Gender Differences in the Relationship Between Financial Stress and Metabolic Abnormalities

Abstract

Background:

Financial stress is associated with higher prevalence of metabolic abnormalities and cardiovascular disease, but the extent to which this association differs by type of metabolic abnormalities or gender is unclear.

Objectives:

The study aims were to: (a) examine the association between financial stress and the prevalence of common metabolic abnormalities; and (b) test the association for gender differences.

Methods:

A cross-sectional secondary analysis was conducted using data from the Retirement and Sleep Trajectories (REST) study, an ancillary study of the Wisconsin Sleep Cohort (WSC) study. Composite indicator structural equation alpha modeling with a stacking approach was applied in the data analysis.

Results:

After controlling for covariates, financial stress was positively associated with the prevalence of abdominal obesity, metabolic syndrome, and dyslipidemia, with significant gender differences. Among men, financial stress was positively associated with the prevalence of hypertriglyceridemia. Among women, financial stress was positively associated with the prevalence of prediabetes, abdominal obesity, metabolic syndrome, and dyslipidemia.

Conclusion:

Men living with financial stress are more likely to have hypertriglyceridemia, a specific metabolic abnormality and risk factor for acute cardiovascular events. However, financial stress in women is associated with a broader array of metabolic abnormalities (e.g., dyslipidemia, prediabetes, abdominal obesity, metabolic syndrome), highlighting a potential risk of multiple chronic conditions later in life.

Financial stress has been defined as a subjective perception of having insufficient financial resources to meet basic requirements for a regular standard of living (Agrigoroaei et al., 2017; Steen & MacKenzie, 2013). As a subjective perception, financial stress consists of several constructs, including perceived inability to pay monthly bills, perceived inability to meet life necessities, and worrying about current or future financial status as an individual or household (Agrigoroaei et al., 2017; Block et al., 2009). Although the economy improved since the Great Recession, financial stress has been rated as the top source of stress for Americans since 2008 (Bethune, 2015). In 2019, survey findings suggested that 25% of Americans experienced severe to extreme levels of financial stress, while 60% experienced moderate levels of financial stress (Holland, 2019). Since the COVID–19 pandemic, the percentage of adults who reported severe to extreme financial stress has doubled, jumping from 25% to 47% (National Endowment for Financial Education, 2020). Compared to other age groups, middle-aged and older adults can be more susceptible to poor health associated with overwhelming financial stress, due to age-related changes in life circumstances. For instance, the increasing medical costs of declining health, income lost with retirement, and life events associated with aging can jointly complicate the health effects of financial stress by limiting healthy behaviors and access to health care. Empirical evidence has shown that delayed health care in response to financial stress are associated with higher risk of cardiovascular disease and incidence of all-cause mortality (Khera et al., 2018).

Financial stress is a type of life stressor that can produce allostatic reactivity and allostatic load. The allostatic load model has been used to explain the intricate psychosocial–pathophysiological pathways underpinning the relationship between financial stress and metabolic abnormalities (Karatsoreos & McEwen, 2011). Allostatic load reflects chronic activation of the physiological stress response and results in physiological wear and tear that affects health (Fava et al., 2019). According to the allostatic load model, this wear and tear triggers neuroendocrine dysregulation in the sympathetic–adrenal–medullary (SAM) and hypothalamic–pituitary–adrenal (HPA) axes, resulting in a hormone dysfunction called glucocorticoid resistance, and further leading to higher risk of metabolic abnormalities (Karatsoreos & McEwen, 2011). Although the stress–metabolism pathway appears to be theoretically consistent with allostatic load model, previous studies examining stress and specific alterations in metabolic biomarkers showed that the effect of stress on metabolic outcomes can vary depending on sources of stress, metabolic phenotypes, and gender (Cosgrove et al., 2012; Kuo et al., 2019; Sui et al., 2016; Tenk et al., 2018). In addition, recent systematic reviews concluded that psychological stress is associated with higher risk of weight gain, abdominal obesity, and metabolic syndrome (Tenk et al., 2018), while there is no significant association between psychological stress and the risk of developing type 2 diabetes mellitus (T2DM; Cosgrove et al., 2012; Sui et al., 2016). These results indicate that the effect of psychological stress on metabolic biomarkers of allostatic load could be source-specific and phenotype-specific. These findings also challenge a central premise of the allostatic load model, where stressors experienced by different people are often quantified using generic perceived stress scale and different metabolic outcomes are categorized into the dysregulation of metabolic function (Karatsoreos & McEwen, 2011). There is a need to examine whether relationships between a specific source of stress (e.g., financial stress) and metabolic outcomes differ depending on the metabolic outcome examined and the population being studied (Epel et al., 2018; Fava et al., 2019).

