Material circumstances,health care access,and self-reported health: A latent class analysis

Abstract

Latent class analysis was used to explore intersections of material circumstances and health care access among 308 adults, and associations between classes with health outcomes. Good fit was found for a four-class model: Resource Stable (Class 1, 62.43%), Unbalanced Meals with Health Care (Class 2, 16.91%), Resource Insecurity with Delayed Health Care (Class 3, 14.75%), and Resource Stability without Access to Health Care (Class 4, 5.91%). Class 1 reported greater well-being and self-rated health than Class 2 and 3. Class 1 reported lower BMI than Class 2. Findings document intersections among economic marginalization indicators with varying health outcomes among classes.

Keywords

economic marginalization health care access latent class analysis material circumstances self-reported health

Multidisciplinary research has documented the effects of economic marginalization on adverse mental and physical health outcomes, such as depressive symptoms and cardiovascular disease (e.g. Galobardes et al., 2006; Lorant et al., 2003). The psychology literature has been limited by singular indicators of economic marginalization (e.g. income) and analytic tests that calculate sample averages (Adler et al., 1994; American Psychological Association, 2019; Liu, 2011). A review of economic marginalization measurement best practices (e.g. Diemer et al., 2013) is beyond the scope of our discussion; however, our study sought to model inequity in material circumstances, relative indicators of poverty that capture the “affective and psychological experience” of economic marginalization (Diemer et al., 2013, p. 103) and have been identified as the most crucial pathway in understanding socioeconomic health disparities (Solar and Irwin, 2010). Documenting intersections between these indicators and health care access has the potential to uncover the varying ways groups of people experience material circumstances and the health system, providing information relevant for dismantling socioeconomic health disparities. Application of person-centered statistical techniques, which identify groups of people with similar response patterns among items (Masyn, 2013; Nylund et al., 2007), has the potential to identify these groups, and explore how they may be related to self-reported health outcomes.

Material circumstances, health care access, and health

The psychology discipline has primarily focused on psychosocial, behavioral, and biological factors which contribute to socioeconomic health disparities (e.g. Adler, 2009; Matthews and Gallo, 2011). Although commonly assessed by family process researchers (Diemer et al., 2013), material circumstances (e.g. food security, housing stability) have received less attention in psychological literature, despite their role in socioeconomic health disparities (Solar and Irwin, 2010). Given the influence of structural factors, such as unemployment rates and housing policies, on material circumstances and health care access, increased attention on these variables in relation to psychological outcomes is in line with calls in the literature to consider the multiple levels that contribute to personal health and well-being (Tucker et al., 2019).

The frequency and impact of deprivation in material circumstances is well-documented. According to the United States (US) Department of Agriculture Economic Research Service (United States Department of Agriculture, n.d.a), for example, 11.1% of US homes were food insecure in 2018. Among homes with an annual household income under 100% of the federal poverty level, an indicator of absolute poverty (Diemer et al., 2013), 35.3% of homes were food insecure (United States Department of Agriculture, n.d.b). Similarly, whereas 2.28% of participants in the 2017 American Housing Survey were unable to pay all or part of rent over the past 3 months, 4.5% of people between 100% and 149% of the federal poverty level reported this circumstance (United States Census Bureau, n.d.a). Further, multidisciplinary research has shown that food insecurity and housing instability contribute to poor health (Adler, 2009). Greater food insecurity has been associated with poorer self-rated health, more physical limitations, and increased likelihood of a mood disorder diagnosis (Gundersen and Ziliak, 2015; Siefert et al., 2004; Stuff et al., 2004), while housing instability has been linked with depression and general anxiety (Suglia et al., 2011).

There are notable overlaps among material circumstances and health care access. For example, of the 7659 adults in a US sample (n = 16,651) who described experiencing food insecurity, 42.4% also reported housing instability and 27.8% indicated no usual source of health care (Kushel et al., 2006). These data reflect difficult decisions associated with navigating economic stressors. In a qualitative analysis of 51 parents in households with young children, parents describe alternating payments each month for housing and utilities or the challenging decision of selecting between food or life-sustaining medication for one’s child (Knowles et al., 2016). Additional data show that policy changes targeting one resource can affect the accessibility of other resources. For instance, following the implementation of the Affordable Care Act, national data demonstrated that as the number of people insured increased, the stress associated with paying for nutritious meals and housing declined among those newly eligible for health care services (Kino et al., 2018).

Despite overlaps among material circumstances and health care access, researchers typically focus on the independent contributions of these variables to mental and physical health outcomes, including psychological well-being, self-rated health, and body mass index (e.g. Knowles et al., 2016; Martin et al., 2019; Moradi et al., 2019; Myers, 2020; Willis et al., 2020). While cumulative indices of material circumstances can be used (Diemer et al., 2013), qualitative data highlight meaningful interactions among these economic stressors (e.g. Knowles et al., 2016). Person-centered quantitative statistical techniques may reveal patterns in how material circumstances and health care access vary among groups of people, and subsequently how those groups relate to health outcomes.

Latent class analysis

Latent class analysis (LCA) is an increasingly popular person-centered statistical approach which describes patterns in variable(s) among groups of people (Masyn, 2013; Nylund et al., 2007). This analysis allows us to model the heterogeneity in material circumstances and health care access among people using a finite number of classes or groups. In other words, these classes allow us to document inequities in experiences of material circumstances and health care access that emerge among groups of people. Pairing LCA with variable-centered analyses provides an opportunity to examine the relationship between class membership and self-reported health outcomes (i.e. distal outcomes; Nylund-Gibson et al., 2019).

