Abstract
Background
Objective (SES) and subjective socioeconomic status (SSS) affect symptom intensity and magnitude of limitations. Identification of potentially modifiable social risk factors might contribute to additional opportunities for optimizing musculoskeletal health.
Questions/purposes
(1) There are no correlations between magnitude of limitations (as measured with Patient-Reported Outcomes Measurement Information System Physical Function [PROMIS PF computer adaptive test]) and components of SES or SSS in people with musculoskeletal disease; (2) There are no factors (including level of social deprivation) independently associated with PROMIS PF.
Methods
One hundred and fifty-nine patients presenting to clinicians specializing in the treatment of a broad variety of musculoskeletal conditions were prospectively enrolled in the study. We recorded patient demographics and assessed patients' socioeconomic status using the MacArthur Sociodemographic questionnaire and physical disability rating using PROMIS PF. Patients deprivation index was retrieved using their 9-digit ZIP codes. We used bivariate analysis to determine correlations between magnitude of limitations and socioeconomic status. We created a stepwise backward multivariable linear regression model to assess factors independently associated with PROMIS PF.
Results
Weak correlations were found on bivariate analysis of PROMIS PF with SSS measured as “Place in community” (r 0.28; P < 0.001) and “Place in the United States of America” (r 0.25; P = 0.002). In the multivariable models, the area deprivation index was not independently associated with physical limitations. Male gender (beta regression coefficient [β] 4.1; 95% CI 0.71 to 7.5; P = 0.018) and having net worth of $5000 - $19,999 (β 6.3; 95% CI 0.35 to 12; P = 0.038) or $20,000 - $99,999 (β 5.8; 95% CI 2.1 to 9.5; P = 0.003) when compared to having net worth of less than $4999 were independently associated with better physical function. Being unemployed or disabled and keeping house, being a student, or retired were independently associated with worse physical function (β −12; 95% CI -18 to −7.0; P < 0.001; β −5.6; 95% CI -9.9 to −1.4; P = 0.009, respectively), when compared to working full-time or part-time.
Conclusions
Objective and subjective measures of socioeconomic status are associated with magnitude of physical limitations in patients with musculoskeletal illness. These factors should be considered when developing treatment plans for patients with musculoskeletal conditions.
Level of evidence
Level II prognostic study.
Keywords: Socioeconomic status, MacArthur questionnaire, Musculoskeletal illness, Physical limitations, Social deprivation
1. Introduction
Patient-reported outcome measures (PROMs) quantify illness (symptom intensity and magnitude of limitations). The Patient-Reported Outcomes Measurement System (PROMIS) Physical Function (PF) computer adaptive test is a PROM based on item-response theory that assesses physical limitations and is scaled to population norms and has little to no ceiling or floor effects.1, 2, 3 For a given disease (objective pathophysiology) there is wide variation in symptom intensity and magnitude of limitations. A substantial amount of this variation in illness is accounted for by differences in mental and social health.4,5 Identification of opportunities for improving musculoskeletal health by addressing social determinants of health has the potential to improve physical function, with or without intervention.
Unlike objective socioeconomic status (SES), traditionally assessed using a person's education, occupation and income, subjective socioeconomic status (SSS) is defined as “a person's subjective perception of their rank, relative to others, in the socioeconomic hierarchy”.6 SSS may better account for a more diverse set of current and future markers of SES.7, 8, 9, 10, 11, 12 SES and SSS are both associated with lower scores on a general health patient-reported outcome (the SF-36).7,8 Greater monthly disposable household income and greater value of household possessions are associated with higher self-rated health.13 Socioeconomic status, specifically SSS, is associated with variation in magnitude of physical limitations by socioeconomic class.14,15 Regions where the most marginalized populations live often have a greater burden of illness and this is often paired with less access to care.16 This may lead to progression or exacerbation of disease, greater distress, and less effective coping strategies all contributing to diminished health. Social deprivation is associated with worse physical function, as measured with the Short Physical Performance Battery and the Health Assessment Questionnaire.17, 18, 19, 20 Furthermore, patients who lived in a more socially deprived community, measured using the Distressed Communities Index (DCI) that combines 7 socioeconomic variables to estimate a community's socioeconomic distress using ZIP codes, were more likely to experience major complications.21 Patients were more likely to experience major complications after surgery with an odds ratio of 1.1 per quartile increase in DCI.21 Improvements in social health may alleviate musculoskeletal symptoms and limitations.
