Cancer,Body,and Mastery at the Intersection of Gender and Race

Abstract

Using the 2006-2014 data from the Health and Retirement Study, the author compares changes in personal mastery after a new cancer diagnosis among white men, white women, black men, and black women. The author further examines the physical burden of cancer (incontinence, fatigue, pain, and decreased strength) as a mechanism mediating the effect of cancer on mastery in each group and finds that white men experience a substantially more pronounced decline in mastery after the onset of cancer than all women and black men, despite white men’s advantaged material resources and favorable cancer-related symptoms. This steepest decline in mastery among white men is entirely due to a disproportionately adverse effect of physical symptoms on mastery. The author argues that the physical burden of cancer might pose a profound threat to white men’s cultural privilege by undermining the masculine body—a critical and highly visible resource for “doing” masculinity.

Keywords

gender mastery psychosocial resources socioeconomic status race/ethnicity

Because of the rapid population aging and improvements in cancer detection and treatment, an unprecedented number of people in the United States are currently living with cancer, and this number will be increasing in the next decades (American Cancer Society 2016). Cancer exerts a profound physical and psychological toll both in the short term and long after the diagnosis and treatment (Kenyon, Mayer, and Owens 2014; Pudrovska 2010b; Watts et al. 2015). Given that more people are living with cancer than ever before, research on psychological well-being of cancer survivors becomes critically important and timely. There is a pressing need to expand current knowledge of psychosocial resources that promote coping and positive adjustment to cancer.

Of all health-enhancing psychosocial resources, “belief in control over one’s own life may be the most important” (Mirowsky and Ross 1986:26). Mastery, or personal control, is a belief in one’s ability to control one’s destiny, influence one’s environment, and bring about desired life outcomes (Pearlin and Schooler 1978). High personal mastery was consistently shown to improve multiple aspects of adjustment to chronic illness, including disease management, adherence to treatment regimens, and even survival (Baum and Posluszny 1999; Bettencourt et al. 2008; Skaff et al. 2003; Surtees et al. 2010). Personal mastery has a powerful effect on mental health, with higher mastery decreasing and lower mastery increasing depressive symptoms and anxiety (Mirowsky, Ross, and Van Willigen 1996; Pearlin et al. 1981; Pudrovska et al. 2005). Mastery as a buffer against psychological distress can be particularly beneficial for cancer patients who are disproportionately affected by depression and anxiety relative to the general population (Deimling et al. 2006; Pudrovska 2010b; Watts et al. 2015).

Uncertainty over the future is inherent in cancer more than in any major chronic disease (J. N. Clarke and Everest 2006). A potentially life-threatening nature of a cancer diagnosis may lead to an acute realization of the volatility and finitude of one’s body and self (Shaha et al. 2008). A prevailing cultural view of cancer as an uncontrollable and “dread” disease can be particularly conducive to the fatalistic beliefs in newly diagnosed patients. Moreover, relative to other chronic conditions, cancer and its treatments exert a particularly devastating effect on the body. Physical symptoms among cancer patients include nausea, pain, fatigue, decreased strength and stamina, urinary and bowel incontinence, limpness, and weakness (American Cancer Society 2016; Ervik and Asplund 2012; McCaughan et al. 2012). Extensive evidence documents a strong adverse effect of physical impairment on the sense of control among older adults (Dewan, MacDermid, and Packham 2013; McIlvane, Schiaffino, and Paget 2007; Mirowsky 1995). Physical symptoms of cancer can undermine beliefs in personal mastery because they impose constraints on daily activities and force individuals to give up or scale down long-term goals and aspirations (Mirowsky 1995; Ross and Mirowsky 2002; Vehling et al. 2012). Moreover, cancer patients might view bodily changes as permanent or indefinite, not amenable to effort and willpower, and eroding one’s abilities and achievements (Calasanti, Pietila, and Ojala 2013). The physical burden of cancer might pose a particular threat to personal mastery as a profound testimony to the limits of individual power to control not only one’s future but also one’s own body (Mirowsky 1995; Pudrovska 2010a; Shaha et al. 2008).

Given the importance of control beliefs for health and well-being, surprisingly little is known about the extent to which a cancer diagnosis undermines personal mastery in midlife and old age. This gap is especially glaring in the context of unprecedented population trends that are changing the demography of cancer in the United States. Cancer incidence increases exponentially after age 50 and slows down slightly only at ages 85 and older (Howlader 2017). Cancer is the leading cause of death for men aged 60 to 74 years old and women aged 50 to 79 years old (Howlader 2017). At the same time, midlife and late life are the life course stages with progressively lower levels of mastery, which peaks in one’s 50s and then declines steadily (Mirowsky 1995; Mirowsky and Ross 2007). This segment of the U.S. population with the highest risk of cancer and the lowest levels of mastery has grown dramatically in recent decades and will continue to expand rapidly (Vincent and Velkoff 2010).

Moreover, the demographic composition of the U.S. cancer survivors is shifting toward greater numbers of racial/ethnic minorities reflecting an unprecedented growth of older non-white population groups (Vincent and Velkoff 2010). There is a conspicuous lack of studies exploring distinct and heterogeneous effects of cancer on mastery by gender and race in an increasingly diverse aging population. Mastery is influenced by objective constraints and opportunities of the social structure and is strongly patterned by socioeconomic status (SES), gender, and race (Mirowsky 1995; Mirowsky and Ross 1986). Compared to women and racial minorities, men and whites tend to have higher levels of mastery, reflecting their advantage in socioeconomic resources, power, and status (Mirowsky and Ross 1986; Pearlin et al. 1981). In contrast, due to dependency, restricted opportunities, and discrimination, women and African Americans are less likely to believe in their personal ability to control their lives (Mirowsky et al. 1996; Mirowsky and Ross 1986).

Our goal is to expand current knowledge of physical and psychological consequences of cancer by exploring the link between cancer-related bodily changes and personal mastery at the intersection of gender and race. Using the 2006-2014 data from the Health and Retirement Study (HRS), we explore changes in personal mastery after the onset of a new cancer among adults aged 50 to 85 years old. We further examine the physical burden of cancer, including incontinence, fatigue, pain, and decreased strength and stamina, as a mechanism mediating the effect of cancer on mastery. Finally, we explore gender and race differences by conducting between-gender comparisons (white men vs. white women and black men vs. black women) and within-gender comparisons (white men vs. black men and white women vs. black women).

An Integrative Theoretical Framework

We integrate structural theories of health stratification and social constructionist approaches to gender and health into a theoretical framework that links individual-level outcomes at the intersection of gender and race to the macro-level social context and the meso-level social psychological processes (Schnittker and McLeod 2005). Structural theories focus on the health benefits of social advantage, whereas the social constructionist approach focuses on the health costs of social advantage (Courtenay 2000; Hill and Needham 2013). Therefore, our framework allows to generate competing—although not necessarily mutually exclusive—predictions about how the fundamental dimensions of social hierarchy shape the effect of cancer on mastery.

Structural Theories of Health Stratification

The fundamental cause theory (Link and Phelan 1995; Phelan and Link 2015) views SES and systemic racism as fundamental causes that generate and sustain health inequality by facilitating or constraining access to material, psychosocial, and cognitive resources that protect health under widely divergent circumstances. Moreover, a structural approach to health stratification postulates that the aggregation of multiple disadvantaged statuses is particularly detrimental to physical and mental health, whereas multiple advantaged positions confer superior health benefits (McGuire and Reskin 1993; Rosenfield 2012). In current cohorts of older adults, education, wealth, income, and other valued resources are the highest among white men and lowest among black women (Federal Interagency Forum on Aging-related Statistics 2016).

Among people diagnosed with cancer, white men’s higher SES and advantage in health-enhancing resources allow them to obtain more definitive and aggressive treatments and better follow-up care (Drabe et al. 2016; Martinez et al. 2014). As a result, white men often experience improved physical functioning, reduced side effects, and a better prognosis relative to women and racial minority patients (Drabe et al. 2016; Powe et al. 2007; Ravi et al. 2014). White women and especially black men and black women are more likely to have poorer functional health after cancer treatment, more severe therapy-related symptoms and side effects, and a greater and more persistent physical burden (Fu et al. 2009; Martinez et al. 2014; Powe et al. 2007). Because physical burden undermines the sense of control (Mirowsky 1995), one might expect lower mastery in cancer patients with worse physical symptoms and higher mastery in patients with better physical symptoms. At the intersection of race and gender, white men are expected to have the highest mastery and black women the lowest. Collectively, these theoretical arguments and empirical evidence suggest the between- and within-gender components of Hypothesis 1:

Hypothesis 1.1. Between-gender: White men will have a lower physical burden of cancer than white women, which in turn will be related to higher mastery in white men. In contrast, the difference in SES and the physical burden of cancer between black men and black women will be smaller due to less pronounced gender differences in resources among African Americans.

