Abstract
Background and Objectives:
As suggested by the Minorities’ Diminished Return theory, the association between socioeconomic status and health is weaker for racial and ethnic minorities compared to Whites. The current study compared Blacks and Whites in terms of the association between marital status and physical health.
Methods:
The State of the State Survey (2017) included 881 adults (92 Blacks and 782 Whites) generalizable to the state of Michigan, the United States. The marital status and self-rated physical health (SRPH), which was measured using a single item, were considered as independent and dependent variables, respectively. In addition, age, gender, education, and employment were covariates. Race/ethnicity was regarded as the moderating factor. Logistic regression was used for data analysis.
Results:
Based on the results, being married was associated with better SRPH, the net of all confounders. A significant interaction was found between race and marital status on SRPH, suggesting a larger association for Blacks compared to Whites. In race stratified models, marital status was related to better SRPH for Whites and Blacks, but the magnitude of this link was larger for Blacks compared to Whites.
Conclusion:
Overall, marital status was differently linked to SRPM for Whites and Blacks. Accordingly, policymakers should be cautious while not assuming that diverse racial and ethnic groups with similar economic resources have similar health status.
Keywords: Socioeconomic position, Self-rated physical health, Inequality, Disparities, Race, Ethnic groups, African American, Black
1. Introduction
The link between socioeconomic status (SES) and health outcomes is very well-known1,2 and several SES indicators such as education, employment, income, and marital status are associated with the reduced risk of morbidity and mortality1–5. High SES is also related to self-rated physical health (SRPH)6.
According to the Minorities’ Diminished Return theory, the members of race/ethnic minority groups and White Americans differ regarding the relationship between their SES and health7,8. The protective effects of SES on physical, mental, and oral health is also found to vary for Blacks9–16 and Hispanics10 compared to Whites.
The unequal gain of equal resources across the racial groups is attributed to a number of social processes such as differential access to the opportunity structure and different distributions of societal and everyday barriers in the very daily lives of racial/ethnic groups7,8. Structural factors such as residential segregation and discrimination in education, labor market, banking, and policing increase the costs that the minority populations pay for upward social mobility, thus the link between SES and health would vary for Whites and non-Whites7,8.
Although racial differences regarding the effects of education, employment, and income are confirmed7,8, less is known about the differential association between marital status and health across racial groups17. Therefore, the current study was conducted to examine the link between marital status and SRPH and to test for racial heterogeneity in the above-mentioned relationship. In line with the empirical evidence that suggests SES differently correlates with health for Whites and non-Whites7,8, we expected a weaker association between marital status and SRPH for Blacks in comparison to Whites.
2. Methods
2.1. Design and Setting
Using a cross-sectional design, this study borrowed data from the 2017 State of the State Survey (SOSS), which is a state-wide representative survey of economic and sociopolitical attitudes and beliefs in Michigan, the United States. The SOSS is conducted by the Michigan State University Institute for Public Policy and Social Research, Lansing, Michigan, the United States.
The SOSS collects data using the telephone survey mode and participants include a random sample of approximately 1,000 Michigan residents. The survey takes 20 minutes to complete on average [41]. The SOSS recruited samples utilizing a stratified random sampling method of adults (i.e., age > = 18 years) who live in the state of Michigan, the United States.
2.2. Eligibility
The eligibility for the SOSS included being within the age range of 18 or more, living in Michigan, and having the ability to complete the interview in English. Institutionalized individuals were excluded from the survey. Meanwhile, only adults in households with a phone (landline telephone or Michigan cell phone number) are included since SOSS is a telephone survey.
2.3. Sampling
The SOSS sample is composed of both new and old participants. Up to 80% of the SOSS sample includes new participants, meaning that they are interviewed by the SOSS for the first time. The SOSS sample is drawn from a list of random-digit-dial (RDD) phone numbers for the state of Michigan. A small proportion of the SOSS sample comes from the previous SOSS surveys (i.e., participants who are a part of SOSS surveys during the past two years) and up to 90% of SOSS participants agree to be re-contacted for a re-interview. Both of the above-mentioned SOSS sub-samples are a representative of the random sample. Although many Michigan residents have no landline, the SOSS sampling frame also includes cellphone users. The SOSS sampling frame is provided by Survey Sampling Inc.