This study assessed the association of SES and SSS with magnitude of physical limitations (measured with PROMIS PF). Specifically, the primary null hypothesis was that there is no correlation between magnitude of limitations and SSS and SES among patients with a musculoskeletal disease. Our secondary hypotheses aimed to assess the level of social deprivation independently associated with magnitude of limitations.
2. Methods
2.1. Study design and setting
After institutional review board approval, 159 new and returning adult (aged 18–89 years old), English or Spanish speaking patients presenting to clinicians specializing in the treatment of a broad variety of musculoskeletal conditions of the upper- and lower extremity were invited to participate. Patients were approached in the outpatient clinic setting in a large urban area in conjunction with their specialist appointment to participate in this cross-sectional study over a two-month period. All questionnaires were administered on an encrypted tablet via a secure, HIPAA-compliant electronic platform called REDCap (Research Electronic Data Capture Application) that is a web-based tool for building and managing online surveys and databases).14 Completion of the questionnaires indicated consent. Three patients (1.9%) were not interested in participating, leaving 156 for analysis.
2.2. Outcome measures
Subjects completed a demographic questionnaire, the MacArthur Sociodemographic questionnaire, and PROMIS PF. The demographic questionnaire consisted of 9-digit ZIP code, gender, marital status, race/ethnicity, insurance status, traumatic vs non-traumatic condition, comorbidities (diabetes, musculoskeletal, cardiovascular, pulmonary, other), smoker status, length, and weight (to calculate body mass index; BMI). PROMIS PF served as our proxy for physical disability (i.e. limitations of physical activity), as its questions detail the extent to which patients can complete daily household activities and other tasks requiring physical labor of various degrees.
2.3. Measurements
The MacArthur Sociodemographic questionnaire is a 17-item questionnaire that aims to assess subjective- and objective socioeconomic status. SSS is tested using two ladders as visual analogue scales. People compare themselves to others in terms of their self-defined standing in their community and their standing in the United States whilst taking education, money, and jobs into consideration. SSS was measured from 0 to 10 on a sliding scale, with a higher score indicating a higher standing.22 SES is rated using a variety of multiple-choice questions on income, household composition, education and profession.11,22,23 The sociodemographic questionnaire additionally assesses SES variables like highest education grade completed, the total number of people living in household, number of people living in the household with income, total combined family income, and financial assets (defined as the combined value of all of patients (and their spouse's) checking and savings accounts, and any stocks and bonds). Net worth was defined as financial assets minus any debt (credit card debt, unpaid loans including car loans, home mortgage).22
The Area Deprivation Index (ADI) uses the 9-digit ZIP-code to rank neighborhoods by socioeconomic deprivation when compared to state level or national level of social deprivation. Data from the American Community Survey in 2013 was used to create the index. It is a multidimensional measure of deprivation that accounts for factors like education, employment, housing quality and income. National rankings extend from 1 to 100 in percentiles, and from 1 to 10 in deciles for states. Higher ADI rankings indicate neighborhoods to be more disadvantaged.24
PROMIS PF uses computerized adaptive testing based on item response theory to fit the most appropriate series of questions for each person based on their preceding responses. It uses a minimum of 4 and a maximum of 12 questions and measures limitations of physical activity.25 Items are scored on a 5-point Likert scale. Higher scores indicate better physical function (fewer limitations of physical activity). PROMIS PF is typically completed in less than a minute and has little or no ceiling or floor effects.21,26,27
2.4. Patient characteristics
Our study sample consisted of 156 participants (82 men; 53%), with a mean age of 50 ± 16 years (Table 1). Most persons were white (N = 109; 70%) and married or in a domestic partnership (N = 77, 51%) and with private insurance (N = 94; 60%). Fifty (32%) participants had an income of $75,000 or greater (Table 2). Mean ADI was 3.1 ± 2.4 when ranked state-specific and 39 ± 26 when compared to national levels of disadvantage whereas mean “place in community” was 62 ± 18 and mean place in the USA was 62 ± 20. Mean PROMIS PF score was 45 ± 9.9.