In addition to metabolic phenotypes, major findings in recent reviews suggest that the effects of stress on metabolism in men and women might involve different physiological and behavioral pathways (Kuo et al., 2019; Love, 2018; Taylor et al., 2018; Tenk et al., 2018). However, studies comparing men and women have also shown that findings for behavioral pathways could differ from those for physiological pathways. For instance, oxytocin has been shown to enhance HPA sensitivity to negative feedback signals that can dampen stress reactivity, reduce cravings for food high in salt and fat, and make weight management easier (Lawson, 2017; Winter & Jurek, 2019). Physiologically, women have higher levels of oxytocin compared to men, which should give them an advantage regarding the effect of stress response on eating behaviors. Yet, many researchers have demonstrated that women were more likely to report stress‐eating compared with men (Beydoun, 2014). Such contradictive relationships between physiological and behavioral pathways highlight the importance of examining whether the relationship between a specific source of stress and metabolic phenotype differs between men and women.

Mounting evidence has suggested that financial stress is associated with higher prevalence of obesity and weight gain in the U.S., especially among those with difficulty paying bills or who have less money to meet life necessities (Block et al., 2009; Conklin et al., 2013). However, it remains unclear whether financial stress is also associated with other metabolic outcomes (e.g., T2DM, hypertension, and dyslipidemia) and whether gender plays a role in these relationships. In the present study, we hypothesized that the relationships between financial stress and metabolic abnormalities differ by types of metabolic outcomes and differ between men and women (Figure 1). The study aims are to: (a) examine the association between financial stress and the prevalence of common metabolic abnormalities; and (b) test the association for gender differences.

Figure 1.

Hypothesized conceptual framework.

Methods

Design, Sample, and Setting

This is a cross-sectional secondary analysis using data from the Retirement and Sleep Trajectories study (REST), an ancillary study of the Wisconsin Sleep Cohort study (WSC). WSC was established in 1988 as a prospective population-based study of the natural history of sleep disorders (Peppard et al., 2013). Participants in WSC were selected from a random sampling pool of payroll records for State of Wisconsin employees aged 30 to 60 years; they have participated in follow-up assessment approximately every 4 years since 1988 (Peppard et al., 2013). As an ancillary study of WSC, REST was a prospective study consisting of four annually mailed surveys—from 2010 to 2015—that collected information regarding stress, health behaviors, sleep disorders, and other psychosocial factors. WSC participants were invited to participate in REST if they had previously responded to at least three WSC mailed surveys or at least one WSC in-laboratory assessment plus one WSC mailed survey (Hagen et al., 2016). The dates of WSC in-laboratory assessments and REST mailed surveys varied for each participant which depended on the schedule of lab visits and returns of questionnaires (Hagen et al., 2016; Peppard et al., 2013).

To be eligible in this cross-sectional data analysis, the participants must have had completed both the first wave of the REST mailed surveys, sent between 2010 and 2011, and a WSC in-laboratory assessment after the first wave of REST mailed survey. In total, 457 participants met the inclusion criteria (See Figure 2). The study was approved by the health science institutional review board (IRB) at the University of Wisconsin–Madison.

Figure 2.

The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: Flow chart of the sample size in the present study.