Hypothesis 1.2. Within-gender: White men will have a lower burden of physical symptoms and better functional health than black men, reflecting white men’s advantage in SES and health-enhancing resources, which in turn will be related to higher mastery in white men than black men. The race gap in mastery will be significantly greater among men than among women.

Social Constructionist Theories of Health Stratification

According to the social constructionist approach, gender is “a set of socially constructed relationships which are produced and reproduced through people’s actions” (Gerson and Peiss 1985:327). Men and women construct (“do”) masculinity and femininity recurrently in everyday interaction with others (Connell 1995; Courtenay 2000; West and Zimmerman 1987). The physical body is a crucial resource for doing masculinity and femininity (Lee and Owens 2002; Messerschmidt 2009; Oliffe 2006). The social constructions of the feminine body prioritize physical attractiveness,whereas masculinity ideals emphasize physical strength (Halliwell and Dittmar 2003; Toscano et al. 2016). These cultural norms affect men’s and women’s self-construals of their bodies even at older ages (L. H. Clarke and Griffin 2008; Halliwell and Dittmar 2003; Hilário 2015). Studies among older adults show that women are concerned about a changing appearance and men fear a loss of bodily control and functionality (Halliwell and Dittmar 2003). Men’s social status and dominance depend on a fully functional and physically formidable body, yet women’s large size, including strength and height, is contrary to culturally normative visual cues of femininity (Rayner, Pyett, and Astbury 2010; Sell, Tooby, and Cosmides 2009; Toscano et al. 2016).

Because the idealized masculine body is strong, muscular, disciplined, and invincible (Calasanti, Pietilä, and Ojala 2013; Lukaszewski et al. 2016), cancer symptoms and side effects are potentially more threatening to the performance of culturally defined masculinity than femininity. Incontinence, pain, weakness, and fatigue undermine men’s ability to comply with the masculine bodily ideals (Burns and Mahalik 2006; Ervik and Asplund 2012; McCaughan et al. 2012). In contrast, women’s bodies are expected to be fragile, svelte, and structurally weaker than men’s bodies (McCaughan et al. 2012; Rayner et al. 2010). Male cancer patients react more adversely than women to a visible lack of toughness and vigor and are more reluctant than women to appear in public if their bodies look physically weak or frail (Calasanti et al. 2013; Solimeo 2008). The physical burden of cancer can undercut men’s social significance and diminish their masculine self-perceptions (Ervik and Asplund 2012; Lukaszewski et al. 2016; Pudrovska 2010b; Wall and Kristjanson 2005). In turn, an actual and perceived loss of social status and a threat to the masculine identity attenuate the sense of personal control among men (Anderson et al. 2012; Lachman and Weaver 1998; McLeod 2015).

Body in Dominant and Marginalized Masculinities

The notion of masculinity in the U.S. culture is not monolithic but rather comprises multiple masculinities, with white heterosexual middle-class masculinity being the dominant gender construction that emphasizes strength, authority, independence, and control (Courtenay 2000; Oliffe 2006). Although black men endorse many traditional norms of hegemonic masculinity, including physical strength, toughness, risk taking, and economic achievement, they also express broader and more flexible views of manhood relative to white men (Gilbert et al. 2016; Hooker et al. 2012). Masculinity ideals of middle-aged and older African American men are consistent with taking care of one’s health, accepting age-related changes, and adapting to physical problems (Griffith et al. 2015; Hooker et al. 2012). Among older black men, physical declines related to aging were viewed through the prism of acceptance, mental repose, and relaxation (Hooker et al. 2012). If the physical body is less central to black men’s constructions of masculinity, black men may be less likely than their white peers to perceive cancer-related bodily symptoms as a status threat (Gilbert et al. 2016; Griffith 2015). Therefore, it is possible that cancer symptoms are less threatening to the marginalized masculinity of black men relative to the hegemonic masculinity of white men. Collectively, these theoretical arguments and empirical evidence suggest the between- and within-gender components of Hypothesis 2:

Hypothesis 2.1. Between-gender: Physical cancer symptoms will have a stronger effect on mastery in white men than white women because this effect is magnified among white men through a potential threat to the self-concept, reflecting the psychological cost of white masculinity relative to femininity. In contrast, we do not expect a significant gender difference among black adults.

Hypothesis 2.2. Within-gender: A decline in mastery after a cancer diagnosis will be greater in white men than black men, reflecting the psychological cost of white masculinity relative to black masculinity. In contrast to men, we hypothesize that there will be no significant race difference among women.

Data and Methods

Data

The Health and Retirement Study (HRS) is a longitudinal study of men and women aged 50 and older in the United States. The study began in 1992 with individuals born between 1931 and 1941 and their spouses. In the initial wave, in-home face-to-face interviews were conducted with 7,608 households (12,652 individuals). Participants have been interviewed approximately every two years since then. Starting in 2006, the HRS has added an enhanced face-to-face (EFTF) interview that includes a self-administered psychosocial questionnaire. Between 2006 and 2014, two randomly selected 50 percent panels participated in the EFTF interview in alternating waves. Panel A provided data in 2006, 2010, and 2014, whereas Panel B provided data in 2008 and 2012. Our focal variable of interest—personal mastery—is included only in the psychosocial questionnaire. Therefore, our analytic subsample is restricted to non-Hispanic black and white men and women who participated in the EFTF interview at least twice and had never been diagnosed with cancer at baseline. Of these individuals, 279 white men, 301 white women, 47 black men, and 52 black women were diagnosed with a new cancer by the follow-up. The numbers of participants in each group without cancer at the follow-up are 2,196, 3,463, 219, and 587, respectively. A table detailing two panels and how they were pooled is provided in the Appendix.

Sample weights and sample attrition

To adjust for the HRS multistage area probability sample design with oversamples of African Americans, Hispanics, and Florida residents, we use overall household-level and individual-level sampling weights for each wave. Additionally, we apply separate individual-level sample weights developed for the psychosocial questionnaire component in 2006, 2008, 2010, 2012, and 2014. Finally, we create an inverse-probability of censoring weight to account for right censoring due to attrition. The final weights are obtained by multiplying core sample weights, the EFTF interview weight, and the censoring weight.

Variables

All variables were assessed at baseline (t₁) and follow-up (t₂). The cancer group did not have cancer at t₁ but was diagnosed with cancer between t₁ and t₂. Thus, t₁ measures for cancer patients reflect pre-cancer characteristics.

The measure of the new cancer onset is based on participants’ self-reports of having been diagnosed by a doctor with a new cancer (excluding minor skin cancers) or malignant tumor since the previous wave. The new cancer variable is coded 1 for people who reported a new cancer at the follow-up and 0 for people who have never been diagnosed with cancer.

Mastery is assessed with 10 items (Lachman and Weaver 1998; Pearlin and Schooler 1978), such as “I often feel helpless in dealing with the problems of life,” “What happens in my life is often beyond my control,” “There is really no way I can solve the problems I have,” “I can do just about anything I really set my mind to,” “Whether or not I am able to get what I want is in my own hands,” and “What happens to me in the future mostly depends on me.” Participants were asked about their extent of agreement or disagreement with each statement. Response categories are 1 = strongly disagree, 2 = somewhat disagree, 3 = slightly disagree, 4 = slightly agree, 5 = somewhat disagree, and 6 = strongly agree. Negatively worded items were reverse-coded. All items were averaged to create a scale (α = .88).

Cancer-related physical symptoms

Incontinence is coded 1 if participants reported having lost any amount of urine beyond their control in the last month. Pain is assessed with three mutually exclusive categories: no pain, pain that does not limit usual activities, and pain that makes it difficult to do usual activities. Fatigue is coded 1 if a person reports any persistent or troublesome severe fatigue or exhaustion. Strength-related limitations are based on three self-reported difficulties because of a health problem: climbing several flights of stairs without resting, lifting or carrying weights over 10 pounds, and pulling or pushing large objects.

The number of chronic illnesses other than cancer is measured as a sum of self-reported diagnoses of stroke, cardiovascular disease, diabetes, and hypertension. Depressive symptoms are assessed as a sum of eight items from the Center for Epidemiologic Studies Depression (CES-D) scale (α = .79). Prior to the summation, each item was coded 1 if a person experienced a given symptom much of the time and 0 otherwise.

Sociodemographic characteristics include gender (coded 1 for women and 0 for men), race (1 = black, 0 = white), and age (coded in years at the beginning of each interview). Socioeconomic status and resources include education (high school or less, some college, bachelor’s or higher degree), the longest occupation (managerial, professional, lower-status), employment status (employed full- or part-time, retired, out of the labor force), wealth, non-mortgage debt, and public and private health insurance plans. Family statuses reflect marital status (coded 1 for persons who are married or partnered) and the total number of living children.