Table 1.
Patient characteristics.
| Variables | N = 156 |
|---|---|
| Age in years¹ | 50 ± 16 (18–81) |
| Men | 82 (53) |
| BMI¹ | 30 ± 7.3 (17–54) |
| Race/Ethnicity | |
| White | 109 (70) |
| Non-white | 47 (30) |
| Marital status¹ | |
| Married | 77 (51) |
| Single | 42 (28) |
| Separated/Dicorced/Widowed | 33 (22) |
| Insurance | |
| Private | 94 (60) |
| Medicare | 24 (15) |
| Other | 33 (21) |
| None | 5 (3.2) |
| Trauma | 53 (34) |
| Smoking | 9 (5.8) |
| Comorbidities | |
| Diabetes | 17 (11) |
| Muskuloskeletal disease | 30 (19) |
| Cardiovascular disease | 27 (17) |
| Pulmonary | 2 (1.3) |
| Other | 27 (17) |
| ADI¹ | |
| State-specific | 3.1 ± 2.4 (1–10) |
| National | 39 ± 26 (1–100) |
| PROMIS PF¹ | 45 ± 9.9 (24–76) |
Continuous variables as mean ± standard deviation (range); Discrete variables as number (percentage); BMI = Body Mass Index (kg/m²); ADI = Area Deprivation Index: National (percentile of block group ADI score); State-specific (decile of block group ADI score); PROMIS PF = Patient-Reported Outcomes Measurement Information System Physical Function; ¹ N = 155 for age, 151 for BMI, 152 for marital status, 139 for state specific and national ADI, 154 for PROMIS PF.
Table 2.
MacArthur sociodemographic questionnaire.
| Variables | N = 154 |
|---|---|
| Place in community | 62 ± 18 (0–100) |
| Place in the USA | 62 ± 20 (0–100) |
| Highest education grade completed | |
| High school or less (grade 1–12) | 36 (23) |
| (Some) College (grade 13–16) | 79 (51) |
| (Some) Graduate or more (grade 17–20+) | 39 (25) |
| Highest degree earned | |
| High school or equivalency (GED) or less | 42 (27) |
| Associate degree (junior college) | 14 (9.1) |
| Bachelor's degree | 54 (35) |
| Masters degree | 24 (16) |
| Doctorate or Professional | 11 (7.1) |
| Other | 9 (5.8) |
| Working status | |
| Working part-time/full-time | 94 (61) |
| Keeping house/Student/Retired | 42 (27) |
| Unemployed/Disabled | 18 (12) |
| Personal income last 12 months | |
| Less than $15,999 | 28 (18) |
| $16,000 - $34,999 | 24 (16) |
| $35,000 - $49,999 | 21 (14) |
| $50,000 - $74,999 | 20 (13) |
| $75,000 and greater | 50 (32) |
| Don't know/No response | 11 (7.1) |
| People in household | |
| One | 31 (20) |
| Two | 58 (38) |
| Three | 24 (16) |
| Four | 25 (16) |
| Five or more | 16 (11) |
| Adults in household | |
| One | 40 (26) |
| Two | 88 (57) |
| Three | 18 (12) |
| Four or more | 8 (5.2) |
| Children in household | |
| None | 102 (66) |
| One | 17 (11) |
| Two | 24 (16) |
| Three or more | 11 (7.1) |
| People in household with income | |
| None | 8 (5.2) |
| One | 71 (46) |
| Two | 62 (41) |
| Three or more | 12 (7.8) |
| Home | |
| Owned | 103 (67) |
| Rented | 41 (27) |
| Other | 10 (6.5) |
| Total combined family income | |
| Less than $15,999 | 13 (8.4) |
| $16,000 - $34,999 | 16 (10) |
| $35,000 - $49,999 | 11 (7.1) |
| $50,000 - $74,999 | 13 (8.4) |
| $75,000 and greater | 87 (56) |
| Don't know/No response | 14 (9.1) |
| Amount of time able to live at current address and standard of living when all sources of family income are lost. | |
| Less than 1 month | 22 (14) |
| 1–2 months | 19 (12) |
| 3–6 months | 27 (18) |
| 7–12 months | 14 (9.1) |
| More than 1 year | 72 (47) |
| Financial assets | |
| Less than $4999 | 33 (21) |