A total of 12,007 phone numbers were used for the 2017 SOSS sample. From this number, 584, 5,897, and 6,500 cases were in the re-contact, new RDD segment, and the new cell phone segments, respectively. Overall, 48.2% of the phone numbers were work telephone numbers (79.8%, 50.2%, and 43.6% for the re-contact, the new RDD, and the new cellphone segments, respectively).
2.4. Data Collection
Data were collected by the Institute for Public Policy and Social Research Office for Survey Research. All interviews were conducted between April 19 and July 30, 2017, applying a Computer Assisted Telephone Interviewing system. According to this system, interviews were scripted and executed from a computer workstation. During the interview, the questions and the instructions were provided for the interviewers on their computer screens and the computer indicated what numeric codes or text could be potentially entered as the responses to each item. In addition, Computer Assisted Survey Execution System software (version 5.5) was used for interviews. This system is collectively developed by the U.S. Department of Agriculture, the U.S. Census Bureau, and the University of California, Berkeley.
2.5. Interviewer Training
A total of 38 trained interviewers collected the SOSS data in 2017. Interviewer training covered the study protocol, the interview questionnaire, as well as the meaning and aim of various questions. The interviewers with previous experience only received two hours of training which was specific to the SOSS 2017 while new interviewers received 13 hours of training including the interview practice.
2.6. Ethics
The SOSS study protocol was approved by the Michigan State University Institutional Review Board. All participants provided informed consent and were financially compensated for their time.
2.7. Measures
2.7.2. Dependent Variable
Self-Rated Physical Health (SRPH)
SRPH was regarded as our outcome variable of interest, which was measured by asking the participants “How would you rate your overall physical health?” The response items were in five levels ranging from 1 (excellent) to 5 (poor). The single-item SRPH measure correlates with multiple-item measures of health, physical activity, health behaviors, and well-being. SRPH was dichotomized as poor/fair (1) and good/very good/excellent (0) and high SES was shown to predict better SRPH6.
2.7.1. Independent Variable
Marital status, as the main independent variable, was operationalized as a dichotomous variable married (1) versus unmarried (0).
2.7.3. Covariates
Demographic Factors
Age and gender (male =1 vs. female =0) were regarded as interval and dichotomous measures.
Education and Income
Similarly, sociodemographic factors included education and employment status (labor market participation) and education was a dichotomous variable (college not completed =0 and college completed =1). Further, employment status was another dichotomous measure (0= non-participation in the labor market and 1= labor market participation).
2.7.4. Moderator
Race/ethnicity
Self-identified race/ethnicity was considered as a dichotomous measure as well (Blacks =1 and Whites =0).
2.8. Statistical Analysis
Stata software (Stata Corp., College Station, Texas), version 13.0 was applied to analyze the data and mean (standard errors) and relative frequencies in the overall sample, along with race were utilized to describe our sample. To understand the pattern of bivariate associations, the Pearson correlation test was applied to estimate the correlation matrix between the study variables. It should be noted that the Pearson correlation test was used due to the large sample size. Then, Blacks and Whites were compared in terms of the study constructs employing the Pearson Chi-square test and independent-sample t test. Four logistic regression models were also utilized to perform multivariable analysis. The odds ratio, 95% confidence interval, and P-value were reported as well. In all logistic regression models, marital status and poor/fair SRPH were regarded as the independent and dependent variable while gender, age, education, employment status, and household income were regarded as the control variables. Models and 2 were fitted in the overall sample. Moreover, Model 1 included no interaction term whereas Model 2 encompassed the race by marital status interaction term. Finally, Models 3 and 4 were estimated for Whites and Blacks, respectively.
3. Results
3.1. Descriptive Statistics
Table 1 demonstrates a summary of descriptive information in the overall sample and by race/ethnicity. As shown, Blacks had lower socioeconomic status indicated by a lower frequency of being married, having lower educational attainment, being employed, and earning lower household income in comparison to Whites. In other words, Blacks had worse self-rated physical health (SRPH) compared to Whites.
Table 1.