Analytic Approach

First, we obtain summary statistics for all study variables by gender and race using one-way ANOVA for continuous variables and χ² test for dummy variables (Table 1).

Table 1.

Descriptive Statistics: The Health and Retirement Study, 2006-2014.

Variables	Men with Cancer		Men without Cancer		Women with Cancer		Women without Cancer
	White	Black	White	Black	White	Black	White	Black
	n = 279	n = 40	n = 2,196	n = 219	n = 301	n = 52	n = 3,463	n = 587
Mastery t₁	4.82^a (1.04)	4.83^b (1.20)	4.89^a (1.01)	4.84^b (1.15)	4.80^c (1.01)	4.55^a,b,c (1.28)	4.80 (1.06)	4.74^a,b (1.15)
Mastery t₂	4.55^a,c (1.17)	4.64^a,b,c (1.25)	4.74^a,b (1.10)	4.73^a (1.24)	4.58^a (1.17)	4.54^a,b (1.11)	4.66^a,b (1.12)	4.68^a (1.21)
Mastery t₂ – t₁	−.27^a,b,c	−.19^a,b,c	−.15^c	−.11^b,c	−.22^a,b,c	−.01^a,b,c	−.14^c	−.06^b,c
Incontinence t₁	.114	.151	.116	.128	.352	.308	.344	.163
Incontinence t₂	.276	.325	.156	.137	.412	.385	.383	.201
Pain disruptive t₁	.172	.233	.177	.168	.235	.269	.229	.241
Pain disruptive t₂	.226	.325	.184	.164	.286	.298	.244	.257
Fatigue t₁	.107	.075	.106	.078	.171	.251	.185	.155
Fatigue t₂	.165	.175	.134	.091	.282	.365	.197	.178
Lift t₁	.139	.225	.101	.183	.222	.403	.249	.485
Lift t₂	.270	.341	.148	.242	.335	.576	.306	.497
Push t₁	.143	.275	.129	.228	.295	.481	.291	.402
Push t₂	.290	.325	.183	.315	.405	.519	.328	.465
Climb stairs t₁	.321	.343	.257	.316	.443	.425	.387	.478
Climb stairs t₂	.404	.425	.328	.391	.481	.577	.437	.512
Education
Bachelor’s or higher	.358	.100	.329	.114	.252	.153	.211	.132
Longest occupation
Professional/executive	.423	.153	.402	.145	.313	.220	.336	.183
Lower status	.577	.847	.598	.856	.688	.779	.664	.817
Household wealth t₁	8.29 (6.51)	5.98 (6.55)	5.22 (6.98)	4.11 (6.09)	7.79 (6.43)	5.32 (6.22)	5.13 (6.79)	3.32 (5.79)
Debt amount t₁	.56 (4.88)	3.60 (5.44)	1.04 (5.07)	3.03 (5.56)	.99 (5.01)	1.95 (5.31)	1.01 (5.01)	2.23 (5.33)
Age t₁	69.16 (8.69)	71.22 (8.71)	68.81 (10.56)	70.79 (10.54)	67.40 (8.36)	69.54 (8.50)	65.33 (7.71)	67.27 (7.93)
Married t₁	.849	.775	.833	.689	.661	.442	.638	.381
Number of children t₁	2.81 (1.73)	3.54 (2.59)	2.88 (1.88)	3.63 (2.49)	2.99 (1.92)	2.87 (2.29)	2.92 (1.92)	3.41 (2.39)
Chronic illnesses t₁	1.93 (1.34)	2.11 (1.75)	1.88 (1.31)	1.81 (1.32)	1.91 (1.31)	2.28 (1.43)	1.90 (1.30)	2.30 (1.30)
Depression t₁	.86 (1.59)	1.02 (1.38)	.85 (1.50)	.84 (1.83)	1.19 (1.71)	2.05 (2.31)	1.20 (1.82)	1.61 (1.99)
Employed t₁	.253	.237	.359	.372	.273	.351	.287	.308
Retired t₁	.716	.672	.613	.565	.621	.458	.585	.565
Not in labor force t₁	.031	.090	.028	.063	.106	.192	.128	.127

Note. Each cell contains means (standard deviations) for continuous variables and proportions for categorical variables.

Significant differences (at least at .05 level) between persons with and without cancer.

Significant differences (at least at .05 level) between genders.

Significant differences (at least at .05 level) within genders.

Second, we use the matching models to estimate the effect of cancer on mastery by comparing individuals with cancer (the “treatment group”) to matched individuals without cancer (the “control group”), who are as similar as possible on all observable characteristics at baseline with the exception of a new onset cancer at the follow-up (Table 2). Within each group (white men, white women, black men, black women), individuals with cancer are matched on baseline (pre-cancer) variables to people who have never been diagnosed with cancer. We combine exact matching on key covariates with propensity score matching. All variables used for matching are shown in the note to Table 2. Our propensity score matching method is 1:2 nearest neighbor matching. Each treated individual is matched to two control individuals with propensity scores that are the closest to the score of a given treated individual. Matched persons in the treatment and control groups are compared to obtain the average treatment effect for the treated (ATT), namely, a change in mastery from the baseline to the follow-up in persons with cancer relative to their matched controls. A hidden bias arising from unobserved heterogeneity is a potential weakness of propensity score matching estimators. We address this problem with a sensitivity analysis based on the bounding approach, which simulates how strongly an unmeasured variable must influence selection into treatment to undermine conclusions of propensity score matching (Ichino, Mealli, and Nannicini 2008). An unobserved binary confounder is simulated in the data based on its joint distribution with the treatment variable, and a comparison of the simulated ATT and the actual ATT indicates the degree of robustness of the latter.

Table 2.

The Effect of Cancer on Mastery among Individuals with Cancer and Matched Controls without Cancer: The Health and Retirement Study, 2006-2014.

Within-group Comparisons
ATT (white men)
n (cancer) = 279	−.173*
n (controls) = 558	(.087)
ATT (white women)
n (cancer) = 301	−.128
n (controls) = 602	(.084)
ATT (black men)
n (cancer) = 47	−.075
n (controls) = 94	(.209)
ATT (black women)
n (cancer) = 52	−.102
n (controls) = 104	(.093)
Between-group Comparisons
Between gender
Δ white = ATT (white men) – ATT (white women)		−.045*** (.008)
Δ black = ATT (black men) – ATT (black women)		.026(.031)
Within gender
Δ men = ATT (white men) – ATT (black men)		−.098* (.046)
Δ women = ATT (white women) – ATT (black women)		−.025(.029)

Note. Each cell contains average treatment effects for the treated (ATT) and standard errors in parentheses. Within each group, persons with cancer and controls without cancer are matched on the following baseline (pre-cancer) variables: education, marital status, and employment and retirement status (via exact matching) and mastery, age, the longest occupation, the number of chronic illnesses, and depressive symptoms (via propensity score matching).

p < .05. ***p < .001.

Third, we use multiple regression models to estimate the mediating effects of physical symptoms and interactive effects of cancer with gender and race. To formally test whether physical symptoms mediate the effect of cancer on mastery (Table 3), we fit the model for the outcome (Equation 1) and the model for the mediator (Equation 2):

\begin{array}{l} E (Y_{i t_{2}}) = γ_{1} (C_{i t_{2}} - C_{i t_{1}}) + γ_{2} (M_{i t_{2}} - M_{i t_{1}}) \\ + γ_{3} (Y_{i t_{1}}) + γ_{4} (X_{i t_{1}}) + γ_{5} (X_{i}) + ε_{i t} \end{array}

\begin{array}{l} g {E (M_{i t_{2}} - M_{i t_{1}})} = β_{1} (C_{i t_{2}} - C_{i t_{1}}) \\ + β_{2} (Y_{i t_{1}}) + β_{3} (X_{i t_{1}}) + β_{4} (X_{i}) + ε_{i t} . \end{array}

Table 3.

Indirect (Mediated) Effects of Cancer on Mastery: The Health and Retirement Study (HRS), 2006-2014.