| $5000 - $19,999 | 19 (12) |
| $20,000 - $99,999 | 21 (14) |
| $100,000 - $499,999 | 27 (18) |
| $500,000 and greater | 31 (20) |
| Don't know/No response | 23 (15) |
| Net worth | |
| Less than $4999 | 58 (38) |
| $5000 - $19,999 | 11 (7.1) |
| $20,000 - $99,999 | 13 (8.4) |
| $100,000 - $499,999 | 23 (15) |
| $500,000 and greater | 26 (17) |
| Don't know/No response | 23 (15) |
Continuous variables as mean ± standard deviation (range); Discrete variables as number (percentage).
2.5. Statistical analysis
Continuous data are presented as mean ± standard deviation (SD) with range and discrete variables as numbers and percentages. We used Pearson's correlation coefficient to compare continuous variables, Student's t-test for continuous and dichotomous variables, and one-way analysis of variance (ANOVA) for nominal and continuous variables. We created a stepwise backward multivariable linear regression model to assess factors independently associated with PROMIS PF (Table 3). We included all factors with P < 0.10 on bivariate analysis (Appendix 1). Semi-partial R-squared (R2) expresses the specific variability of a given independent variable in the model and adjusted R2 indicates the amount of variability explained for PROMIS PF. An a priori power analysis indicated that a minimum a sample size of 136 participants would provide 80% statistical power with a small to medium effect size of 0.15, for a regression with 5 predictors. To account for 15% incomplete responses, we enrolled 156 patients.
Table 3.
Stepwise multivariable regression analyses of factors associated with PROMIS PF CAT.
| Dependent variables | Retained variables¹ | Regression coefficient [β] (95% CI) | Standard error (SE) | P value | VIF | Semipartial R² | Adjusted R² |
|---|---|---|---|---|---|---|---|
| PROMIS PF | Sex | 0.26 | |||||
| Women | Reference value | ||||||
| Men | 4.1 (0.71–7.5) | 1.7 | 0.018 | 1.0 | 0.01 | ||
| Working status | |||||||
| Working part-time/full-time | Reference value | ||||||
| Unemployed/Disabled | −12 (−18 to −7.0) | 2.8 | < 0.001 | 1.0 | 0.10 | ||
| Keeping house/Student/Retired | −5.6 (−9.9 to −1.4) | 2.1 | 0.009 | 1.2 | 0.03 | ||
| Net worth | |||||||
| Less than $4999 | Reference value | ||||||
| $5000 - $19,999 | 6.3 (0.35–12) | 3.0 | 0.038 | 1.1 | 0.02 | ||
| $20,000 - $99,999 | 5.8 (2.1–9.5) | 1.9 | 0.003 | 1.2 | 0.03 |
Bold indicates statistically significant difference; ¹ Only significant variables displayed; CI = Confidence interval; VIF = Variance Inflation Factor; PROMIS PF = Patient-Reported Outcomes Measurement Information System Physical Function; Variables applied to stepwise backward regression model (P < 0.10 on bivariate analysis): age, sex, BMI, marital status, insurance status, trauma, diabetes, musculoskeletal comorbidity, cardiovascular comorbidity, pulmonary comorbidity, other comorbidity; ADI = Area Deprivation Index national and state-specific, place in community, place in USA, level of eduction, work status, personal income, people in household, adults in household, children in household, people in household with income, total combined family income, amount of time able to live at current address and standard of living when all sources of family income are lost, financial assets, and net worth.