Descriptive Data of the Overall Sample and by Race/Ethnicity
| Characteristics | All | Whites | Blacks | |||
|---|---|---|---|---|---|---|
| Mean | 95% CI | Mean | 95% CI | Mean | 95% CI | |
| Age (years)* | 48.24 | 46.63–49.86 | 50.25 | 48.57–51.94 | 43.48 | 38.17–48.79 |
| % | 95% CI | % | 95% CI | % | 95% CI | |
| Race | ||||||
| Whites | 84.76 | 80.94–87.93 | ||||
| Blacks | 15.24 | 12.07–19.06 | ||||
| Gender | ||||||
| Women | 52.80 | 48.72–56.84 | 51.98 | 47.75–56.18 | 57.34 | 44.42–69.33 |
| Men | 47.20 | 43.16–51.28 | 48.02 | 43.82–52.25 | 42.66 | 30.67–55.58 |
| Education (≥12 years)*a | ||||||
| Less than college | 55.46 | 51.36–59.49 | 52.47 | 48.22–56.69 | 72.08 | 59.22–82.10 |
| Completed college | 44.54 | 40.51–48.64 | 47.53 | 43.31–51.78 | 27.92 | 17.90–40.78 |
| Employment*a | ||||||
| Not in Labor Force | 36.59 | 32.89–40.45 | 38.62 | 34.68–42.72 | 25.28 | 16.71–36.31 |
| In Labor Force | 63.41 | 59.55–67.11 | 61.38 | 57.28–65.32 | 74.72 | 63.69–83.29 |
Source: The State of the State Survey (2017).
Note.
P<0.05 for Whites versus Blacks;
Pearson Chi-square test; Independent samples t student test; CI: Confidence interval.
3.2. Bivariate Correlations
Likewise, the bivariate correlation matrix in the overall sample is presented in Table 2. Based on the obtained data, race/ethnicity, education employment, and marital status were associated with SRPH. Additionally, Blacks were younger and more employed than Whites. Eventually, Blacks were less likely to be married and had worse SRPH compared to Whites.
Table 2.
Spearman Correlations in the Pooled Sample and by Race/Ethnicity
| Characteristics | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
|---|---|---|---|---|---|---|---|
| All | |||||||
| 1 Race (Black) | 1 | ||||||
| 2 Gender (Men) | −0.05 | 1 | |||||
| 3 Age | −0.17*** | −0.07* | 1 | ||||
| 4 Education (1–4) | −0.05 | −0.01 | −0.05 | 1 | |||
| 5 Employment (In labor force) | 0.08* | 0.12*** | −0.45*** | 0.12** | 1 | ||
| 6 Marital status (Married) | −0.18*** | 0.10** | −0.05 | 0.12** | 0.06# | 1 | |
| 7 Poor SRPH | 0.10** | −0.04 | 0.03 | −0.21*** | −0.20*** | −0.14*** | 1 |
Source: The State of the State Survey (2017).
Note. SRPH: Self-rated physical health; Logistic regressions
P<0.1,
P<0.05
P<0.01
P<0.001
3.3. Multivariable Logistic Regression Models in the Overall Sample
Table 3 provides a summary of the results of two logistic regressions in the pooled sample. Model 1 showed a protective effect of being married against the odds of poor SRPH above and beyond all covariates. In addition, Model 2 documented a significant interaction between marital status and race/ethnicity on poor SRPH.
Table 3.
Association Between Marital Status and Poor SRPH in the Pooled Sample
| Characteristics |
Model 1 Main Effects |
Model 2 Model 1 + Interactions |
||
|---|---|---|---|---|
| B | 95% CI | B | 95% CI | |
| Race (Black) | 1.74* | 1.02–2.97 | 2.44** | 1.31–4.54 |
| Age | 0.99# | 0.98–1.00 | 0.99# | 0.98–1.00 |
| Gender (Men) | 0.81 | 0.56–1.16 | 0.81 | 0.57–1.16 |
| Education (1–4) | 0.59*** | 0.48–0.73 | 0.58*** | 0.47–0.72 |
| Employment (In Labor Force) | 0.35*** | 0.23–0.53 | 0.35*** | 0.23–0.53 |
| Marital Status (Married) | 0.57** | 0.40–0.82 | 0.65* | 0.45–0.95 |
| Marital Status* Race | 0.25* | 0.06–1.00 | ||
Source: The State of the State Survey (2017).
Note. SRPH: Self-rated physical health; Logistic regressions
P< 0.1
P<0.05
P<0.01
P<0.001.