LCA has relevance for capturing variation in economic marginalization. A study with 750 White and Black adults found distinct classes of socioeconomic mobility by race using individual objective (e.g. education level, income, governmental assistance) and subjective (e.g. perceived availability of money) measures of economic marginalization (Surachman et al., 2020). These patterns in socioeconomic mobility were associated with immune dysregulation among White adults, but not Black adults (Surachman et al., 2020). While LCA has also been utilized in epidemiological studies investigating indicators of economic marginalization and health (e.g. Goodwin et al., 2018), existing research has not examined whether latent classes emerge based upon material circumstances and health care access, crucial pathways linking economic marginalization with health (Solar and Irwin, 2010).

Theoretical framework

The Commission on Social Determinants of Health (CSDH) framework provides theoretical rationale for considering the overlaps among material circumstances and health care access in relation to personal health and well-being (Solar and Irwin, 2010). This model considers both structural and intermediary determinants of health disparities, with structural determinants operationalized as contextual (e.g. policies, societal values) and individual (e.g. income, gender, race/ethnicity) factors and intermediary determinants operationalized as material circumstances, psychosocial factors, behavioral and biological pathways, and the health system. As described in this model, social inequities contribute to health outcomes via intermediary determinants of health (Solar and Irwin, 2010). Using socioeconomic health disparities as an example, this model would posit socioeconomic-based contextual (e.g. policies, values) and individual (e.g. income, education level, occupation) factors influence material circumstances (e.g. food insecurity), psychosocial factors, behavioral and biological pathways, and health care access, all of which ultimately give rise to health outcomes. As described earlier, the critical gap in the literature is modeling the overlap among material circumstances and health care access and examining associations with personal health outcomes.

Current study

The current study employed an LCA to explore heterogeneity in material circumstances and health care access among a sample of 308 adults in the US, broadly reflective of the US Census Bureau demographics (United States Census Bureau, n.d.b). While we expected unequal classes to emerge, reflecting existing inequities in material circumstances and health care access in the US, LCA is an exploratory approach (Masyn, 2013). To provide some evidence of validity for the classes, we also examined associations among class membership and self-reported indices of mental and physical health that have been shown to differ by material circumstances (e.g. Knowles et al., 2016; Martin et al., 2019; Moradi et al., 2019; Myers, 2020; Willis et al., 2020): well-being, self-rated health, and body mass index.

Method

Study design and participants

Cross-sectional data were collected from 322 participants using Amazon Mechanical Turk (MTurk) in 2018. MTurk has been increasingly used in the social sciences; the benefits and psychometric properties for this approach are detailed elsewhere (Paolacci and Chandler, 2014; Schleider and Weisz, 2015; Shapiro et al., 2013). Relevant to this study, participants recruited via MTurk have been shown to be more diverse than samples obtained via other mediums (i.e. social media crowdsourcing and undergraduate students) in terms of race/ethnicity and socioeconomic position (Casler et al., 2013). The broader study aim was to examine intrapersonal (e.g. mindfulness, repetitive thinking) and interpersonal (e.g. interpersonal mindfulness, co-rumination) modifiers of socioeconomic stress and self-reported health resulting in the following inclusion criteria: age 18 or older, living in the US at the time of participation, and being in a romantic relationship. Socioeconomic health inequities vary internationally (Mackenbach et al., 2008), thus the study was restricted by geographic location. Because the study considered the influence of both intrapersonal and interpersonal resources in the relationship between socioeconomic position and self-rated health, relationship status was an inclusion criterion. To promote high-quality data, participants were also expected to hold an MTurk Masters Qualification, meaning that they have demonstrated and maintained attentiveness and accuracy across other MTurk assignments or tasks (FAQs, n.d.).

A total of 14 participants who reported living outside of the US were excluded from further analyses. The remaining participants identified as residing in the US with one participant not responding. In addition, 44 participants identified as single, divorced, widowed, or preferred not to answer the relationship status item. Because interpersonal modifiers were not under examination in the current analyses, these participants were retained for analyses. Sample characteristics are described in Table 1.

Table 1.

Descriptive statistics for demographic characteristics.

	% yes (n)/M (SD; range)
1 Age (year)	39.49 (10.7; 22–72)
2 Race*
American Indian/Alaskan Native	2.3 (7)
Asian	6.2 (19)
Black/African American	6.5 (20)
Biracial	2.9 (9)
Latinx	2.6 (8)
Multiracial	0.6 (2)
Native Hawaiian, Pacific Islander	0 (0)
White	77.6 (239)
Prefer not to answer	1.3 (4)
3Ethnicity
Hispanic or Latinx	7.5 (23)
Not Hispanic or Latinx	92.5 (285)
4 Gender
Female	53.6 (165)
Genderqueer	0 (0)
Male	45.5 (140)
Nonbinary	0 (0)
Prefer not to answer	1.0 (3)
5 Education level
Less than high school	0 (0)
High school diploma	12.7 (39)
Some college	21.1 (65)
Technical degree or certification	3.2 (10)
Associate’s degree	15.9 (49)
Bachelor’s degree	33.4 (103)
Graduate degree	13.3 (41)
Prefer not to answer	0.3 (1)
6 Receipt of government support
No	91.6 (282)
Yes	8.4 (26)
7 Annual household income
$10,000 or less	2.3 (7)
$10,001−$20,000	9.1 (28)
$20,001−$35,000	18.5 (57)
$35,001−$50,000	16.2 (50)
$50,001−$100,000	40.6 (125)
$100,001−$150,000	8.8 (27)
$150,001 or more	3.2 (10)
Prefer not to answer	1.3 (4)
8 Employment status
Employed	89.3 (275)
Not employed	10.7 (33)
9 Relationship status
Divorced	1.6 (5)
In a relationship	36.7 (113)
Married	49.0 (151)
Separated	0 (0)
Single	12.0 (37)
Widowed	0.3 (1)
Prefer not to answer	0.3 (1)
10 Sexual orientation^#
Bisexual	8.1 (25)
Gay	1.3 (4)
Heterosexual	88.6 (273)
Lesbian	1.0 (3)
Prefer not to answer	0.6 (2)

n = 308.