3. Results
3.1. Correlation between SSS and SES with physical function
For SSS, “Place in community” (r 0.28; P < 0.001) and “Place in the United States of America” (r 0.25; P = 0.002) had weak correlations with PROMIS PF (Appendix 1).
For objective SES, state-specific ADI (r −0.16; P = 0.062) did not correlate with PROMIS PF. ADI, when compared to national level of social deprivation, had a weak negative correlation with PROMIS PF (r −0.19; P = 0.024).
3.2. Factors associated with physical function
Using stepwise backward multivariable linear regression analysis to account for potential confounding, ADI was not retained in the final model of factors associated with PROMIS PF, but male sex (β 4.1; 95% CI 0.71 to 7.5; P = 0.018; semipartial R2 0.01), and net worth of $5000 - $19,999 (β 6.3; 95% CI 0.35 to 12; P = 0.038; semi-partial R2 0.02) or $20,000 - $99,999 (β 5.8; 95% CI 2.1 to 9.5; P = 0.003; semi-partial R2 0.03; Adjusted R2 full model 0.26) compared to net worth of less than $4999 were independently associated with better physical function (Table 3). Being unemployed or disabled (β −12; 95% CI -18 to −7.0; P < 0.001; semipartial R2 0.10) or keeping house, being a student or being retired (β −5.6; 95% CI -9.9 to −1.4; P = 0.009; semipartial R2 0.03) when compared to working full-time or part-time were independently associated with worse physical function. In other words, unemployed or disabled patients had 12 points lower on PROMIS PF than working patients.
4. Discussion
An improved understanding of the effect of SES and SSS on symptom intensity and magnitude of limitations might identify potentially modifiable factors that could alleviate symptoms and improve physical function. This study addressed the influence of objective and subjective measures of socioeconomic status on magnitude of physical limitations.
The results of this study should be interpreted with an awareness of several limitations. First, we analyzed a large number of variables, mainly because the MacArthur Sociodemographic questionnaire had 12 separate questions. To account for this, we checked for indications of collinearity (Appendix 2) and used stepwise backward regression analysis. Given the large number of socioeconomic variables significantly associated with each other and with PROMIS PF in bivariate analysis, we used stepwise regression, which may have overestimated R2 and included fewer socioeconomic variables in the model because of correlation between variables. Second, the most recent ADI uses data from the 2013 American Community Survey, which may be out of date. This urban area is one of the fastest growing cities in the United States with an annual population growth rate approaching 3%.28,29 Additionally, median per capita income, household income, and sales price of homes are increasing.30,31 With this magnitude and pattern of growth, it is likely that the distribution of social disadvantage in this urban area has shifted, causing the ADI to be unrepresentative. In addition, there were 17 new zip codes since 2013 that represented missing values for ADI in the analysis. New neighborhoods in a rapidly developing city may be comprised of more socially advantaged individuals, thereby potentially skewing our results. Third, we did not account for the known influence of psychological factors on measures of symptoms and limitations, which may lead to an overestimation of the independent effect of socioeconomic factors on physical limitations. Fourth, our study addressed a general mix of people, all new and returning persons with a musculoskeletal condition presenting to a specialist in a relatively wealthy urban area. Among persons with specific pathophysiology, from other geographical areas, and at other specific times in the care continuum findings might be different. Fifth, low SES individuals were relatively unrepresented. The associations might be stronger in other populations.