Mediators (M)^a	White Men			White Women			Black Men			Black Women
	Model 1			Model 2			Model 3			Model 4
	(a)	(b)	(c)	(a)	(b)	(c)	(a)	(b)	(c)	(a)	(b)	(c)
	Cancer → M	M → Mastery	Indirect Effect^b	Cancer → M	M → Mastery	Indirect Effect^b	Cancer → M	M → Mastery	Indirect Effect^b	Cancer → M	M → Mastery	Indirect Effect^b
Incontinence	.113***	−.125***	−.014***	.016	−.102**	−.002	.366**	−.069	−.025*	.258**	−.079*	−.020**
Pain^c	.051*	−.257***	−.013**	.029	−.233***	−.007*	.077*	−.116*	−.009	.019	−.058	−.001
Fatigue	.072	−.138***	−.010*	.145**	−.112***	−.016***	.161**	−.108*	−.017*	.288***	−.065	−.019**
Lifting	.094***	−.152***	−.014***	.035*	−.134***	−.005	.167*	−.032	−.005	.077*	−.036	−.003
Pushing	.105***	−.129***	−.014***	.056**	−.105***	−.006*	.021	−.025	−.001	.018	−.029	−.001
Climbing	.072***	−.135***	−.010*	.037*	−.081***	−.003	.069*	−.061	−.004	.094**	−.089**	−.008

Note. Each cell shows standardized coefficients (β). All models control for baseline mastery, survey year, HRS panel (A or B), age, education, employment and retirement status, the longest occupation, the number of chronic illnesses, and depressive symptoms.

M refers to a given mediator.

Indirect effects = (Cancer → M) × (M → Mastery).

Pain refers to pain that interferes with activities.

p < .05. **p < .01. ***p < .001.

In both equations, the subscript i denotes an individual, the subscripts t₁ and t₂ denote the baseline and follow-up measurements, $C_{i t_{2}} - C_{i t_{1}}$ is the new cancer onset, $M_{i t_{2}} - M_{i t_{1}}$ is the change in a given physical symptom (mediator), $Y_{i t_{1}}$ is baseline (pre-cancer) mastery, and $X_{i t_{1}}$ and $X_{i}$ are baseline and time-invariant covariates, respectively. The direct effect of cancer on mastery equals γ₁, the indirect effect of cancer on mastery (mediated by a given physical symptom) equals γ₂β₁, and the total effect of cancer net of control variables equals γ₁ + γ₂β₁.

To test for significant between-gender differences, we estimate models with an interaction between cancer and gender among white men and women (Table 4). To explore significant within-gender differences, we include an interaction between cancer and race among black and white men (Table 5).

Table 4.

Interactive Effects of Cancer and Gender among White Men and Women: The Health and Retirement Study (HRS), 2006-2014 (N = 6,239).

Variables	(1)	(2)	(3)	(4)
Variables	b (SE)	b (SE)	b (SE)	b (SE)
Cancer	−.140*** (.042)	−.147*** (.042)	−.140*** (.043)	−.094* (.042)
Female	−.032(.026)	−.036(.027)	−.012(.028)	.063* (.028)
Cancer × female	.145* (.068)	.156* (.069)	.144* (.071)	.128* (.065)
Debt		−.010*** (.003)
Net worth		.012*** (.002)
Married or partnered			.010(.029)
Number of living children			.009(.007)
Incontinence				−.057* (.028)
No pain (reference)
Pain does not interfere				.018(.037)
Pain interferes				−.053(.035)
Fatigue				−.248*** (.035)
Difficulties with activities
Lifting				−.144*** (.037)
Pushing				−.154*** (.035)
Climbing stairs				−.081** (.029)
Constant	3.308	3.426	3.297	3.339
Adjusted R²	.235	.240	.235	.260

Note. Each cell contains unstandardized regression coefficients (b) and standard errors in parentheses. All models control for baseline mastery, survey year, HRS panel (A or B), age, education, employment and retirement status, the longest occupation, the number of chronic illnesses, and depressive symptoms.

p < .05. **p < .01. ***p < .001.

Table 5.

Interactive Effects of Cancer and Race among White and Black Men: The Health and Retirement Study (HRS), 2006-2014 (N = 2,733).

Variables	(1)	(2)	(3)	(4)
Variables	b (SE)	b (SE)	b (SE)	b (SE)
Cancer	−.135* (.044)	−.146* (.043)	−.135* (.044)	−.093(.043)
Black	−.080(.052)	−.082(.052)	−.082(.052)	−.081(.051)
Cancer × black	.191* (.096)	.194* (.095)	.189* (.095)	.155(.094)
Debt		−.008* (.004)
Net worth		.011** (.004)
Married or partnered			.029(.049)
Number of living children			.008(.010)
Incontinence				−.060(.052)
No pain (reference)
Pain does not interfere				.045(.057)
Pain interferes				−.050(.057)
Fatigue				−.294*** (.060)
Difficulties with activities
Lifting				−.108(.065)
Pushing				−.112(.059)
Climbing stairs				−.154*** (.045)
Constant	3.209	3.297	3.176	3.241
Adjusted R²	.223	.224	.221	.244

p < .05. **p < .01. ***p < .001.

Finally, we use the following formula to obtain statistics for formal tests of between-group differences in means and proportions, ATTs, and mediating pathways (Paternoster et al. 1998):

\frac{E_{1} - E_{2}}{\sqrt{s^{2} (E_{1}) + s^{2} (E_{2})}},

where E₁ and E₂ are estimates for each group under comparison and s² is the sampling variance of the estimates adjusted for the nonindependence of the two samples.

Results

Descriptive Statistics

Table 1 shows descriptive statistics by gender, race, and cancer status. Comparisons of persons with and without cancer within each group reveal that the decline in mastery was significantly more pronounced among white men with cancer than among their peers without cancer (–.27 vs. –.15, p < .05). Similarly, white women with cancer (–.22) and black men with cancer (–.19) experienced a significantly greater decline in mastery than white women without cancer (–.14) and black men without cancer (–.11). In contrast, among black women, mastery declined less in the cancer group than the non-cancer group (–.01 vs. –.06). Among persons with cancer, white men had a more pronounced decline in mastery than white women in between-gender comparisons (–.27 vs. –.22, p < .05) and a much greater decline than black men in within-gender comparisons (−.27 vs. –.19, p < .001). Black women with cancer had the lowest decline in mastery of all cancer groups. A comparison of physical symptoms indicates that incontinence after the onset of cancer increased substantially for black and white men, and this increase was significantly greater in black men than white men (p < .01). Among black and white women, the difference in incontinence after cancer was much smaller than among men. White men with cancer had the lowest incontinence prevalence than all other groups. After the onset of cancer, pain that disrupts activities and chronic fatigue were the highest in black men and women and the lowest in white men. White men started with fewer difficulties with lifting, pushing, andclimbing and experienced the smallest increase in these symptoms after the onset of cancer. Among persons with cancer, white men had higherlevels of socioeconomic resources than other groups, including higher levels of collegeeducation, more prestigious occupations, higher net worth, and lower debt. In contrast, black womenwere the most disadvantaged socioeconomically.

Matching Models

Table 2 shows within- and between-group differences in the effect of cancer on mastery. Within-group comparisons are presented in Model 1. The difference in mastery between white men with cancer and their matched controls is large and statistically significant, as indicated by the ATT of –.173 (p < .05). In contrast, the respective difference is smaller and not statistically significant among white women (ATT = –.128, p > .05). The ATTs among black men (–.075) and black women (–.102) are small relative to whites and far from statistical significance. Thus, white men with cancer experienced the largest decline in mastery relative to their matched controls without cancer, whereas black men with cancer had the smallest decline in mastery. A sensitivity analysis (not shown in the table) indicates that the ATT among white men is robust to potential bias from unobserved heterogeneity. The original ATT (–.173) is very similar to the ATT with simulated unobserved covariates (–.169, p < .05).

The analysis presented in Model 2 formally tests between- and within-gender differences based on the within-group ATTs reported in Model 1. Between-gender comparisons (within the same race) indicate that the adverse effect of cancer on mastery is significantly stronger in white men relative to white women (ATT [white men] – ATT [white women] = –.045, p < .001). In contrast, the difference in ATTs between black men and black women is small and not statistically significant. Within-gender comparisons reveal that white men’s ATT is significantly greater than black men’s ATT (ATT [white men] – ATT [black men] = –.098, p < .05), whereas there is no significant race difference among women. Collectively, the between- and within-group comparisons suggest that the decline in mastery after the onset of cancer is substantially more pronounced in white men than in white women and black men, whereas differences among other groups are smaller and not statistically significant.

Indirect (Mediated) Effects of Cancer on Mastery

The formal test of physical symptoms as mediators of the effect of cancer on mastery is shown in Table 3. Within each group, column a shows the effect of cancer on each bodily change, column b shows the effect of the body on mastery, and column c shows the indirect effects of cancer on mastery mediated by each bodily symptom. All effects are given as standardized coefficients to facilitate comparisons across groups; therefore, the indirect effect in column c is the product of the two effects in columns a and b. White women experienced a lower increase in physical symptoms after the onset of cancer than white men, with the exception of fatigue. Compared to whites, the physical symptoms burden was larger in black men and black women, especially incontinence, fatigue, and pain. Yet, in contrast to the weaker effect of cancer on symptoms, the effect of cancer on mastery was stronger in white men and women relative to black men and women. Among whites, the effect of cancer on mastery was significantly stronger in men than women. All physical symptoms are significant mediators of the adverse effect of cancer on mastery among white men, as indicated by large and statistically significant indirect effects in Model 1. A decomposition of the indirect effect in columns a and b reveals that the effect of cancer on symptoms is smaller than the effect of symptoms on mastery. Thus, the large mediating effect of physical symptoms in white men is primarily driven by a particularly strong impact of physical symptoms on the decline in mastery. Bodily symptoms have a disproportionately adverse effect on mastery among white men compared to white women and especially to black men and black women.