Latinx was also included as an option for race. Eight participants who self-reported as biracial shared further details, with four participants identifying as Latinx and White, two participants identifying as Asian and White, one participant identifying as Black and White, and one participant identifying as American Indian and White. In addition, one participant who endorsed a multiracial identify further reported being Asian, White, and Native Hawaiian/Pacific Islander.

One person preferred to self-describe and identified as asexual.

Study procedures

Participants were recruited using a study notification on MTurk. MTurkers interested in the study reviewed a preamble and agreed to participate in the survey. After completion of the survey, participants were compensated $4.00. While this reimbursement is above the MTurk median hourly rate of $1.38 (Horton and Chilton, 2010), this price is more in line with suggested best practices of MTurk and the minimum federal hourly wage of $7.25 in the US (U.S. Department of Labor, n.d.). The study was approved by the University of Louisville Institutional Review Board.

Measures

Material circumstances

Food insecurity was measured using two items from 2012 United States Department of Agriculture Household Food Security Survey Module capturing information from the past 12 months: “The food that (I/we) bought just didn’t last, and (I/we) didn’t have money to get more” and “(I/we) couldn’t afford to eat balanced meals” (United States Department of Agriculture, n.d.c). Participants responded to both items using the following scale: often true, sometimes true, never true, or don’t know/refused (United States Department of Agriculture, n.d.c). Two-item screeners for food insecurity have demonstrated appropriate sensitivity and specificity in the literature (e.g. Gundersen et al., 2017) and have been associated with health outcomes (e.g. Harrison et al., 2019). Most two-item screeners of food insecurity focus on food access (e.g. Gundersen et al., 2017); however, given the relevance of food quality and variety for measuring the complete food insecurity continuum (United States Department of Agriculture, n.d.c), we selected the two aforementioned items. Housing instability was assessed using two items with a dichotomous yes or no response: “Have you had trouble paying rent, mortgage or utility bills in the past year?” and “Have you had to move in with family or friends because you had no other choice?” (Kushel et al., 2006). These items have been associated with poor access to health care (Kushel et al., 2006).

Health care access

Consistent with prior studies, health care access was measured using three items with a dichotomous yes or no response: “Do you have a usual medical provider to go to when sick?,” “Do you experience a delay in obtaining healthcare?,” and “Do you experience a delay in obtaining prescription medication?” (Kushel et al., 2006; Vargas Bustamante et al., 2012). These items have been associated with food insecurity and housing instability (Kushel et al., 2006).

Demographic variables

Age, race, ethnicity, gender, annual household income, education level, employment status, receipt of government support, and sexual orientation were collected. Age was obtained using an open-ended question. Race, gender, and sexual orientation were collected with the option of selecting a specific identifier or self-describing using text. Ethnicity, annual household income, education level, employment status, and receipt of government support were collected using multiple choice.

Self-reported mental and physical health outcomes

Well-being

To measure self-reported mental health, the World Health Organization-Five (WHO-5; World Health Organization, 1998) Well-Being Index was used. This measure asks individuals to respond to 5 items assessing well-being over the past 2 weeks using a 6-point Likert scale ranging from 5 (all of the time) to 0 (none of the time). A total score was obtained by summing the items. In addition to strong reliability and validity in measuring subjective well-being (Krieger et al., 2014), the WHO-5 Well-Being Index has demonstrated acceptable sensitivity and specificity as a screener of depression (Topp et al., 2015). In the current study, internal consistency for the WHO-5 Well-Being Index was α = 0.93, similar to previous findings in which Cronbach’s alphas ranged from 0.89 to 0.95 (Hajos et al., 2013; Krieger et al., 2014).

Self-rated health

Self-reported physical health was assessed using the global self-rated health item from the Medical Outcomes Study Short-Form Health Survey (“[I]n general, would you say your health is”) with the response options of excellent, very good, good, fair, or poor (Ware and Sherbourne, 1992). Poorer health, as measured using this item, has been associated with mortality risk (e.g. DeSalvo et al., 2006).

Body mass index (BMI)

BMI was calculated utilizing self-reported weight and height. BMI has been associated with adverse health outcomes (e.g. Dwivedi et al., 2020).

Statistical analyses

Analyses were conducted in SPSS 24.0 and Mplus 8.3. Two participants responded “don’t know or refused” in reference to one item assessing food insecurity; these responses were treated as missing. To explore the heterogeneity in material circumstances (i.e. food insecurity and housing instability) and health care access with all cases, maximum likelihood estimation was used. Unconditional LCA models were fit using seven dichotomous items measuring food insecurity, housing instability, and health care access. Consistent with best practice (Nylund-Gibson and Masyn, 2016), this process was conducted until an unidentified model was encountered. To determine the best fitting model, fit indices were evaluated and compared: Akaike’s Information Criterion (AIC), Bayesian Information Criterion (BIC), sample-size adjusted BIC (saBIC), Approximate Weight of Evidence Criterion (AWE), Lo-Mendell Rubin Likelihood Ratio Test (LMR-LRT), and Bootstrapped LRT (McLachlan and Peel, 2000; Nylund et al., 2007). Increased attention was placed on indices that have been shown to perform well when differentiating categorical LCA models, specifically the saBIC and BLRT (Henson et al., 2007; Nylund et al., 2007). Sample sizes greater than 300 have been suggested as suitable for LCA models based on methodological studies, although the specific number can vary depending on multiple factors (e.g. type of indicator, number of classes; Nylund-Gibson and Choi, 2018). While we expected unequal classes to emerge, the LCA model estimated for this study was relatively simple and used seven dichotomous indicators routinely incorporated into studies assessing material circumstances and health care access.