Our finding of a weak correlation between subjective socioeconomic status and physical limitations is consistent with prior studies.14,32 Although subjective ranking may not correspond precisely with objective SES, it might better incorporate past and future circumstances, which – in a way – may make it a type of composite and multidimensional measure of socioeconomic status.11 Prior studies found that higher subjective socioeconomic status is associated with greater self-rated health as measured on a 5-point Likert scale or 10-rung self-anchoring scale.10,11
Social deprivation as measured using ADI was not independently associated with greater physical limitations in this study. Prior research with people with a similar SES distribution found that residents who live in more deprived neighborhoods had a diminished physical function (e.g. mobility difficulties, like walking 100 yards).17, 18, 19, 20 An updated ADI or a population with patients of lower SES might have shown larger correlations. On the other hand, working status (unemployed, disabled, housekeeping, being a student or being retired) and lower net worth were independently associated with greater limitations of physical activity. This is in line with prior research finding associations between components of objective socioeconomic status and physical limitations. Lower level of education, lower per capita and household income, lesser working status and fewer financial assets associated with greater limitations of physical activity as measured by aerobic endurance, gait speed, lower body strength, the performance-oriented mobility assessment, and limitations in instrumental activities of daily living and activities of daily living.14,15,33,34
Our findings emphasize the effect of perceived and objective socioeconomic status on physical limitations. Clinicians and policymaker awareness of these influences can lead to better models of care and improved health. Clinicians might consider accounting for SES and measuring SSS when helping patients with a musculoskeletal disease get and stay healthy. Awareness of the influence of social health on symptoms and limitations might influence clinicians to rely less on low yield tests and discretionary interventions with inconsistent, debated, or limited effects on patient-reported outcomes. This is particularly relevant given the growing number of operative procedures that do not outperform non-operative or sham treatments in randomized trials.35, 36, 37 It also runs counter to current financial incentives and to the hope that people seeking care often place on passive, powerful other approaches to health, which might affect satisfaction scores, recommendations to acquaintances, and referring clinicians. Nevertheless, this study adds to the mounting evidence that the focus in musculoskeletal health might increasingly be on comprehensive care, opening a dialogue with a broader care team that prioritizes attention to mental and social health. Areas for future studies include the influence of psychological factors (e.g. symptoms of depression or the effectiveness of coping strategies) on SSS, and the ability of interventions to improve social health to alleviate symptoms and improve function in people with musculoskeletal illness.
Author contribution
LR, JK, DB, JZ, and SG certify that they have no commercial associations (e.g. consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article.
KB has or may receive payment or benefits in the form of: Grants from Agency for Healthcare Research and Quality (AHRQ), grants from California Public Employees' Retirement System (CalPERS), personal fees from Harvard Business School, personal fees from Centers for Medicare and Medicaid Services, other from American Joint Replacement Registry (AJRR), grants from the Agency for Health Care Research & Quality (AHRQ), other from American Academy of Orthopaedic Surgeons (AAOS), other from American Association of Hip and Knee Surgeons (AAHKS), other from Orthopaedic Research and Education Foundation (OREF), and personal fees from Carrum Health.
DR has or may receive payment or benefits from Skeletal Dynamics, Wright Medical for elbow implants, Deputy Editor for Clinical Orthopaedics and Related Research, Universities and Hospitals, Lawyers outside the submitted work.
Ethical committee approval
This study received approval from the Institutional Review Board of the University of Texas at Austin. This study has been performed in accordance with the ethical standards in the 1964 Declaration of Helsinki. This study has been carried out in accordance with relevant regulations of the US Health Insurance Portability and Accountability Act (HIPAA).
Statement of location
This study was performed at The Dell Medical School – The University of Texas.
Declaration of competing interest
No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
Contributor Information
Léon Rijk, Email: rijk.leon@gmail.com.
Joost T.P. Kortlever, Email: kortlever.joost@gmail.com.
David L.J.I. Bandell, Email: d.bandell@gmail.com.
Juliana Zhang, Email: julianazhang921@gmail.com.
Sean M. Gallagher, Email: sean.gallagher@austin.utexas.edu.
Kevin J. Bozic, Email: kevin.bozic@austin.utexas.edu.
David Ring, Email: david.ring@austin.utexas.edu.