Interactive Effects of Cancer, Gender, and Race on Mastery

Between-gender comparisons

The effect of cancer on mastery among black individuals (not shown in the table) does not differ significantly by gender (b = –.089, SE = .244, p > .05). Therefore, Table 4 presents a statistically significant interactive effect of cancer and gender among white men and women. The main effect of cancer in Model 1 applies to men only (female = 0) and reveals a significantly greater decline in mastery among white men with cancer relative to white men without cancer (b = .140, SE = .042, p < .001). The small and nonsignificant coefficient for the main effect of gender in Model 1 indicates that the change in mastery among persons without cancer (cancer = 0) does not differ significantly between men and women. The statistically significant cancer × female interaction term suggests that cancer has a weaker effect on the decrease in mastery in white women than white men (b = .145, SE = .068, p < .05). In other words, among white cancer patients, the decline in mastery is more pronounced in men than women. As shown in Models 2 and 3, financial resources and family statuses do not mediate either main or interactive effects of cancer and gender. In contrast, after physical symptoms are added in Model 4, the interaction coefficient declines from .145 in Model 1 to .128 in Model 4 (by 11.7 percent). Similarly, the main effect of cancer (the change in mastery among men with cancer relative tomen without cancer) declines from –.140 to −.094 (by 32.8 percent). This reduction in effect sizes reflects both a higher increase in physical symptoms and a stronger effect of physical symptoms on mastery in white men than white women (shown in Models 1 and 2 of Table3). A significantly greater decline in mastery after the onset of cancer in white men than white women is explained by a somewhat greater physical burden and a much greater adverse effect of physical burden on mastery in white men.

Within-gender comparisons

Table 5 shows an interaction between cancer and race among men. The respective interaction term among women (not shown in the table) is not significant (b = .066, SE = .157, p > .05). The main effect of cancer in Model 1 applies to white men only (black = 0) and suggests that the decline in mastery is significantly greater in white men who were diagnosed with cancer by the follow-up compared to white men who did not develop cancer. The interactive effect of cancer and race indicates that the adverse effect of cancer on mastery is significantly stronger in white men than black men (b= .191, SE = .096, p < .05). The main and interactive effects of cancer are not explained by financial resources and family statuses (Models 2 and 3). In contrast, when the physical symptoms are included in Model 4, the interaction coefficient decreases by 18.8 percent (from .191 to .155). This reduction of the race gap in mastery among men reflects a disproportionately stronger adverse effect of physical symptoms on mastery in white men than black men despite white men’s lower physical burden of cancer relative to their black peers. Although the effect of cancer on physical symptoms is much smaller in white men than black men (Models 1a and 3a in Table 3), cancer-related bodily changes undermine mastery more dramatically in white men (Models 1b and 3b in Table 3).

Discussion

Personal mastery reflects “the extent to which one regards one’s life chances as being under one’s own control in contrast to being fatalistically ruled” (Pearlin and Schooler 1978:211). This study explores changes in physical symptoms and personal mastery after the onset of cancer among white men, black men, white women, and black women. Using within- and between-group comparisons, we find that white men with cancer experienced the greatest decline in mastery than all women and black men, despite white men’s most advantaged material resources and favorable cancer-related symptoms (a somewhat higher physical burden relative to white women but a much lower physical burden relative to black men and women). White men—a socially advantaged group of cancer patients with the largest access to resources and information—are particularly vulnerable to the cancer-related decline in personal mastery relative to white women, black men, and black women. In contrast, black women exhibit the smallest decline in mastery despite their highest physical burden and lowest levels of socioeconomic resources. White men’s decline in mastery after the onset of cancer reflects the fact that physical symptoms undermine mastery disproportionately stronger in white men than all other groups. White men start with the highest mastery prior to the cancer diagnosis but also experience the largest decline in mastery after the diagnosis due to a particularly strong effect of body changes on mastery. In other words, white men’s decrease in mastery is higher than would be expected based on the increase in their cancer symptoms.

By expanding current knowledge of the ways in which physical illness affects psychological outcomes, our study can potentially contribute to three prominent themes in the sociology of mental health. First, our findings emphasize the importance of integrating the structural and cultural approaches to psychological well-being. Incorporating cultural beliefs that affect the “social psychological processes” of status and identity can provide new insights into “unexpected patterns of mental health inequalities” (McLeod 2015:149), including our finding of white men’s disadvantage in mastery that cannot be fully explained by the leading structural theories of health stratification. Second, social stress research has typically viewed psychosocial resources as a relatively stable property of the socially advantaged positions (Mirowsky and Ross 1986; Pearlin et al. 1981; Phelan and Link 2015). Yet, our findings underscore the importance of adopting a dynamic view because psychosocial resources are not immutable and may be threatened and depleted by stressors and life transitions (Pudrovska et al. 2005). Third, existing studies of health disparities have overwhelmingly relied on binary comparisons between monolithic gender or race categories (Courtenay 2000; K. M. Keyes, Barnes, and Bates 2011; Meyer, Schwartz, and Frost 2008; Phelan and Link 2015; Powe et al. 2007; Pudrovska 2010b). This broad focus obscures heterogeneity within each category and boundary permeability between them. We emphasize the need for within- and between-group comparisons to uncover complex mental health outcomes at the intersection of multiple axes of stratification. In the three subsequent sections, we provide a more detailed discussion of each of these potential contributions.

A Contextual Framework Integrating Structural and Cultural Approaches

The effects of cancer on physical symptoms are consistent with the structural theories of health disparities predicting the health benefits of health-enhancing resources of social advantage (Phelan and Link 2015). In contrast, the effect of cancer on mastery is consistent with the social constructionist perspective on gender and health predicting the health costs of white hegemonic masculinity. Whereas earlier research on the stress process suggested that men are less vulnerable to stressors than women because men have more resources to cope with stressors, more recent studies argue that gender differences in psychological adjustment are better explained by the fact that the same stressors may have different meanings and relevance for men and women (Caputo and Simon 2013; Simon 1992). Identity-relevant stressors that threaten core beliefs about one’s self-concept and self-worth exert a particularly detrimental psychological impact relative to other stressors (Caputo and Simon 2013; McLeod 2015). Adjustment to physical declines in late life is profoundly influenced by gendered cultural norms (L. H. Clarke and Griffin 2008; Hilário 2015). Women are more likely than men to accept the loss of bodily abilities. In contrast, physical losses are more disruptive and demoralizing for men because they contradict masculinity ideals of strength, virility, and bodily control (Hilário 2015). We argue that the cultural norms of masculinity and femininity imbue cancer and its symptoms with gender-specific cultural meanings, which affect the differential relevance of cancer-related bodily changes for men’s and women’s self-concepts. Cancer is a powerful threat to masculinity, particularly the hegemonic masculinity of white men (Pudrovska 2010b). More than any other chronic illness, cancer and its treatment threaten men’s ability to display bodily ideals, including strength, muscularity, toughness, and physical invincibility and discipline. The physical burden of cancer might pose a profound threat to white men’s cultural privilege by disrupting the masculine body as a critical resource for demonstrating masculinity. Failure to comply with cultural expectations of the idealized masculine body can reduce a man’s status in masculine hierarchies and a man’s own self-perceptions of manhood. In turn, threats of social devaluation and status loss are particularly detrimental to the sense of control of high-status people (Anderson et al. 2012).

A Dynamic View of Psychosocial Resources

Mastery has been consistently viewed as a crucial psychosocial pathway through which social advantage improves physical health and psychological well-being (Pearlin et al. 1981; Pudrovska et al. 2005; Seeman et al. 2014). Existing research has primarily focused on the undeniable benefits of psychosocial resources and thus has espoused a static view of resources as relatively stable attributes of socially advantaged positions. Yet, it is equally important to recognize the dynamic and contextual nature of coping resources that are not immutable and may be threatened or depleted by stressors. While high mastery is indeed an effective coping resource among older adults, it can be undermined by chronic stressors, and diminished mastery in turn increases physical and psychological vulnerability to subsequent hardships (Pearlin et al. 1981; Pudrovska et al. 2005). Our findings reveal simultaneously male advantage and disadvantage in psychosocial resources. White men exhibit the highest mastery at the baseline but also experience the steepest decline in mastery after the onset of cancer. A dynamic approach to resources allows to interpret this seeming contradiction through the prism of status threat (Goymann and Wingfield 2004; Sapolsky 2004). The psychosocial resources of high status depend on the stability of status hierarchies—whether the dominant group’s status is secure or contested and challenged (Anderson et al. 2012; Gruenewald et al. 2006). High-status people have the most beneficial psychosocial resources when their status is secure and uncontested but are the most adversely affected when their status is threatened (Goymann and Wingfield 2004; Sapolsky 2004). It is dominant individuals who suffer “the greatest sense of loss of control and predictability” in unstable hierarchies relative to subordinate group members (Sapolsky 2004:407). Collectively, these findings suggest that white men in our study experienced the greatest deterioration in mastery despite their baseline advantage due to a potential threat to the high status of masculinity precipitated by the meaning of cancer as a loss of control over one’s life and one’s body.