To examine associations among class membership with self-reported mental and physical health outcomes, latent class means were estimated using the automated BCH method for the following outcomes: well-being, self-rated health, and body mass index (Nylund-Gibson et al., 2019). Data were missing for one participant on well-being, and four participants reported a body mass index considered severely underweight (below 14) and unlikely given life-threatening circumstances; these were treated as missing. The authors confirm that the data supporting the findings of this study are available as Supplemental Materials.

Results

Descriptive statistics and correlations

The five items assessing housing instability and health care access used a dichotomous (yes/no) response system; distributions of these items can be found in Supplemental Table 1. Two items measuring food insecurity were captured using a four-point response system (United States Department of Agriculture, n.d.c). In response to item one, “[t]he food that (I/we) bought just didn’t last, and (I/we) didn’t have money to get more,” participants responded often (n = 11; 3.6%), sometimes (n = 57; 18.5%), or never true (n = 240; 77.9%); no participants responded don’t know or refused. In reference to item two, “(I/we) couldn’t afford to eat balanced meals,” participants responded often (n = 22; 7.1%), sometimes (n = 66; 21.4%), or never true (n = 218; 70.8%); as described earlier, two participants responded don’t know/refused (0.6%) and their responses were coded as missing. The two items measuring food insecurity were recoded into yes (often or sometimes true) or no (never true) responses for the LCA and these distributions are provided in Supplemental Table 1. Descriptive statistics and correlation coefficients (phi and point-biserial) for material circumstances, health care access, and self-reported health outcomes can be found in Supplemental Table 1.

Unconditional LCA model enumeration

LCA models were fit using seven items assessing material circumstances and health care access. Supplemental Table 2 provides model fit indices for nine latent class models. Consistent with best practice (Nylund-Gibson and Masyn, 2016), models were fit until an unidentified model was encountered. Model 10 was unidentified and therefore excluded from further consideration.

With smaller values indicative of better fit, review of the AIC and saBIC pointed toward a four-class model, while the BIC suggested a three-class model. Of these models, the smaller AWE value was obtained in a three-class model; however, this index may under-extract classes (Kaar et al., 2018). The LMR-LRT and BLRT p-values were also reviewed; models that are not significantly improved by an additional class (i.e. non-significant p-value) are reflective of good fit. When comparing a three-class model to a four-class model, the LMR-LRT p-value was not significant while the BLRT p-value was significant; however, when comparing a four-class model to a five-class model, both the LMR-LRT and BLRT p-values were non-significant. Entropy values for the three- and four-class models were 0.91 and 0.90, respectively. Simulation studies have shown that the saBIC and BLRT perform well in categorical LCA models with smaller sample sizes (Henson et al., 2007; Nylund et al., 2007). Thus, after placing increased attention on those indices and considering the substantive theoretical meaning of the three- and four-class model, the four-class model was selected for further analyses. Classes comprising the four-class model were named after close consideration of the conditional item probabilities reflective of high homogeneity (>0.70 or <0.30; Nylund-Gibson and Choi, 2018): Unbalanced Meals with Health Care (Class 1, 17.5%), Food and Housing Insecurity with Delayed Health Care (Class 2, 13.38%), Food and Housing Security with Health Care (Class 3, 62.55%), and Food Security without Health Care (Class 4, 6.58%). Conditional item probabilities for each class can be found in Table 2, with a visual depiction in supplemental materials (Supplemental Figure 1).

Table 2.

Item probabilities among latent classes and mean comparisons between latent classes on self-reported health outcomes.

Class count¹	Unbalanced meals with health care	Food and housing insecurity with delayed health care	Food and housing security with health care	Food security without health care
Class count¹	n = 53.91 (17.5%)	n = 41.20 (13.38%)	n = 192.64 (62.55%)	n = 20.25 (6.58%)
Item probabilities
Food Insecurity 1	0.61	0.75	0.02	0.00
Food Insecurity 2	0.86	0.92	0.00	0.19
Housing Instability 1	0.66	0.75	0.13	0.47
Housing Instability 2	0.09	0.22	0.02	0.12
Health Care Access 1	0.80	0.37	0.85	0.09
Health Care Access 2	0.08	1.00	0.04	0.91
Health Care Access 3	0.04	0.72	0.00	0.60
Mean (standard error)
Well-being	11.89 (0.89)^a	11.02 (1.09)^a	15.67 (0.43)^b	14.32 (1.42)^ab
Self-rated Health²	2.86 (0.12)^a	3.10 (0.16)^a	2.34 (0.06)^b	2.63 (0.21)^ab
Body Mass Index	28.79 (0.95)^a	28.08 (1.68)^ab	26.03 (0.42)^b	26.69 (1.70)^ab

Item probabilities reflective of high homogeneity (>0.70 or <0.30) are presented in bold. Means in a row sharing the same superscript do not differ statistically significantly from each other based on pairwise Wald chi-square tests (p < 0.05).

Counts represent the probable membership based on the estimated model.

Better self-reported health is reflected in lower values.

Class membership and self-reported health outcomes

Estimated means for well-being, self-rated health, and BMI can be found for each class in Table 2. The Food and Housing Security with Health Care class reported greater well-being and higher self-rated health than the Unbalanced Meals with Health Care (ps < 0.001) and Food and Housing Insecurity with Delayed Health Care (ps < 0.001) classes. In addition, the Food and Housing Security with Health Care class reported lower BMI than the Unbalanced Meals with Health Care (p < 0.01) class. No significant differences between the Food Security without Health Care class and other classes emerged on these outcomes.