Appendix 1. Bivariate analysis of factors associated with PROMIS PF
| Variables | PROMIS PF | P value |
|---|---|---|
| Age (r) | −0.17 | 0.035 |
| Sex | ||
| Women | 43 ± 9.5 | 0.064 |
| Men | 46 ± 10 | |
| BMI (r) | −0.19 | 0.022 |
| Race/Ethnicity | ||
| White | 45 ± 9.7 | 0.524 |
| Non-White | 44 ± 10 | |
| Marital status | ||
| Married | 46 ± 9.8 | 0.031 |
| Single | 45 ± 11 | |
| Separated/Dicorced/Widowed | 41 ± 8.4 | |
| Insurance | ||
| Private | 47 ± 9.8 | 0.005 |
| Medicare | 41 ± 8.1 | |
| Other | 41 ± 10 | |
| None | 44 ± 8.7 | |
| Trauma | ||
| No | 46 ± 10 | 0.091 |
| Yes | 43 ± 9.4 | |
| Smoker | ||
| No | 45 ± 10 | 0.267 |
| Yes | 41 ± 7.6 | |
| Comorbidities | ||
| Diabetes | ||
| No | 45 ± 9.7 | 0.001 |
| Yes | 37 ± 8.6 | |
| Muskuloskeletal disease | ||
| No | 46 ± 9.8 | 0.013 |
| Yes | 40 ± 9.3 | |
| Cardiovascular disease | ||
| No | 45 ± 9.9 | 0.047 |
| Yes | 41 ± 9.4 | |
| Pulmonary | ||
| No | 45 ± 9.8 | 0.092 |
| Yes | 33 ± 5.1 | |
| Other | ||
| No | 45 ± 9.7 | 0.083 |
| Yes | 42 ± 10 | |
| ADI (r) | ||
| State specific | −0.16 | 0.062 |
| National | −0.19 | 0.024 |
| Place in community (r) | 0.28 | < 0.001 |
| Place in the USA (r) | 0.25 | 0.002 |
| Highest education grade completed | ||
| High school or less (grade 1–12) | 42 ± 11 | 0.021 |
| (Some) College (grade 13–16) | 44 ± 9.1 | |
| (Some) Graduate or more (grade 17–20+) | 48 ± 9.8 | |
| Highest degree earned | ||
| High school or equivalency (GED) or less | 42 ± 9.6 | 0.078 |
| Associate degree (junior college) | 44 ± 9.7 | |
| Bachelor's degree | 45 ± 9.2 | |
| Masters degree | 48 ± 10 | |
| Doctorate or Professional | 49 ± 10 | |
| Other | – | |
| Working status | ||
| Working part-time/full-time | 47 ± 9.6 | < 0.001 |
| Unemployed/Disabled | 36 ± 7.5 | |
| Keeping house/Student/Retired | 43 ± 9.0 | |
| Variables | PROMIS PF | P value |
|---|---|---|
| Personal income last 12 months | ||
| Less than $15,999 | 40 ± 9.6 | 0.022 |
| $16,000 - $34,999 | 44 ± 10 | |
| $35,000 - $49,999 | 44 ± 9.7 | |
| $50,000 - $74,999 | 46 ± 11 | |
| $75,000 and greater | 47 ± 8.9 | |
| People in household | ||
| One | 41 ± 7.7 | 0.003 |
| Two | 45 ± 9.4 | |
| Three | 41 ± 9.3 | |
| Four | 48 ± 10 | |
| Five or more | 50 ± 12 | |
| Adults in household | ||
| One | 42 ± 8.0 | 0.008 |
| Two | 46 ± 9.8 | |
| Three | 40 ± 9.5 | |
| Four or more | 49 ± 14 | |
| Children in household | ||
| None | 44 ± 9.8 | 0.006 |
| One | 40 ± 8.0 | |
| Two | 48 ± 8.9 | |
| Three or more | 51 ± 11 | |
| People in household with income | ||
| None | 40 ± 8.9 | 0.030 |
| One | 43 ± 9.0 | |
| Two | 47 ± 9.7 | |
| Three or more | 44 ± 14 | |
| Home | ||
| Owned | 45 ± 9.6 | 0.245 |
| Rented | 44 ± 9.8 | |
| Other | 40 ± 12 | |
| Total combined family income | ||
| Less than $15,999 | 37 ± 11 | < 0.001 |
| $16,000 - $34,999 | 43 ± 7.3 | |
| $35,000 - $49,999 | 43 ± 11 | |
| $50,000 - $74,999 | 39 ± 5.4 | |
| $75,000 and greater | 48 ± 9.5 | |
| Amount of time able to live at current address and standard of living when all sources of family income are lost. | ||