A Paradox of Intersectionality

Our findings about the physical burden of cancer are consistent with the double jeopardy approach positing the accumulated psychological burden of multiple disadvantages. In our study, black patients have more cancer symptoms than white patients, and black women have the most symptoms of all cancer groups. In contrast to physical symptoms, our findings about mastery are contrary to the double jeopardy predictions. We document a greater resilience to cancer in black women relative to other groups. Black women’s mastery is undermined by cancer the least, despite their lowest material resources and worst physical symptoms. A more favorable psychological adjustment to cancer among the socially disadvantaged is not consistent with structural theories of stratification that are more suited to explain vulnerability than resilience. To understand the roots of positive well-being of black women, it is important to incorporate social psychological perspectives. In particular, research on personal growth in the face of adversity suggests that racism and multiple challenges that African Americans endure on a daily basis enhance psychological hardiness and stoicism (C. L. M. Keyes 2009; Ryff, Keyes, and Hughes 2003). Psychological resilience against racial inequality and racism may itself become a protective resource for mental health of African Americans (Watkins 2012).

Further, our results indicate that within-gender differences in the effect of cancer on mastery are significantly stronger among men than women. In other words, cancer undermines mastery in white men much more than in black men. This race difference among men can be explained by the existence of multiple masculinities in the U.S. culture (Courtenay 2000). Masculinities of racial minority men are profoundly shaped by their marginalization in the mainstream society (Connell and Messerschmidt 2005; Rogers, Sperry, and Levant 2015). African American men are confronted with constant threats to their masculinity resulting from pervasive discrimination and racial adversity (Rogers et al. 2015). This marginalization may attenuate the link between masculinity and black men’s psychological well-being. In contrast, white men’s self-concepts and mental health are strongly linked to hegemonic masculinity because the norms of dominant masculinity are more favorable and beneficial to their lives (Levant and Wong 2013). Stressors that are potential or actual status threats are more detrimental to incumbents of high-status than low-status positions (Anderson et al. 2015; Taylor 2014). When individuals achieve high status, they dread its loss even more and thus experience a particularly demoralizing reaction to status threats (Anderson et al. 2015; Gruenewald et al. 2006). White men may be more vulnerable tocancer than black men, reflecting the cost of losing a highly valued status because a threat to the culturally dominant masculinity is potentially more adverse psychologically than a threat to a marginalized masculinity (Pudrovska 2010b; Schmutte et al. 2009).

Our study also reveals that between-gender differences in the effect of cancer on mastery are smaller among black than white individuals. Masculinity and femininity may be less polarized in black men and women, who generally endorse more egalitarian, less rigid, and less conventional gender beliefs than their white peers (Kane 2000; Rosenfield 2012). Conceptualizations of gender among African Americans are more interchangeable and less strictly demarcate the private sphere of femininity and the public sphere of masculinity (Kane 2000). Our findings are consistent with a recent study showing that black women exhibit better overall mental health relative to white women and all men (Rosenfield 2012). According to Rosenfield (2012), the mental health effects of multiple disadvantaged statuses at the intersection of race and gender are “paradoxical” because they refute predictions of multiple jeopardy. These paradoxes emphasize the importance of intersectionality and increased attention to multiple masculinities and femininities as well as within-group heterogeneity in addition to differences between monolithic gender categories.

Limitations and Future Directions

Our study is based on older cohorts characterized by the gender-typed division of labor, with men as primary breadwinners and women as primary homemakers and caregivers. Family and work domains of men and women have become more similar in younger cohorts (Casper and Bianchi 2002). Gender beliefs and cultural norms of masculinity have been changing as well (Pudrovska 2015). Our next step will be to explore physical and psychosocial burden of cancer in younger generations of cancer survivors. Further, we could not explicitly test the masculinity explanation in our analysis because of the lack of any relevant measures in the HRS. In general, very few data sets contain measures of gender beliefs and masculine self-concepts. Masculinity is a multidimensional construct that includes not only cultural beliefs and stereotypes but also personality traits, behaviors, and self-evaluations. Therefore, an important direction for future research will be to develop instruments that directly measure multiple dimensions of masculinity.

Further, potential gender and race biases in self-reports of cancer symptoms might potentially affect our conclusions (Dowd and Todd 2011). For example, men may be less likely than women to report embarrassing symptoms, and white men may be more reluctant than other gender × race groups to report symptoms that are incompatible with social dominance. Interestingly, however, the physical burden of cancer reported by white men in our study had a disproportionately strong effect on their mastery. If this burden was underreported and white men in fact had experienced more severe symptoms than they admitted, the true link between body and mastery would be even more adverse than we observed. Ideally, future research should incorporate objective measures of cancer-related physical sequelae, including biomarkers and reports of external observers.

A specific type and site of cancer is an important factor that contributes to the complexity of cancer experiences. Yet, the information about cancer type is highly restricted in the HRS. Ourfindings most likely provide conservative estimates of the effect of cancer on mastery because our sample is likely dominated by more prevalent and less fatal cancers. Cancers with poor prognosis, such as lung or pancreatic, may be associated with lower personal mastery than cancers with better survival rates, such as breast or prostate. Another important limitation of our study is a limited number of black cancer patients, especially black men. This group is underrepresented in research among older adults. A crucial direction for future research is to expand the focus on older race/ethnic minorities and include not only black older adults but also other race/ethnic groups.

Despite these limitations, our study expands current research by integrating structural and social psychological perspectives into a unified framework that allows us to place psychological resilience and vulnerability of cancer survivors in a larger socio-historical and cultural context. We view cancer as a profound developmental experience relevant to the identity and the self-concept that involves individual-level coping and psychological adjustment shaped by the macro-social patterns. Our findings emphasize the need for a dynamic and contextual approach to address increasing complexity and heterogeneity of health stratification processes in the U.S. aging population.

Footnotes

Appendix

References

American Cancer Society. 2016. Cancer Facts & Figures 2016. Atlanta, GA: Author.

Anderson

Cameron

Hildreth

John A.

Howland

Laura

. 2015. “Is the Desire for Status a Fundamental Human Motive? A Review of the Empirical Literature.” Psychological Bulletin 141(3):574-601.

Anderson

Cameron

Kraus

Michael W.

Galinsky

Adam D.

Keltner

Dacher

. 2012. “The Local-ladder Effect: Social Status and Subjective Well-being.” Psychological Science 23(7):764-71.

Baum

Andrew

Posluszny

Donna M.

1999. “Mapping Biobehavioral Contributions to Health and Illness.” Annual Review of Psychology 50:137-63.

Bettencourt

B. Ann

Talley

Amelia E.

Molix

Lisa

Schlegel

Rebecca

Westgate

Steven J.

2008. “Rural and Urban Breast Cancer Patients: Health Locus of Control and Psychological Adjustment.” Psycho-Oncology 17(9):932-39.

Burns

Shaun

Mahalik

James R.

2006. “Physical Health, Self-reliance, and Emotional Control as Moderators of the Relationship between Locus of Control and Mental Health among Men Treated for Prostate Cancer.” Journal of Behavioral Medicine 29(6):561-72.

Calasanti

Toni

Pietilä

Ilkka

Ojala

Hanna

. 2013. “Men, Bodily Control, and Health Behaviors: The Importance of Age.” Health Psychology 32(1):15-23.

Caputo

Jennifer

Simon

Robin W.

2013. “Physical Limitation and Emotional Well-being: Gender and Marital Status Variations.” Journal of Health and Social Behavior 54(2):241-57.

Casper

Lynne M.

Bianchi

Suzanne M.

2002. Continuity and Change in the American Family. Thousand Oaks, CA: SAGE Publications.

10.

Clarke

Juanne N.

Everest

Michelle M.

2006. “Cancer in the Mass Print Media: Fear, Uncertainty and the Medical Model.” Social Science and Medicine 62(10):2591-600.

11.

Clarke

Laura Hurd

Griffin

Meridith

. 2008. “Failing Bodies: Body Image and Multiple Chronic Conditions in Later Life.” Qualitative Health Research 18(8):1084-95.