Discussion

The current study found good fit for a four-class unconditional LCA model using material circumstances and health care access data from 308 adults: Unbalanced Meals with Health Care, Food and Housing Insecurity with Delayed Health Care, Food and Housing Security with Health Care, and Food Security without Health Care. As intended, this exploratory study provides proof-of-concept for the use of LCA in capturing the ways in which groups of people experience material circumstances and health care access. Prior studies examining the relationship between material circumstances and health care access have supported positive relationships among these variables, where, for example, greater food insecurity would be associated with greater housing instability (Kino et al., 2018; Kushel et al., 2006). However, analyzing these associations using variable-centered statistical approaches does not take into consideration patterns that exist among people, overlooking the complexity that is embedded in the experience of navigating economic marginalization. The classes that materialized in this study help us unpack the ways in which access to food, housing, and health care are interconnected.

Unequal classes did emerge in the LCA, which was consistent with our expectation given the inequity that exists among material circumstances and health care access in the US. Classes were named based on item probabilities reflecting high homogeneity: Unbalanced Meals with Health Care, Food and Housing Insecurity with Delayed Health Care, Food and Housing Security with Health Care, and Food Security without Health Care. While one class was characterized by stable access to food, housing, and health care (i.e. Food and Housing Security with Health Care), the other three classes reflect other resource patterns that may emerge among people. The Unbalanced Meals with Health Care class was characterized by being unable to afford balanced meals alongside having access to a health provider and timely medical care. Participants in this class indicated that their financial situation affected their selection of food which could be a result of structural variables, such as food deserts and the availability or cost of healthy foods. Given their proximity to healthcare services, it is possible that participants in this class are engaged in dialogue with healthcare providers about their health but may be unable to act upon medical advice due to financial constraints. In the Food and Housing Insecurity with Delayed Health Care class, participants described experiencing difficulties purchasing enough food, affording food for balanced meals, and paying for rent, mortgage, or utilities. While the probability of having a healthcare provider was high, participants noted delays in healthcare services. These participants reported that their financial situation affected their access to food security, housing stability, and timely medical care. Finally, the Food Security without Health Care class was characterized by food stability without a healthcare provider and delayed healthcare services. While it is unclear whether participants in this class are seeking healthcare access, it is important to acknowledge the likely unavailability of access should the need arise. Reasons for this could include barriers associated with accessing coverage through the Affordable Care Act and prior adverse experiences with the healthcare system (e.g. discrimination).

The variability in how material circumstances and health care access were experienced by this sample may be the result of survival strategies and trade-offs by participants facing oppression. At the personal level, qualitative data have shown that people engage in immediate trade-offs among resources, such as alternating bills paid and foregoing a meal to purchase medication for a child, when faced with financial insecurity (e.g. Knowles et al., 2016). In some ways, these decisions reflect the ways by which individuals and families survive and protect their loved ones; at the same time, these narratives document the severe consequences associated with overlaps with resource insecurity. At the structural level, these classes provide one way to conceptualize how interlocking socioeconomic-related systems (i.e. food, housing, and health care) affect personal health.

Of note, the item asking participants whether they had to move in with family or friends did not distinguish any of the classes in this sample. This item can be conceptualized to capture crowding or even forced moves, two dimensions of housing insecurity (Routhier, 2019). Other data have found that these dimensions are notably less common than housing unaffordability (Routhier, 2019). This is consistent with the current study which found that difficulty paying rent, mortgage, or utility bills was endorsed by 32.8% of the sample while moving in with family or friends was reported by 6.8% of the sample. While the percentage of people in this study who indicated they needed to move in with family or friends is consistent with other metropolitan data (Routhier, 2019), it is possible that the relevance of this item for distinguishing classes may be amplified in a sample where people report more structural barriers related to housing stability (e.g. previous evictions, poor credit history).

In addition to the identification of classes, this study found that class membership was associated with self-reported health outcomes, as expected. The Food and Housing Security with Health Care class reported greater well-being and higher self-rated health than the Unbalanced Meals with Health Care and Food and Housing Insecurity with Delayed Health Care classes. In addition, the Food and Housing Security with Health Care class reported lower BMI than the Unbalanced Meals with Health Care class. No significant differences were found between the Food Security without Health Care class and other classes on these outcomes. Unsurprisingly, these findings highlight that greater stability across material circumstances and health care was associated with better self-reported health, consistent with data focused on one economic resource (Gundersen and Ziliak, 2015; Siefert et al., 2004; Stuff et al., 2004; Suglia et al., 2011). Importantly, findings associated with BMI indicate that these classes may reflect an experience of resources with different implications for health indicators. However, it is crucial to acknowledge the limitations of BMI as a measurement tool, including challenges to its validity in Black, Indigenous, and People of Color (BIPOC) populations (e.g. Burkhauser and Cawley, 2008; Nuttall, 2015). Instead of drawing conclusions about the meaning of different BMI values or other self-reported health outcomes among groups, we view these findings as preliminary support for these classes in their relationships with health and health risk.

Though LCAs offer a meaningful method to capture the variation in social inequity, few studies have employed this analytic approach in understanding socioeconomic inequity. In a study using samples representative of households in Belgium and Great Britain, analyses supported a multidimensional conceptualization of economic marginalization using objective and subjective indicators related to financial stress and housing (Dewilde, 2004). As described earlier, Surachman et al. (2020) found that socioeconomic mobility taxonomies emerged among a sample of White and Black adults and that patterns of socioeconomic mobility were associated with immune dysregulation among White adults. The current study built upon this existing literature by identifying novel classes reflective of material circumstances, arguably the most crucial pathway in socioeconomic health disparities (Solar and Irwin, 2010), and health care access, and demonstrating relationships between these classes with self-reported mental and physical health indicators. Together, these findings highlight the importance of considering a multidimensional conceptualization of economic marginalization and its association with health. Future research examining economic marginalization classes in relation to mental and physical health indicators may inform interventions that can be employed to meaningfully support groups of people experiencing different expressions of socioeconomic-related distress.