Measures

Metabolic Abnormalities

The metabolic abnormalities included metabolic syndrome, T2DM, prediabetes, hypertension, dyslipidemia, hypertriglyceridemia, and abdominal obesity. Participants with metabolic abnormalities were identified through WSC in-laboratory assessments of metabolic biomarkers, including waist circumference, resting blood pressure (BP), and blood tests that measured fasting glucose, triglycerides, high-density lipoprotein cholesterol (HDL), and low-density lipoprotein cholesterol (LDL). Medication reconciliation was performed through individual interviews and pill bottle assessments to identify participants who took prescribed medication to control BP, cholesterol, triglycerides, and glucose. Diagnostic criteria were based on current clinical guidelines from the American Heart Association (AHA), the American Diabetes Association (ADA), and the World Health Organization (WHO; see Table 1). All the metabolic abnormalities were coded as “present” or “not present” as binary outcomes.

Table 1.

Current guidelines for metabolic outcomes in the present study

Metabolic outcomes	Criteria a	Guidelines b
Metabolic syndrome	At least three of the criteria below:	AHA/NCEP
	• BP ≥ 130/85 mmHg or medicated  • FPG ≥ 100 mg/dL or medicated  • TG ≥ 150 mg/dL or medicated  • HDL < 40 (men), <50 (women)  • WC > 40 (men), >35 (women)
Hypertension	BP ≥ 130/80 mmHg or medicated	ACC/AHA
T2DM	FPG ≥ 126 mg/dL or medicated	ADA
Prediabetes	FPG ≥100	ADA
Dyslipidemia	Any one of the criteria below:  • TG > 200 mg/dL or medicated  • HDL < 40 (men), <50 (women) mg/dL or medicated  • LDL > 160 mg/dL or medicated	AHA/NCEP
Hypertriglyceridemia	TG > 200 mg/dL or medicated	AHA/NCEP
Abdominal obesity	WC > 40 inches (men), >35 inches (women)	WHO

Note. MetS=metabolic syndrome; T2DM=type 2 diabetes mellitus; BP=blood pressure; FPG=fasting plasma glucose; TG=triglycerides; HDL=high-density lipoprotein; LDL=low-density lipoprotein; WC=waist circumference; AHA=American Heart Association; NCEP=National Cholesterol Education Program; ACC=American College of Cardiology; ADA=American Diabetes Association; WHO=World Health Organization.

^aAll the cutoff points are based on the latest guidelines.

^bThe references are listed in supplement (SDC1.).

Financial Stress

Financial stress was collected through REST mailed survey using a three-item financial stress scale adapted from Boston Longitudinal Study (BOLOS), a Boston area oversample of the Midlife in the United States (MIDUS; Agrigoroaei et al., 2017). This adapted scale was intended to assess respondents’ financial stress on three different constructs, including current financial situation, ability to meet life necessities or financial needs, and ability to pay monthly bills. The three items are on Likert-type scales:

1. “How would you rate your current financial situation?” with responses ranging from 0 (worst) to 10 (best).

2. “In general, would you say you (and your family living with you) have more money than you need, just enough for your needs, or not enough to meet your needs” with responses ranging from 1 (money more than you need) to 3 (not enough money).

3. “How difficult is it for you (and your family) to pay your monthly bills?” with responses ranging from 1 (very difficult) to 4 (not at all difficult).

The first and the third items were coded reversely, so that higher scores indicate higher levels of financial stress on that construct. The standardized Cronbach’s alpha is 0.82 in the present study, indicating good internal consistency.

Covariates

Age, alcohol consumption (number of drinks/week), cigarette smoking status (current smoker or not), depressive symptoms, and numbers of comorbidities were controlled in the data analysis as potential confounding variables. Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES–D; Lewinsohn et al., 1997). Comorbidities included cancer, myocardial infarction, heart failure, Parkinson’s disease, Alzheimer’s disease, glaucoma, restless leg syndrome, emphysema, chronic obstructive pulmonary disease, and kidney disease (Plante et al., 2017). Lastly, because the exposure variable (financial stress) and outcome variables (metabolic biomarkers) were collected at different time points depending on participants’ schedule of lab visits and returns of questionnaires, the duration between the occurrence of exposure variable (REST mailed survey) and the occurrence of outcome variables (WSC in-laboratory assessment) was transformed into days and included in the model as a covariate.