12.

Connell

Raewyn W

. 1995. Masculinities. Cambridge, UK: Polity Press.

13.

Connell

R. W.

Messerschmidt

James W.

2005. “Hegemonic Masculinity: Rethinking the Concept.” Gender and Society 19(6):829-59.

14.

Courtenay

Will H

. 2000. “Constructions of Masculinity and Their Influence on Men’s Well-being: A Theory of Gender and Health.” Social Science & Medicine 50(10):1385-401.

15.

Deimling

Gary T.

Bowman

Karen F.

Sterns

Samantha

Wagner

Louis J.

Kahana

Boaz

. 2006. “Cancer-related Health Worries and Psychological Distress among Older Adult, Long-term Cancer Survivors.” Psychooncology 15(4):306-20.

16.

Dewan

Neha

MacDermid

Joy C.

Packham

Tara

. 2013. “Role of a Self-efficacy-based Model of Intervention.” Critical Reviews in Physical and Rehabilitation Medicine 25(3-4):241-59.

17.

Dowd

Jennifer Beam

Todd

Megan

. 2011. “Does Self-reported Health Bias the Measurement of Health Inequalities in U.S. Adults? Evidence Using Anchoring Vignettes from the Health and Retirement Study.” Journals of Gerontology: Psychological Sciences and Social Sciences 66(4):478-49.

18.

Drabe

Natalie

Steinert

Hans

Moergeli

Hanspeter

Weidt

Steffi

Strobel

Klaus

Jenewein

Josef

. 2016. “Perception of Treatment Burden, Psychological Distress, and Fatigue in Thyroid Cancer Patients and Their Partners.” Psychooncology 25(2):203-09.

19.

Ervik

Bente

Asplund

Kenneth

. 2012. “Dealing with a Troublesome Body: A Qualitative Interview Study of Men’s Experiences Living with Prostate Cancer Treated with Endocrine Therapy.” European Journal of Oncology Nursing 16(3):103-08.

20.

Federal Interagency Forum on Aging-related Statistics. 2016. Older Americans 2016: Key Indicators of Well-being. Washington, DC: U.S. Government Printing Office.

21.

Olivia S.

Crew

Katherine D.

Jacobson

Judith S.

Greenlee

Heather

Gary

Campbell

Julie

Ortiz

Yvette

Hershman

Dawn L.

2009. “Ethnicity and Persistent Symptom Burden in Breast Cancer Survivors.” Journal of Cancer Survivorship 3(4):241-50.

22.

Gerson

Judith M.

Peiss

Kathy

. 1985. “Boundaries, Negotiation, Consciousness: Reconceptualizing Gender Relations.” Social Problems 32(4):317-31.

23.

Gilbert

Keon L.

Ray

Rashawn

Siddiqi

Arjumand

Shetty

Shivan

Baker

Elizabeth A.

Elder

Keith

Griffith

Derek M.

2016. “Visible and Invisible Trends in Black Men’s Health: Pitfalls and Promises for Addressing Racial, Ethnic, and Gender Inequities in Health.” Annual Review of Public Health 37:295-311.

24.

Goymann

Wolfgang

Wingfield

John C.

2004. “Allostatic Load, Social Status and Stress Hormones: The Costs of Social Status Matter.” Animal Behavior 67(3):591-602.

25.

Griffith

Derek M.

2015. “‘I Am a Man’: Manhood, Minority Men’s Health and Health Equity.” Ethnicity and Disease 25(3):287-93.

26.

Griffith

Derek M.

Brinkley-Rubinstein

Lauren

Bruce

MarinoA.

Thorpe

Roland J.

Metzl

Jonathan M.

2015. “The Interdependence of African American Men’s Definitions of Manhood and Health.” Family and Community Health 38(4):284-96.

27.

Gruenewald

Tara L.

Seeman

Teresa E.

Ryff

Carol D.

Karlamangla

Arun S.

Singer

Burton H.

2006. “Combinations of Biomarkers Predictive of Later Life Mortality.” Proceedings of the National Academy of Sciences 103(38):14158-63.

28.

Halliwell

Emma

Dittmar

Helga

. 2003. “A Qualitative Investigation of Women’s and Men’s Body Image Concerns and Their Attitudes toward Aging.” Sex Roles 49(11):675-84.

29.

Hilário

Ana Patrícia

. 2015. “Making Sense of a Changed Physical Body: Why Gender Matters at the End of Life.” Journal of Aging Studies 33:58-66.

30.

Hill

Terrence D.

Needham

Belinda L.

2013. “Rethinking Gender and Mental Health: A Critical Analysis of Three Propositions.” Social Science and Medicine 92:83-91.

31.

Hooker

Steven P.

Wilcox

Sara

Burroughs

Ericka L.

Rheaume

Carol E.

Courtenay

Will

. 2012. “The Potential Influence of Masculine Identity on Health-improving Behavior in Midlife and Older African American Men.” Journal of Men’s Health 9(2):79-88.

32.

Nadia

Howlader

, ed. 2017. SEER Cancer Statistics Review, 1975-2014. Bethesda, MD: National Cancer Institute.

33.

Ichino

Andrea

Mealli

Fabrizia

Nannicini

Tommaso

. 2008. “From Temporary Help Jobs to Permanent Employment: What Can We Learn from Matching Estimators and Their Sensitivity?” Journal of Applied Econometrics 23(3):305-27.

34.

Kane

Emily W.

2000. “Racial and Ethnic Variation in Gender-related Attitudes.” Annual Review of Sociology 26:419-39.

35.

Kenyon

Megan

Mayer

Debora K.

Owens

Ann Kat

. 2014. “Late and Long-term Effects of Breast Cancer Treatment and Surveillance Management for the General Practitioner.” Journal of Obstetric, Gynecologic, & Neonatal Nursing 43(3):382-98.

36.

Keyes

Corey L. M.

2009. “The Black-white Paradox in Health: Flourishing in the Face of Social Inequality and Discrimination.” Journal of Personality 77(6):1677-706.

37.

Keyes

Katherine M.

Barnes

David M.

Bates

Lisa M.

2011. “Stress, Coping, and Depression: Testing a New Hypothesis in a Prospectively Studied General Population Sample of U.S.-born Whites and Blacks.” Social Science & Medicine 72(5):650-59.

38.

Lachman

Margie E.

Weaver

Suzanne L.

1998. “The Sense of Control as a Moderator of Social Class Differences in Health and Well-being.” Journal of Personality and Social Psychology 74(3):763-73.

39.

Lee

Christina

Owens

R. Glynn

2002. “Issues for a Psychology of Men’s Health.” Journal of Health Psychology 7(3):209-17.

40.

Levant

Ronald F.

Wong

Y. Joel

. 2013. “Race andGender as Moderators of the Relationship between the Endorsement of Traditional Masculinity Ideology and Alexithymia: An Intersectional Perspective.” Psychology of Men & Masculinity 14(3):329-33.

41.

Link

Bruce

Phelan

. 1995. “Social Conditions as Fundamental Causes of Disease.” Journal of Health and Social Behavior 35:80-94.

42.

Lukaszewski

Aaron W.

Simmons

Zachary L.

Anderson

Cameron

Roney

James R.

2016. “The Role of Physical Formidability in Human Social Status Allocation.” Journal of Personality and Social Psychology 110(3):385-406.

43.

Martinez

Kathryn A.

Snyder

Claire F.

Malin

Jennifer L.

Sydney M.

2014. “Is Race/Ethnicity Related to the Presence or Severity of Pain in Colorectal and Lung Cancer?” Journal of Pain and Symptom Management 48(6):1050-59.

44.

McCaughan

Eilis

Prue

Gillian

Parahoo

Kader

McIlfatrick

Sonja

McKenna

Hugh

. 2012. “Exploring and Comparing the Experience and Coping of Men and Women with Colorectal Cancer after Chemotherapy Treatment: A Qualitative Longitudinal Study.” Psycho-Oncology 21(1):64-71.

45.

McGuire

Gail M.

Reskin

Barbara

. 1993. “Authority Hierarchies at Work: The Impacts of Race and Sex.” Gender and Society 7(4):487-506.

46.

McIlvane

Jessica M.

Schiaffino

Kathleen M.

Paget

Stephen A.

2007. “Age Differences in the Pain-depression Link for Women with Osteoarthritis. Functional Impairment and Personal Control as Mediators.” Women’s Health Issues 17(1):44-51.

47.

McLeod

Jane D.

2015. “Why and How Inequality Matters.” Journal of Health and Social Behavior 56(2):149-65.

48.

Messerschmidt

James W.

2009 “‘Doing Gender’: The Impact and Future of a Salient Sociological Concept.” Gender and Society 23(1):85-88.

49.

Meyer

Ilan H.

Schwartz

Sharon

Frost

David M.