In 2019, the American Psychological Association (APA) released guidelines for the Psychological Practice for People with Low-Income and Economic Marginalization (LIEM). In addition to providing important strategies for clinicians to consider in practice, these guidelines underscored critical issues in the scientific study of economic marginalization (American Psychological Association, 2019). Among the issues summarized in the guidelines was an overarching concern of the varied measurement of economic marginalization (American Psychological Association, 2019). This concern has been raised in the past (Adler et al., 1994; American Psychological Association, 2019; Liu, 2011). While studies have established a clear link between economic marginalization and poorer health (e.g. Galobardes et al., 2006; Lorant et al., 2003), targeted efforts to influence policy and further support communities and people facing socioeconomic-related oppression would benefit from quantitative research which considers the intersecting nature of socioeconomic-related variables (e.g. LCA). Future LCAs examining socioeconomic variables across multiple levels (e.g. policy, community, individual) will be critical in addressing socioeconomic health disparities.

Limitations

While this study provides novel data regarding patterns in how people experience material circumstances and health care access in the US and associations with health indicators, findings should be understood within the context of study limitations. This sample was focused on the experiences of adults in the US. Notably, socioeconomic inequities vary internationally (Mackenbach et al., 2008), and future research examining how classes may emerge differently depending on Gini indexes, a country-level indicator of income inequality, would be meaningful. In addition, while the platform used for data collection has been shown to recruit diverse samples (e.g. Casler et al., 2013), there were some notable differences between this sample and the broader US population. For example, although 25% of the sample identified as BIPOC, which is consistent with the US Census Bureau, people who identify as Black/African American, Native Hawaiian/Pacific Islander, and multiracial were underrepresented in this sample compared to the broader US population (United States Census Bureau, n.d.b). Further, people with Bachelor’s degrees had greater representation in this sample than the broader US population (United States Census Bureau, n.d.b). Oversampling people experiencing economic marginalization may be meaningful in future studies. Consistent with intersectionality theory (Crenshaw, 1989, 1991) and data (Surachman et al., 2020; Williams et al., 2016), experiences of economic marginalization materialize differently based on interlocking social identities (e.g. race, gender) and related experiences of oppression and privilege (Diemer et al., 2013). LCAs inclusive of indicators of economic marginalization as well as aspects of oppression associated with other social identities (e.g. racism, sexism) will be fruitful for understanding the nuanced experience of distress related to economic marginalization depending on one’s social location (Diemer et al., 2013). For example, the inclusion of measures assessing disparate health care access and treatment or discrimination (e.g. racism, sexism) may be meaningful when building an economic marginalization taxonomy that more fully considers oppression. While access to health care was assessed using multiple items in this study, the presence or absence of health insurance was not directly measured, nor was the type of government assistance received (if participants indicated yes). This information may provide important contextual details about the sample in future studies. Finally, the current study collected physical health data that was available via self-report. This included BMI, which has been shown to have significant limitations in its application to clinical contexts (e.g. Shachar and Williams, 2017). Other physical health data, such as medical diagnoses, waist-hip ratios, or biological markers (e.g. inflammatory markers) may be relevant depending on research questions and the relationships between researchers and community partners. Researchers may consider these study findings as a launching point to understanding how classes such as these may capture outcomes associated with experiences of economic marginalization.

Conclusion

In sum, the findings of the current study highlight the complexity of the relationship between economic marginalization and self-reported health. Nuanced, person-centered approaches, like LCA, make space to consider patterns of experience that would not be captured with average-centered statistical methods. Notably, four classes emerged with differential health outcomes among those classes. These findings emphasize the importance of economic marginalization in the US at a time when food insecurity, housing instability, and health care access operate in heterogeneous ways. Future research unpacking economic marginalization taxonomies in relation to mental and physical health outcomes relevant in the context of clinical interventions would be fruitful.

Research Data

sj-csv-1-hpq-10.1177_13591053221132899 – for Material circumstances, health care access, and self-reported health: A latent class analysis

sj-csv-1-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).

Supplemental Material

sj-docx-10-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-docx-10-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-docx-12-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-docx-12-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-docx-13-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-docx-13-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-docx-14-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-docx-14-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Research Data

sj-docx-2-hpq-10.1177_13591053221132899 – for Material circumstances, health care access, and self-reported health: A latent class analysis

sj-docx-2-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-docx-8-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-docx-8-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-15-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-15-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-16-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-16-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-17-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-17-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-18-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-18-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-19-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-19-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-20-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-20-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-21-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-21-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-22-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-22-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-23-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-23-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-24-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-24-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-4-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-4-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-5-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-5-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-out-6-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-out-6-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-pdf-7-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-pdf-7-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-pdf-9-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-pdf-9-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Research Data

sj-sav-3-hpq-10.1177_13591053221132899 – for Material circumstances, health care access, and self-reported health: A latent class analysis

sj-sav-3-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Supplemental Material

sj-spv-11-hpq-10.1177_13591053221132899 – Supplemental material for Material circumstances, health care access, and self-reported health: A latent class analysis

Supplemental material, sj-spv-11-hpq-10.1177_13591053221132899 for Material circumstances, health care access, and self-reported health: A latent class analysis by Amanda M Mitchell, Hannah K Heitz, Stephen M Leach and Kate J Berghuis in Journal of Health Psychology

Footnotes

Acknowledgements

We would like to thank our study participants.

Data sharing statement

The current article is accompanied by the relevant raw data generated during and/or analysed during the study, including files detailing the analyses and either the complete database or other relevant raw data. These files are available in the Figshare repository and accessible as Supplemental Material via the Sage Journals platform. Ethics approval, participant permissions, and all other relevant approvals were granted for this data sharing.

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a University of Louisville College of Education and Human Development Research & Faculty Development Grant awarded to Amanda M Mitchell in 2018.

Ethics approval

All procedures performed in the study involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

This study received institutional ethics approval to use a preamble which all participants reviewed.

Welfare of animals

This article does not contain any studies with animals conducted by any of the authors.