Statistical Analyses

Management of missing data and statistical analyses were performed using SAS software (SAS Institute Inc., 2020) and Mplus software (Muthén & Muthén, 1998–2017). All reported p-values were two-tailed, with p-values less than 0.05 considered as significant. The missing data diagnosis indicated that none of the participants contained more than 27% of missing data, so we included all the participants (N = 457) in the data analysis. We performed multiple imputations using Monte Carlo Markov chain (MCMC) method, which allows us to obtain five imputed data sets based on 100 MCMC iterations (Asparouhov & Muthén, 2010). A Markov chain is a sequence of random variables in which the distribution of a variable depends only on the values of previous variables (Lin, 2010). This algorism draws imputed values from the distribution by iteratively simulating the steps of the chain (Asparouhov & Muthén, 2010; Lin, 2010). See the online Supplemental Digital Content for simulation results.

We used two-phase structural equation modeling to examine the relationship between financial stress and each metabolic abnormality. For the measurement (first) phase, we used Composite Indicator Structural Equation (CISE) alpha modeling because the exposure variable (financial stress) in the present study was a composite variable based on a self-report instrument. To account for measurement error, we built a measurement error term into the composite variable based on the reliability estimates derived from the following formula: ([1 - Cronbach’s alpha of the instrument] x variance of composite variable; McDonald et al., 2005). In the structural (second) phase, we estimated a probit model using weighted least squares mean and variance (WLSMV). WLSMV is a robust estimation for ordinal outcome variables because it does not assume normal distribution (Li, 2016). To interpret the relationship between financial stress and metabolic abnormalities with clinical significance, we transformed the probit coefficient into odds ratio (OR) using OR = EXP(probit*1.7; Thrane, 2019). To ensure the estimated model was reasonably consistent with the current sample, we used McKelvey and Zavoina’s Pseudo R² as the goodness-of-fit index because all outcome variables were binary (McKelvey & Zavoina, 1975). The McKelvey and Zavoina’s pseudo R² was assessed via the R² for each outcome variable based on the probit model (with WLSMV) built in Mplus software (Muthén & Muthén, 1998–2017). Although various types of pseudo R² exist, we chose the pseudo R² suggested by McKelvey–Zavoina, because researchers have found that the pseudo R² suggested by McKelvey–Zavoina performs the best for binary probit models (Veall & Zimmermann, 1994).

To examine gender differences in the relationship between financial stress and the prevalence of metabolic abnormalities, we used a stacking approach with CISE alpha modeling to simultaneously estimate the relationships between financial stress and the prevalence of metabolic abnormalities in men and women. A stacking approach allows researchers to meaningfully categorize participants into subgroups (men and women) and observe how the relationship between the exposure and outcome differs by each subgroup (Byrne, 1998). Furthermore, the stacking approach also permits the estimation of the interaction effect between the exposure and the moderator on each outcome variable by constraining the model with the interaction term (Byrne, 1998).

Results

Sample Characteristics

In this sample of middle-aged and older adults (N = 457), the mean age was 63.96 (SD = 6.99). Among them, 253 (55%) participants were men, 400 (96%) participants were White, 220 (48%) participants had a college degree or higher, and 34 (7%) participants had two or more comorbidities. Detailed sample characteristics are presented in Table 2.

Table 2.

