)

A. Baum et al.

Gender, stress and health

Health Psychology

(1991)

M. Blaxter

Health and lifestyles

(1990)

C. D’Arcy

Social distribution of health among Canadians

B. de Vries et al.

A lifetime of eventsAge and gender variations in the life story

International Journal of Aging and Human Development

(1996)

D. Feeny et al.

Multiattribute and single attribute utility functions for the Health Utilities Index mark 3 System

Medical Care

(2002)

J. Hirdes et al.

Estimates of the relative risk of mortality based on the Ontario Longitudinal Study of Aging

Canadian Journal on Aging

(1989)

J.S. House et al.

Social relationships and health

J.B. Jemmot et al.

Psychological factors, immunologic mediation, and human susceptibility to infectious diseasesHow much do we know?

Psychological Bulletin

(1984)

R.C. Kessler et al.

Sex differences in vulnerability to life events

American Sociological Review

(1984)

Cited by (540)

Childhood socioeconomic position and later-life cognitive functioning in the U.S.: A critical review
2024, Developmental Review
An emerging body of research suggests that later-life cognitive functioning may be partly the result of influences across the life course. Low socioeconomic position in childhood is associated with disparities in cognitive functioning in older adulthood. Framed by the life course perspective, several explanations for this association exist: the latency model, positing that the conditions of low early-life socioeconomic position are directly linked to later-life cognitive functioning; the pathway hypothesis, suggesting that the association is mediated through adult socioeconomic position; and the accumulation of (dis)advantage hypothesis, proposing that the combined conditions of the childhood and adulthood contexts are more impactful than either socioeconomic context alone. The purpose of this critical review was to assess the empirical evidence supporting each hypothesis through a synthesis of the extant literature on the association between childhood socioeconomic position and later-life cognitive functioning. We reviewed 29 studies with U.S. samples and found the strongest evidence for the pathway hypothesis, followed by the accumulation hypothesis. Support for the latency model is present but weaker than the other explanations. The influence of childhood socioeconomic position on cognitive functioning is stronger when cognitive functioning is assessed at a single time point rather than as change over time, suggesting that childhood socioeconomic position might not affect the rate at which cognition declines in later life but does impact performance measured at any designated testing occasion. We conclude with a discussion of the limitations of the state of the literature, directions for future research, and implications for policy.
Age and sex differences in the association between neighborhood socioeconomic environment and incident diabetes: Results from the diabetes location, environmental attributes and disparities (LEAD) network
2023, SSM - Population Health
Worse neighborhood socioeconomic environment (NSEE) may contribute to an increased risk of type 2 diabetes (T2D). We examined whether the relationship between NSEE and T2D differs by sex and age in three study populations.
We conducted a harmonized analysis using data from three independent longitudinal study samples in the US: 1) the Veteran Administration Diabetes Risk (VADR) cohort, 2) the REasons for Geographic and Racial Differences in Stroke (REGARDS) cohort, and 3) a case-control study of Geisinger electronic health records in Pennsylvania. We measured NSEE with a z-score sum of six census tract indicators within strata of community type (higher density urban, lower density urban, suburban/small town, and rural). Community type-stratified models evaluated the likelihood of new diagnoses of T2D in each study sample using restricted cubic splines and quartiles of NSEE.
Across study samples, worse NSEE was associated with higher risk of T2D. We observed significant effect modification by sex and age, though evidence of effect modification varied by site and community type. Largely, stronger associations between worse NSEE and diabetes risk were found among women relative to men and among those less than age 45 in the VADR cohort. Similar modification by age group results were observed in the Geisinger sample in small town/suburban communities only and similar modification by sex was observed in REGARDS in lower density urban communities.
The impact of NSEE on T2D risk may differ for males and females and by age group within different community types.
Inflation hardship, gender, and mental health
2023, SSM - Population Health
Inflation hit a 40 year high in the United States in 2022, yet the impact of inflation related hardships on distress is poorly understood, particularly the impact on women, whose income is already more limited. Using data from the US Household Pulse Survey (September–November 2022), we test whether exposure to inflation hardships is associated with greater distress and whether this association is moderated by gender (n = 119,531). We draw on a list of eighteen inflation related hardships (e.g., purchasing less food, working additional jobs, delaying medical treatment) to construct an ordinal measure of exposure to inflation hardship ranging from “no inflation hardship” to “five or more inflation hardships.” We observe that an increasing number of inflation hardships is associated with higher levels of distress. We find no evidence of gender differences in the magnitude of that association at lower levels of inflation hardship (four inflation hardships or less). However, our findings suggest that exposure to five or more inflation hardships is more strongly associated with distress among men compared to women. The current study provides new insights into the cumulative burden of inflation hardships on mental health and the role that gender plays in this association.
Ecological momentary assessment of state affect prior to and following loss of control eating in young men
2023, Eating Behaviors
Loss of control (LOC) eating is prevalent but understudied among young men. Affect regulation models propose that LOC eating functions as a maladaptive effort to escape from distressing affective states. As such, negative affect is thought to increase before and decrease after LOC eating. However, examinations with young men are lacking and it remains unclear whether specific emotional experiences are differentially implicated in their LOC eating.
The current study examined the temporal roles of affect in LOC eating in 31 young men (18–35 years; Mage = 25.74 ± 5.61y; 46.7 % White; 30 % Black/African American; 10 % Hispanic/Latino, 10 % South Asian) who reported engaging in recurrent LOC eating. Participants completed a 14-day ecological momentary assessment protocol and recorded all eating episodes each day and their state affect five times per day. Generalized linear mixed models were conducted to examine the trajectories of global and item-level negative and positive affect pre- and post-LOC eating episodes.
Negative affect did not change significantly before or after LOC eating (ps > .05). Positive affect did not change significantly before LOC eating (ps > .05). Global positive affect, excitement, and happiness decreased significantly after LOC eating (ps ≤ .001).
Study findings contradict extant theory and empirical data largely from female samples. Negative affect did not increase risk for LOC eating, nor did LOC eating function to improve participants' mood; rather, positive mood slightly decreased after LOC eating. Further investigation around the observed decline in positive affect after LOC eating will clarify if this is a relevant intervention point in this population.
Activities of daily living limitations and the use of physical examination among older adults with informal care in China: do gender and residence make differences?
2024, BMC Geriatrics
The sex difference in self-rated health among older Turkish and Moroccan migrants in the Netherlands: an exploratory study of contributing determinants
2024, BMC Public Health

View all citing articles on Scopus

View full text