3.4. Logistic Regression Specific to Race/Ethnicity
Data related to both logistic regressions specific for Whites and Blacks are reported in Table 4. Based on Models 3 and 4 in Whites and Blacks, respectively, being married was associated with better SRPH for both groups.
Table 4.
Association Between Marital Status and Poor SRPH in Whites and Blacks
| Characteristics |
Model 3 Whites |
Model 4 Blacks |
||
|---|---|---|---|---|
| B | 95% CI | B | 95% CI | |
| Age | 0.99 | 0.98–1.00 | 0.98 | 0.94–1.02 |
| Gender (Men) | 0.76 | 0.52–1.12 | 1.21 | 0.38–3.86 |
| Education (Completed college) | 0.58*** | 0.46–0.72 | 0.61 | 0.31–1.20 |
| Employment (In Labor Force) | 0.40*** | 0.26–0.61 | 0.18** | 0.06–0.59 |
| Marital Status (Married) | 0.66* | 0.45–0.96 | 0.15** | 0.04–0.61 |
Source: The State of the State Survey (SOSS), 2017.
Note.
P<0.1
P<0.05
P<0.01
P<0.001;
SRPH: Self-rated physical health; Outcome: Poor SRPH.
4. Discussion
The current study was performed to test whether marital status is linked to self-rated physical health (SRPH) and if this link is different for Whites and Blacks. Based on the findings, being married was related to better SRPH, overall, and the marital the status-SRPH link differed between Blacks and Whites.
The result regarding the link between marital status and SRPH is consistent with the Social Determinants of Health18,19 and Fundamental Cause3 theories that consider social resources as the root causes of health. The race is also known to moderate the link between socioeconomic status (SES) and health7,8. However, most of these studies are related to education rather than the other SES indicators including marital status. According to some studies, the effects of education on a wide range of health outcomes such as smoking, drinking, obesity, depression, chronic disease, and mortality are stronger for Whites than Blacks7,8, which is partially because education generates more economic prosperity for White as compared to Black families20.
The current study is not the first one to document that race alters the SES-health link. However, most of the existing literature has focused on the other SES indicators such as education, employment, and income7,8 while little is known about marital status17. The unique contribution of this study is to extend what we know about racial differences in SES-health relationship to the link between marital status and SRPH.
Differential links by race are not limited to health but extend to psychological assets like coping21. The differential effects of psychological assets by race are possibly one of the mechanisms that explains racial and ethnic variation concerning the effects of SES on health since such assets, at least in part, mediate the SES-health link. Accordingly, more research should test if race/ethnic variation in SES-psychological assets and psychological assets – health explains the Black-White differences in the SES-health link.
4.2. Limitations
Our study has some limitations. Although this study used a random sample, participants were limited to individuals who had either a landline or a local cellphone. The study was also limited to individuals who were English speakers. Due to a cross-sectional design, the current study fails to establish causation. In addition, socioeconomic status (SES) and health have bidirectional associations, and reverse causation from poor health to downward social mobility is possible as well. Therefore, future research should investigate the effects of change in social status and marital status over the life course on the health of Blacks and Whites. Moreover, future studies may use multiple and repeated observations that are needed to test the bidirectional link between SES and health. Our outcome (SRPH) was also a single-item measure. Thus, these results should be replicated utilizing other types of data and other physical health outcomes. The omitted confounders were another limitation of the study. This study failed to include important confounders such as health insurance, access to the health care system, and chronic medical disease. Therefore, other studies should seek to find whether health care access and differential treatment in the health care system explain the differential effects of SES on health. Similar to any other cross-racial study, the differential validity of SRPH is a threat to the validity of the current findings. Research has shown that SRPH may reflect different health problems across racial groups. Studying the quality of the relationship between partners and spouses by SES is required to test if social relationships explain the differential effects of marital status for Whites and Blacks. Finally, these findings need to be replicated in other settings, particularly for other racial and ethnic groups.
5. Conclusion
In general, the link between marital status and SRPH may depend on race and ethnicity. Research is still required to better understand group differences respecting the health of various racial groups with similar socioeconomic status resources.
Acknowledgments
The authors wish to thank Mohammed Saqib for his input to this paper.