Demographic, psychological, and metabolic characteristics of the study sample

Characteristics	N=457
Age, mean (SD)	63.96 (6.99)
Gender, n (%) men	253 (55%)
Race, n (%) White	400 (96%)
Education, n (%)
High school or less	88 (19%)
Some college	149 (33%)
College or more	220 (48%)
Covered by health insurance, n (%)	446 (98%)
BMI, mean (SD)	31.20 (6.88)
Number of comorbidities, n(%)
None	291 (64%)
One	132 (29%)
Two or more	34 (7%)
Alcohol per week (drinks), mean (SD)	4.04 (5.66)
Current smoker, n (%)	35 (8%)
Financial stress, mean (SD) a
Paying bills [response range 1–4]	1.74 (0.78)
Paying life necessities [response range 1–3]	1.77 (0.59)
Current financial strain [response range 0–10]	3.26 (1.80)

Note. BMI=body mass index.

^aHigher scores indicate higher levels of financial stress on that construct.

Association Between Financial Stress and Metabolic Abnormalities

Results of our CISE alpha modeling are shown in Table 3. Financial stress was significantly associated with the prevalence of abdominal obesity (OR = 1.577; p = .001), metabolic syndrome (OR = 1.637; p < .001), and dyslipidemia (OR = 1.384; p = .042), after controlling for age, number of comorbidities, alcohol consumption, current smoking status, and depressive symptoms. The standardized results showed that one standard deviation (SD) increase in financial stress was associated with 40% higher odds of abdominal obesity (OR* = 1.40), 42% higher odds of metabolic syndrome (OR* = 1.42), and 26% higher odds of dyslipidemia (OR* = 1.26). Pseudo R² suggested a reasonable model fit, where financial stress explained 12% of the variance in hypertension, 9% of the variance in T2DM, 7% of the variance in prediabetes, 10% of the variance in hypertriglyceridemia, 11% of the variance in dyslipidemia, 10% of the variance in abdominal obesity, and 15% of the variance in metabolic syndrome.

Table 3.

Associations between financial stress and the prevalence of metabolic abnormalities in men and women

Relationships	All (N=457)			Men (n=253)			Women (n=204)			Interaction Terms
	OR	95% CI	p-value	OR	95% CI	p-value	OR	95% CI	p-value	βmen – βwomen (Δβ)	p-value
FS → HTN	1.113	[0.861, 1.439]	.494	1.036	[0.705, 1.527]	.878	1.273	[0.923, 1.755]	.216	−0.121	.502
FS → Prediabetes	1.228	[0.949, 1.588]	.191	0.967	[0.646, 1.449]	.892	1.674	[1.140, 2.458]	.027	−0.323	.106
FS → T2DM	1.167	[0.877, 1.553]	.372	1.115	[0.693, 1.792]	.707	1.202	[0.844, 1.708]	.394	−0.044	.836
FS → AO	1.577	[1.258, 1.977]	.001	1.115	[0.767, 1.623]	.631	2.178	[1.705, 2.778]	<.001	−0.393	.014
FS → MetS	1.637	[1.304, 2.056]	<.001	1.322	[0.925, 1.892]	.199	2.098	[1.629, 2.699]	<.001	−0.271	.083
FS → Dyslipidemia	1.384	[1.063, 1.801]	.042	1.202	[0.788, 1.832]	.475	1.626	[1.202, 2.204]	.008	−0.178	.338
FS → Hypertriglyceridemia	1.228	[0.978, 1.545]	.139	1.671	[1.119, 2.496]	.035	1.094	[0.803, 1.489]	.632	0.249	.168

Note. FS=financial stress; HTN=hypertension; T2DM=type 2 diabetes mellitus; AO=abdominal obesity; MetS=metabolic syndrome; 95% CI=95% confidence interval; OR=Odds ratio; β=probit coefficient. Age, number of comorbidities, alcohol consumption, current smoking status, and depressive symptoms were controlled as covariates in this model.