ORCID iDs

Amanda M Mitchell

Stephen M Leach

References

Adler

(2009) Health disparities through a psychological lens. American Psychologist 64(8): 663–673.

Adler

Boyce

Chesney

, et al (1994) Socioeconomic status and health: The challenge of the gradient. American Psychologist 49(1): 15–24.

American Psychological Association (2019) Guidelines for psychological practice for people with low income and economic marginalization. Available at: www.apa.org/about/policy/guidelines-lowincome.pdf (accessed 1 October 2020).

Burkhauser

Cawley

(2008) Beyond BMI: The value of more accurate measures of fatness and obesity in social science research. Journal of Health Economics 27(2): 519–529.

Casler

Bickel

Hackett

(2013) Separate but equal? A comparison of participants and data gathered via Amazon’s MTurk, social media, and face-to-face behavioral testing. Computers in Human Behavior 29(6): 2156–2160.

Crenshaw

(1991) Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review 43(6): 1241–1299.

Crenshaw

(1989) Demarginalizing the intersection of race and sex: A Black feminist critique of antidiscrimination doctrine, feminist theory and antiracist politics. University of Chicago Legal Forum 1989(1): 139–167.

DeSalvo

Bloser

Reynolds

, et al (2006) Mortality prediction with a single general self-rated health question. A meta-analysis. Journal of General Internal Medicine 21(3): 267–275.

Dewilde

(2004) The multidimensional measurement of poverty in Belgium and Britain: A categorical approach. Social Indicators Research 68(3): 331–369.

10.

Diemer

Mistry

Wadsworth

, et al (2013) Best practices in conceptualizing and measuring social class in psychological research. Analyses of Social Issues and Public Policy 13(1): 77–113.

11.

Dwivedi

Dubey

Cistola

, et al (2020) Association between obesity and cardiovascular outcomes: Updated evidence from meta-analysis studies. Current Cardiology Reports 22(4): 25.

12.

FAQs (n.d.) Available at: https://www.mturk.com/help

13.

Galobardes

Smith

Lynch

(2006) Systematic review of the influence of childhood socioeconomic circumstances on risk for cardiovascular disease in adulthood. Annals of Epidemiology 16(2): 91–104.

14.

Goodwin

Gazard

Aschan

, et al (2018) Taking an intersectional approach to define latent classes of socioeconomic status, ethnicity and migration status for psychiatric epidemiological research. Epidemiology and Psychiatric Sciences 27(6): 589–600.

15.

Gundersen

Engelhard

Crumbaugh

, et al (2017) Brief assessment of food insecurity accurately identifies high-risk US adults. Public Health Nutrition 20(8): 1367–1371.

16.

Gundersen

Ziliak

(2015) Food insecurity and health outcomes. Health Affairs 34(11): 1830–1839.

17.

Hajos

TRS

Pouwer

Skovlund

, et al (2013) Psychometric and screening properties of the who-5 well-being index in adult outpatients with type 1 or type 2 diabetes mellitus. Diabetic Medicine 30(2): e63–e69.

18.

Harrison

Davis

Smallwood

, et al (2019) Validation of a 2-item food insecurity screen among adult general medicine outpatients [Conference presentation abstract]. In: American college of epidemiology annual meeting, Pasadena, CA, 19 December. https://doi.org/10.1089/pop.2020.0183

19.

Henson

Reise

Kim

(2007) Detecting mixtures from structural model differences using latent variable mixture modeling: A comparison of relative model fit statistics. Structural Equation Modeling A Multidisciplinary Journal 14(2): 202–226.

20.

Horton

Chilton

(2010) The labor economics of paid crowdsourcing. In: Proceedings of the ACM conference on electronic commerce, 209–218. New York, NY: Association for Computing Machinery.

21.

Kaar

Schmiege

Vadiveloo

, et al (2018) Sleep duration mediates the relationship between health behavior patterns and obesity. Sleep Health 4(5): 442–447.

22.

Kino

Sato

Kawachi

(2018) Spillover benefit of improved access to healthcare on reducing worry about housing and meal affordability. International Journal for Equity in Health 17(1): 174.

23.

Knowles

Rabinowich

Ettinger de Cuba

, et al (2016) “Do You Wanna Breathe or Eat?”: Parent perspectives on child health consequences of food insecurity, trade-offs, and toxic stress. Maternal and Child Health Journal 20(1): 25–32.

24.

Krieger

Zimmermann

Huffziger

, et al (2014) Measuring depression with a well-being index: Further evidence for the validity of the WHO Well-being index (WHO-5) as a measure of the severity of depression. Journal of Affective Disorders 156: 240–244.

25.

Kushel

Gupta

Gee

, et al (2006) Housing instability and food insecurity as barriers to health care among low-income Americans. Journal of General Internal Medicine 21(1): 71–77.

26.

Liu

(2011) Social Class and Classism in the Helping Professions: Research, Theory, and Practice. Thousand Oaks, CA: Sage Publications, Inc.

27.

Lorant

Deliège

Eaton

, et al (2003) Socioeconomic inequalities in depression: A meta-analysis. American Journal of Epidemiology 157(2): 98–112.

28.

Mackenbach

Stirbu

Roskam

, et al (2008) Socioeconomic inequalities in health in 22 European countries. New England Journal of Medicine 358(23): 2468–2481.

29.

Martin

Liaw

Bazemore

, et al (2019) Adults with housing insecurity have worse access to primary and preventive care. The Journal of the American Board of Family Medicine 32(4): 521–530.

30.

Masyn

(2013) Latent class analysis and finite mixture modeling. In: Little

(ed.) The Oxford Handbook of Quantitative Methods in Psychology, vol. 2. New York, NY: Oxford University Press, pp.551–611.

31.

Matthews

Gallo

(2011) Psychological perspectives on pathways linking socioeconomic status and physical health. Annual Review of Psychology 62: 501–530.