2008. “Social Patterning of Stress and Coping: Does Disadvantaged Social Status Confer More Stress and Fewer Coping Resources?” Social Science & Medicine 67(3):368-79.

50.

Mirowsky

John

. 1995. “Age and the Sense of Control.” Social Psychology Quarterly 58(1):31-43.

51.

Mirowsky

John

Ross

Catherine E.

1986. “Social Patterns of Distress.” Annual Review of Sociology 12(1):23-45.

52.

Mirowsky

John

Ross

Catherine E.

2007. “Life Course Trajectories of Perceived Control and their Relationship to Education.” American Journal of Sociology 112(5):1339-82.

53.

Mirowsky

John

Ross

Catherine E.

Van Willigen

Marieke

. 1996. “Instrumentalism in the Land of Opportunity: Socioeconomic Causes and Emotional Consequences.” Social Psychology Quarterly 59(4):322-37.

54.

Oliffe

John

. 2006. “Embodied Masculinity and Androgen Deprivation Therapy.” Sociology of Health and Illness 28(4):410-32.

55.

Paternoster

Raymond

Brame

Robert

Mazerolle

Paul

Piquero

Alex

. 1998. “Using the Correct Statistical Test for Equality of Regression Coefficients.” Criminology 6(4):859-66.

56.

Pearlin

Leonard I.

Lieberman

Morton

Menaghan

Elizabeth

Mullan

1981. “The Stress Process.” Journal of Health and Social Behavior 22(4):337-56.

57.

Pearlin

Leonard I.

Schooler

Carmi

. 1978. “The Structure of Coping.” Journal of Health and Social Behavior 19(1):2-21.

58.

Phelan

Jo C.

Link

Bruce G.

2015. “Is Racism a Fundamental Cause of Inequalities in Health?” Annual Review of Sociology 41:311-30.

59.

Powe

Barbara D.

Hamilton

Jill

Hancock

Nichole

Johnson

Natasha

Finnie

Ramona

Jean

Brooks

Patrice

Boggan

Maurice

Jr . 2007. 2007. “Quality of Life of African American Cancer Survivors. A Review of the Literature.” Cancer 109(2 suppl):435-45.

60.

Pudrovska

Tetyana

. 2010a. “Cancer and Mastery: Do Age and Cohort Matter?” Social Science and Medicine 71(7):1285-91.

61.

Pudrovska

Tetyana

. 2010b. “Why Is Cancer More Depressing for Men than Women among White Older Adults?” Social Forces 89(2):535-58.

62.

Pudrovska

Tetyana

. 2015. “Gender and Health Control Beliefs among Middle-aged and Older Adults.” Journal of Aging and Health 27(2):284-303.

63.

Pudrovska

Tetyana

Schieman

Scott

Pearlin

Leonard

Nguyen

Kim

. 2005. “The Sense of Mastery as a Mediator and Moderator in the Association between Economic Hardship and Health in Late Life.” Journal of Aging and Health 17(5):634-60.

64.

Ravi

Praful

Karakiewicz

Pierre I.

Roghmann

Florian

Gandaglia

Giorgio

Choueiri

Toni K.

Menon

Mani

McKay

Rana R.

Nguyen

Paul L.

Sammon

Jesse D.

Sukumar

Shyam

Varda

Briony

Sun

Maxine

Trinh

Quoc-Dien

. 2014. “Mental Health Outcomes in Elderly Men with Prostate Cancer.” Urological Oncology 32(8):1333-40.

65.

Rayner

Jo-Anne

Pyett

Priscilla

Astbury

Jill

. 2010. “The Medicalisation of ‘Tall’ Girls: A Discourse Analysis of Medical Literature on the Use of Synthetic Oestrogen to Reduce Female Height.” Social Science and Medicine 71(6):1076-83.

66.

Rogers

Baron K.

Sperry

Heather A.

Levant

Ronald F.

2015. “Masculinities among African American Men: An Intersectional Perspective.” Psychology of Men & Masculinity 16(4):416-25.

67.

Rosenfield

Sarah

. 2012. “Triple Jeopardy? Mental Health at the Intersection of Gender, Race, and Class.” Social Science and Medicine 74(11):1791-801.

68.

Ross

Catherine E.

Mirowsky

John

. 2002. “Age and the Gender Gap in the Sense of Personal Control.” Social Psychology Quarterly 65(2):125-45.

69.

Ryff

Carol D.

Keyes

Corey L. M.

Hughes

Diane L.

2003. “Status Inequalities, Perceived Discrimination, and Eudaimonic Well-being: Do the Challenges of Minority Life Hone Purpose and Growth?” Journal of Health and Social Behavior 44(3):275-91.

70.

Sapolsky

Robert M.

2004. “Social Status and Health in Humans and Other Animals.” Annual Review of Anthropology 33:393-418.

71.

Schmutte

Timothy

O’Connell

Maria

Weiland

Melissa

Lawless

Samuel

Davidson

Larry

. 2009. “Stemming the Tide of Suicide in Older White Men: A Call to Action.” American Journal of Men’s Health 3(3):189-200.

72.

Schnittker

Jason

McLeod

Jane D.

2005. “The Social Psychology of Health Disparities.” Annual Review of Sociology 31:105-25.

73.

Seeman

Melvin

Stein Merkin

Sharon

Karlamangla

Arun

Koretz

Brandon

Seeman

Teresa

. 2014. “Social Status and Biological Dysregulation: The ‘Status Syndrome’ and Allostatic Load.” Social Science & Medicine 118:143-51.

74.

Sell

Aaron

Tooby

John

Cosmides

Leda

. 2009. “Formidability and the Logic of Human Anger.” Proceedings of the National Academy of Science 106(35):15073-78.

75.

Shaha

Maya

Cox

Carol L.

Talman

Kirsi

Kelly

Daniel

. 2008. “Uncertainty in Breast, Prostate, and Colorectal Cancer: Implications for Supportive Care.” Journal of Nursing Scholarship 40(1):60-67.

76.

Simon

Robin W.

1992. “Parental Role Strains, Salience of Parental Identity, and Gender Differences in Psychological Distress.” Journal of Health and Social Behavior 33(1):25-35.

77.

Skaff

Marilyn McKean

Mullan

Joseph T.

Fisher

Lawrence

Chesla

Catherine A.

2003. “A Contextual Model of Mastery Beliefs, Behavior, and Health: Latino and European Americans with Type 2 Diabetes.” Psychology and Health 18(3):295-312.

78.

Solimeo

Samantha

. 2008. “Sex and Gender in Older Adults’ Experience of Parkinson’s Disease.” Journals of Gerontology: Social Sciences 63(1):S42-48.

79.

Surtees

Paul G.

Wainwright

Nicholas W.

Luben

Robert

Wareham

Nicholas J.

Bingham

Sheila A.

Khaw

Kay-Tee

. 2010. “Mastery Is Associated with Cardiovascular Disease Mortality in Men and Women at Apparently Low Risk.” Health Psychology 29(4):412-20.

80.

Taylor

Catherine J.

2014. “Physiological Stress Response to Loss of Social Influence and Threats to Masculinity.” Social Science and Medicine 103: 51-59.

81.

Toscano

Hugo

Schubert

Thomas W.

Dotsch

Ron

Falvello

Virginia

Todorov

Alexander

. 2016. “Physical Strength as a Cue to Dominance: A Data-driven Approach.” Personality and Social Psychology Bulletin 42(12):1603-16.

82.

Vehling

Sigrun

Lehmann

Claudia

Oechsle

Karin

Bokemeyer

Carsten

Krull

Andreas

Koch

Uwe

Mehnert

Anja

. 2012. “Is Advanced Cancer Associated with Demoralization and Lower Global Meaning? The Role of Tumor Stage and Physical Problems in Explaining Existential Distress in Cancer Patients.” Psycho-Oncology 21(1):54-63.

83.

Vincent

Grayson K.

Velkoff

Victoria A.

2010. The Older Population in the United States: 2010 to 2050. Washington, DC: US Census Bureau.

84.

Wall

David

Kristjanson

Linda

. 2005. “Men, Culture and Hegemonic Masculinity: Understanding the Experience of Prostate Cancer.” Nursing Inquiry 12(2):87-97.

85.

Watkins

Daphne C.

2012. “Depression Over the Adult Life Course for African American Men: Toward a Framework for Research and Practice.” American Journal of Men’s Health 6(3):194-210.

86.

Watts

Sam

Prescott

Philip

Mason

Jessica

McLeod

Natalie

Lewith

George

. 2015. “Depression and Anxiety in ovarian cancer: A Systematic Review and Meta-analysis of Prevalence Rates.” BMJ Open 5(11):e007618.

87.

West

Candace

Zimmerman

Don H.

1987. “Doing Gender.” Gender and Society 1(2):125-51.