Gender Differences in the Association Between Financial Stress and Metabolic Abnormalities

Results of our CISE alpha modeling with stacking approach are presented in Table 3. Among men, financial stress was positively associated with the prevalence of hypertriglyceridemia (OR = 1.671; p =.035) after controlling for age, number of comorbidities, alcohol consumption, current smoking status, and depressive symptoms. Among women, financial stress was positively associated with the prevalence of prediabetes (OR = 1.674; p = .027), abdominal obesity (OR = 2.178; p < .001), metabolic syndrome (OR = 2.098; p < .001), and dyslipidemia (OR = 1.626; p = .008) after controlling for the same covariates. The interaction terms between financial stress and gender for the prevalence of metabolic abnormalities are also presented in Table 3.

Discussion

Prior evidence suggested that psychological stress is associated with increased prevalence of metabolic syndrome (Kuo et al., 2019; Tenk et al., 2018). The present study adds to the body of literature by further examining the relationship between financial stress and different phenotypes of metabolic abnormalities and the gender differences.

Decades of research have described the link between psychological stress and metabolic abnormalities, but the underlying mechanisms of the link are not yet fully understood. Within the allostatic load model, the wear and tear accumulated from stressful life demands could alter adults’ diurnal cortisol rhythm and decrease the capacity of glucocorticoids to suppress endotoxin-stimulated cytokine production, resulting in a hormone dysfunction called glucocorticoid resistance (Fava et al., 2019; McEwen, 2020). In addition to this physiological pathway, psychologists have suggested that long-term exposure to psychological stress could alter an individual’s perception of needs and how they should prioritize needs; this in turn could lead to feelings of inadequate personal control. Specifically, Shah et al. (2012) proposed the scarcity mindset hypothesis, which states that long-term exposure to financial stress alters adults’ perceptions about what they need most and what they should prioritize. It is possible that adults with financial stress are more likely to prioritize financial needs instead of physical health (Shah et al., 2015). As a consequence, adults who must prioritize their financial needs tend to choose less expensive unhealthy food over more expensive healthier food. Furthermore, adults living with financial stress may be more likely to live in economically depressed areas with higher chances of being food deserts/swamps, such that they might lack access to sources of healthy nutrition in the first place (Brown & Brewster, 2015; Pan et al., 2012). Taken together, the underlying mechanisms in the relationship between financial stress and metabolic syndrome might involve both the direct inflammatory pathway and the indirect behavioral pathway. Longitudinal studies using precise measures of perception and inflammatory biomarkers are needed to examine and interpret the psychological and behavioral restrictions inherent to financial stress, and consequently, to this physiological mechanism in the stress–metabolism relationship.

In this cross-sectional analysis, we found that financial stress was significantly associated with increased prevalence of abdominal obesity, dyslipidemia, and metabolic syndrome, but we found no relationship between financial stress and the prevalence of T2DM or hypertension. This result is consistent with previous literature reviews which showed that chronic stress is associated with lipid dysfunction as manifested by abdominal obesity and dyslipidemia (Tenk et al., 2018). Although the underlying mechanisms are not fully understood, we hypothesize that the behavioral and physiological influence of stress on metabolic function may start with accumulation of visceral fat and dyslipidemia, either from excessive energy imbalance (e.g., a sedentary lifestyle), increased consumption of saturated fat, or dysregulation of the HPA axis (e.g., abnormal diurnal cortisol pattern).

Worldwide, nearly one third of adults with abdominal obesity are considered metabolically healthy without chronic conditions (Phillips, 2016). However, under the continuous influence of chronic stress (e.g., financial stress) and the dysregulation of HPA axis, the accumulation of visceral fat and dyslipidemia might accelerate the development of chronic conditions through the desensitization of β cells and damage in the endothelium of vessels, leading to multiple chronic conditions over time (Karatsoreos & McEwen, 2011; Mongraw-Chaffin et al., 2016). Prospective studies designed to examine the development of these metabolic alterations in relation to financial stress and the acceleration of metabolic alteration triggered by HPA dysregulation are needed. Our findings support public health programs and nursing symptom science aimed at identifying key cardiometabolic biomarkers and mechanisms and improving the clinical management of early metabolic abnormalities in financially vulnerable populations (National Institute of Nursing Research, n.d.).