32.

McLachlan

Peel

(2000) Finite Mixture Models. New York, NY: Wiley.

33.

Moradi

Mirzababaei

Dadfarma

, et al (2019) Food insecurity and adult weight abnormality risk: A systematic review and meta-analysis. European Journal of Nutrition 58(1): 45–61.

34.

Myers

(2020) Food insecurity and psychological distress: A review of the recent literature. Current Nutrition Reports 9(2): 107–118.

35.

Nuttall

(2015) Body Mass Index: Obesity, BMI, and health: A critical review. Nutrition Today 50(3): 117–128.

36.

Nylund-Gibson

Choi

(2018) Ten frequently asked questions about latent class analysis. Translational Issues in Psychological Science 4(4): 440–461.

37.

Nylund-Gibson

Grimm

Masyn

(2019) Prediction from latent classes: A demonstration of different approaches to include distal outcomes in mixture models. Structural Equation Modeling A Multidisciplinary Journal 26(6): 967–985.

38.

Nylund-Gibson

Masyn

(2016) Covariates and mixture modeling: Results of a simulation study exploring the impact of misspecified effects on class enumeration. Structural Equation Modeling A Multidisciplinary Journal 23(6): 782–797.

39.

Nylund

Asparouhov

Muthén

(2007) Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling A Multidisciplinary Journal 14(4): 535–569.

40.

Paolacci

Chandler

(2014) Inside the turk: Understanding mechanical turk as a participant pool. Current Directions in Psychological Science 23(3): 184–188.

41.

Routhier

(2019) Beyond worst case needs: Measuring the breadth and severity of housing insecurity among urban renters. Housing Policy Debate 29(2): 235–249.

42.

Schleider

Weisz

(2015) Using Mechanical Turk to study family processes and youth mental health: A test of feasibility. Journal of Child and Family Studies 24(11): 3235–3246.

43.

Shachar

Williams

(2017) The obesity paradox in cancer – Moving beyond BMI. Cancer Epidemiology Biomarkers & Prevention 26(1): 13–16.

44.

Shapiro

Chandler

Mueller

(2013) Using mechanical turk to study clinical populations. Clinical Psychological Science 1(2): 213–220.

45.

Siefert

Heflin

Corcoran

, et al (2004) Food insufficiency and physical and mental health in a longitudinal survey of welfare recipients. Journal of Health and Social Behavior 45(2): 171–186.

46.

Solar

Irwin

(2010) A conceptual framework for action on the social determinants of health. Discussion paper for the commission on social determinants of health. Geneva: World Health Organization.

47.

Stuff

Casey

Szeto

, et al (2004) Household food insecurity is associated with adult health status. Nutrition Journal 134(9): 2330–2335.

48.

Suglia

Duarte

Sandel

(2011) Housing quality, housing instability, and maternal mental health. Journal of Urban Health 88(6): 1105–1116.

49.

Surachman

Rice

Bray

, et al (2020) Association between socioeconomic status mobility and inflammation markers among white and black adults in the United States: A latent class analysis. Psychosomatic Medicine 82(2): 224–233.

50.

Topp

Østergaard

Søndergaard

, et al (2015) The WHO-5 well-being index: A systematic review of the literature. Psychotherapy and Psychosomatics 84(3): 167–176.

51.

Tucker

Roncoroni

Buki

(2019) Counseling psychologists and behavioral health: Promoting mental and physical health outcomes. The Counseling Psychologist 47(7): 970–998.

52.

United States Census Bureau (n.d.a) American housing survey (AHS) table creator. Available at: https://www.census.gov/programs-surveys/ahs/data/interactive/ahstablecreator.html?s_areas=00000&s_year=2017&s_tablename=TABLES08&s_bygroup1=10&s_bygroup2=1&s_filtergroup1=1&s_filtergroup2=1

53.

United States Census Bureau (n.d.b.) QuickFacts. Available at: https://www.census.gov/quickfacts/fact/table/US/PST045221

54.

United States Department of Agriculture (n.d.a) Food security and nutrition assistance. Available at: https://www.ers.usda.gov/data-products/ag-and-food-statistics-charting-the-essentials/food-security-and-nutrition-assistance/ (accessed 20 March 2020).

55.

United States Department of Agriculture (n.d.b) Key statistics & graphics. Available at: https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-us/key-statistics-graphics/#trends (accessed 20 March 2020).

56.

United States Department of Agriculture (n.d.c) Measurement. Available at: https://www.ers.usda.gov/topics/food-nutrition-assistance/food-security-in-the-u-s/measurement/ (accessed 1 May 2022).

57.

U.S. Department of Labor (n.d.) Minimum wage. Available at: https://www.dol.gov/general/topic/wages/minimumwage (accessed 20 March 2020).

58.

Vargas Bustamante

Fang

Garza

, et al (2012) Variations in healthcare access and utilization among Mexican immigrants: The role of documentation status. Journal of Immigrant and Minority Health 14(1): 146–155.

59.

Ware

Sherbourne

(1992) The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Medical Care 30(6): 473–483.

60.

Williams

Priest

Anderson

(2016) Understanding associations among race, socioeconomic status, and health: Patterns and prospects. Health Psychology 35(4): 407–411.

61.

Willis

Simonsen

Hemmert

, et al (2020) Food insecurity and the risk of obesity, depression, and self-rated health in women. Women’s Health Reports 1(1): 308–317.

62.

World Health Organization (1998) Wellbeing Measures in Primary Health Care/the Depcare Project. Copenhagen: WHO Regional Office for Europe.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.05 MB

0.02 MB

0.01 MB

0.03 MB

0.04 MB

0.05 MB

0.06 MB

0.07 MB

0.08 MB

0.09 MB

0.04 MB

0.09 MB

0.11 MB

0.36 MB

0.11 MB