Funding
Bazargan is supported by the Center for Medicare and Medicaid Services (CMS) Grant 1H0CMS331621. Bazargan and Assari are supported by the NIH under Awards 54MD008149, R25 MD007610, 2U54MD007598, and U54 TR001627. Assari is also supported by D084526–03 and CA201415–02 NIH grants.
Footnotes
Conflicts of Interest
The authors declare no conflicts of interest.
References
- 1.Ross CE, Mirowsky J. Refining the association between education and health: the effects of quantity, credential, and selectivity. Demography. 1999;36(4):445–460. [PubMed] [Google Scholar]
- 2.Ross CE, Mirowsky J. Does employment affect health? J Health Soc Behav. 1995;36(3):230–243. [PubMed] [Google Scholar]
- 3.Phelan JC, Link BG, Diez-Roux A, Kawachi I, Levin B. “Fundamental causes” of social inequalities in mortality: a test of the theory. J Health Soc Behav. 2004;45(3):265–285. [DOI] [PubMed] [Google Scholar]
- 4.Link BG, Phelan J. Social conditions as fundamental causes of disease. J Health Soc Behav. 1995;Spec No:80–94. [PubMed] [Google Scholar]
- 5.Mirowsky J, Ross CE. Education, Health, and the Default American Lifestyle. J Health Soc Behav. 2015;56(3):297–306. [DOI] [PubMed] [Google Scholar]
- 6.Assari S, Moghani Lankarani M. Demographic and socioeconomic determinants of physical and mental self-rated health across 10 ethnic groups in the United States. International Journal of Epidemiologic Research. 2017;4(3):185–193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Assari S Health Disparities due to Diminished Return among Black Americans: Public Policy Solutions. Social Issues and Policy Review. 2018;12(1):112–145. [Google Scholar]
- 8.Assari S Unequal Gain of Equal Resources across Racial Groups. Int J Health Policy Manag. 2017;7(1):1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Assari S, Caldwell CH. Family Income at Birth and Risk of Attention Deficit Hyperactivity Disorder at Age 15: Racial Differences. Children (Basel). 2019;6(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Assari S, Farokhnia M, Mistry R. Education Attainment and Alcohol Binge Drinking: Diminished Returns of Hispanics in Los Angeles. Behav Sci (Basel). 2019;9(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Assari S Family Socioeconomic Position at Birth and School Bonding at Age 15; Blacks’ Diminished Returns. Behav Sci (Basel). 2019;9(3). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Assari S The Benefits of Higher Income in Protecting against Chronic Medical Conditions Are Smaller for African Americans than Whites. Healthcare (Basel). 2018;6(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Assari S, Caldwell CH, Mincy R. Family Socioeconomic Status at Birth and Youth Impulsivity at Age 15; Blacks’ Diminished Return. Children (Basel). 2018;5(5). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Assari S, Hani N. Household Income and Children’s Unmet Dental Care Need; Blacks’ Diminished Return. Dent J (Basel). 2018;6(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Assari S, Mistry R. Educational Attainment and Smoking Status in a National Sample of American Adults; Evidence for the Blacks’ Diminished Return. Int J Environ Res Public Health. 2018;15(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Assari S, Thomas A, Caldwell CH, Mincy RB. Blacks’ Diminished Health Return of Family Structure and Socioeconomic Status; 15 Years of Follow-up of a National Urban Sample of Youth. J Urban Health. 2018;95(1):21–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Assari S, Caldwell CH, Zimmerman MA. Family Structure and Subsequent Anxiety Symptoms; Minorities’ Diminished Return. Brain Sci. 2018;8(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Marmot M Social determinants of health inequalities. Lancet. 2005;365(9464):1099–1104. [DOI] [PubMed] [Google Scholar]
- 19.Economic Marmot M. and social determinants of disease. Bull World Health Organ. 2001;79(10):988–989. [PMC free article] [PubMed] [Google Scholar]
- 20.Assari S Parental Education Better Helps White than Black Families Escape Poverty: National Survey of Children’s Health. Economies. 2018;6(2):30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Assari S General Self-Efficacy and Mortality in the USA; Racial Differences. J Racial Ethn Health Disparities. 2017;4(4):746–757. [DOI] [PMC free article] [PubMed] [Google Scholar]