Our major study findings also suggest potential gender differences in the metabolic profiles of women and men experiencing financial stress. In the present study, women with financial stress had higher prevalence of prediabetes, abdominal obesity, metabolic syndrome, and dyslipidemia. This metabolic profile in financially stressed women covers a broad array of metabolic abnormalities, highlighting a potential risk of multiple chronic conditions later in life. From a life-course perspective, compared with men, women more often are the primary care givers for elderly parents and children and, due to pregnancy and child-rearing, are more likely to have their education or employment careers interrupted or delayed. These life circumstances might jointly lead to greater likelihood of financial stress for older women compared to older men (Herpolsheimer, 2015). In contrast, the metabolic cost of financial stress in men manifests in different outcomes. Our findings indicated that men with financial stress showed higher prevalence of hypertriglyceridemia, highlighting a potential risk of atherosclerosis, stroke, and myocardial infarction in the future. In line with our findings, Carlsson et al. (2014) found that financial stress was associated with higher risk of cardiovascular events and all-cause mortality among men compared with women. Hypertriglyceridemia is often asymptomatic and not manifested in body weight. As a silent killer, hypertriglyceridemia damages the endothelial vessels through chronic inflammation (Ali et al., 2018); financial stress might exacerbate this damage through HPA dysregulation and chronic inflammation (Epel et al., 2006). Furthermore, due to lower levels of peripheral oxytocin in men, men are less likely to report emotional and physical symptoms of stress and tend to withdraw socially when experiencing stress (McKenzie et al., 2018). Therefore, given the fact that gender plays an important role in the metabolic costs of financial stress, researchers should consider different physiological and behavioral pathways underlying gender differences in the stress–metabolism relationship when designing interventions to address cardiometabolic risk in financially stressed adults.

Cardiometabolic diseases tend to be underdiagnosed in financially stressed adults (Amrock et al., 2017). In reality, financially stressed adults are more likely to be covered by Medicare or Medicaid than nonfinancially stressed adults (Sommers & Oellerich, 2013). Given the association between financial stress and the prevalence of dyslipidemia and hypertriglyceridemia in middle-aged and older adults, there is a need for the Centers for Medicare and Medicaid Services (CMS) to increase access to screening for dyslipidemia. Unfortunately, CMS currently reimburses the cost of a lipid test only every 5 years (Medicare.gov, n.d.). Our findings highlight the practical utility of conducting longitudinal risk analysis of stress-related poor health outcomes to design effective dyslipidemia screening protocols for adults enrolled in Medicare and Medicaid programs.

Limitations

The strengths of the present study include a relatively large sample size, strong statistical power, measuring financial stress using a reliable instrument, and assessing a comprehensive panel of metabolic abnormalities. However, the present study also has limitations. Due to the cross-sectional design, causal inference and recursive relationships between the exposure and outcome variables cannot be concluded. Nevertheless, there is still a sequential order in the relationship between the exposure variable and outcome variables. Specifically, based on the study design, financial stress was assessed in REST surveys between 2010 and 2011, while metabolic abnormalities were assessed during WSC in-laboratory visits after 2011—which ensures a logical sequential order between the exposure and outcome variables. Finally, the current sample is based on WSC, which has little racial/ethnic diversity (96% of participants were White). Thus, caution should be taken when generalizing the results to other racial/ethnic backgrounds.

Conclusion

The effect of financial stress on metabolic outcomes differs by types of metabolic abnormalities and gender differences. Women with financial stress show a broader range of metabolic abnormalities (e.g., prediabetes, abdominal obesity, metabolic syndrome, and dyslipidemia), highlighting a potential risk of multiple chronic conditions later in life. On the other hand, men with financial stress show higher prevalence of hypertriglyceridemia, highlighting a potential risk of future cardiovascular events. Longitudinal studies are needed to examine the long-term behavioral and physiological effects of financial stress on cardiometabolic outcomes in men and women. Gender differences in cardiometabolic risk factors may help explain the morbidity observed in financially vulnerable women.