| Less than 1 month | 39 ± 9.6 | 0.056 |
| 1–2 months | 48 ± 7.2 | |
| 3–6 months | 45 ± 9.4 | |
| 7–12 months | 45 ± 7.1 | |
| More than 1 year | 45 ± 11 | |
| Financial assets | ||
| Less than $4999 | 40 ± 9.6 | 0.038 |
| $5000 - $19,999 | 43 ± 8.6 | |
| $20,000 - $99,999 | 45 ± 11 | |
| $100,000 - $499,999 | 47 ± 8.6 | |
| $500,000 and greater | 46 ± 9.5 | |
| Net worth | ||
| Less than $4999 | 42 ± 9.3 | 0.023 |
| $5000 - $19,999 | 49 ± 9.7 | |
| $20,000 - $99,999 | 44 ± 11 | |
| $100,000 - $499,999 | 47 ± 8.6 | |
| $500,000 and greater | 47 ± 9.6 | |
Bold indicates statistically significant difference; Pearson correlation indicated by (r); Continuous variables as mean ± standard deviation; PROMIS PF = Patient-Reported Outcomes Measurement Information System Physical Function; BMI = Body Mass Index (kg/m²); ADI = Area Deprivation Index.
Bold indicates statistically significant difference; Continuous variables as mean ± standard deviation; PROMIS PF = Patient-Reported Outcomes Measurement Information System Physical Function; Financial assets = checking and savings accounts, and any stocks and bonds; Net worth = Financial assets minus any debt (credit card debt, unpaid loans including car loans, home mortgage).
Appendix 2. Bivariate analyses of socioeconomic variables
| Variables | ADI state specific (r) | ADI national | Insurance status | Highest grade completed | Highest degree earned | Working status | Personal income | People in household with income | Home | Family income | Sources | Financial assets | Net worth |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ADI state specific | x | ||||||||||||
| ADI national | 0.962 (r) P < 0.001 | x | |||||||||||
| Insurance status | P < 0.001 | P < 0.001 | x | ||||||||||
| Highest grade completed | P = 0.002 | P = 0.003 | P < 0.001 | x | |||||||||
| Highest degree earned | P < 0.001 | P < 0.001 | P < 0.001 | P < 0.001 | x | ||||||||
| Working status | P = 0.134 | P = 0.182 | P < 0.001 | P < 0.001 | P = 0.001 | x | |||||||
| Personal income | P < 0.001 | P < 0.001 | P = 0.003 | P < 0.001 | P = 0.001 | P < 0.001 | x | ||||||
| People in household with income | P = 0.075 | P = 0.049 | P = 0.255 | P = 0.034 | P = 0.520 | P = 0.015 | P = 0.003 | x | |||||
| Home | P = 0.015 | P = 0.087 | P = 0.063 | P = 0.046 | P = 0.056 | P = 0.023 | P = 0.017 | P = 0.165 | x | ||||
| Family income | P < 0.001 | P < 0.001 | P < 0.001 | P < 0.001 | P = 0.001 | P < 0.001 | P < 0.001 | P = 0.011 | P = 0.004 | x | |||
| Sources | P = 0.058 | P = 0.047 | P = 0.095 | P = 0.024 | P = 0.001 | P = 0.370 | P = 0.006 | P = 0.024 | P < 0.001 | P < 0.001 | x | ||
| Financial assets | P < 0.001 | P < 0.001 | P = 0.003 | P = 0.027 | P = 0.048 | P = 0.352 | P < 0.001 | P = 0.178 | P < 0.001 | P < 0.001 | P < 0.001 | x | |
| Net worth | 3.0 ± 2.4 P = 0.038 | 38 ± 26 P = 0.009 | P = 0.050 | P = 0.015 | P = 0.028 | P = 0.576 | P < 0.001 | P = 0.302 | P = 0.047 | P < 0.001 | P < 0.001 | P < 0.001 | x |
Bold indicates statistically significant difference; Pearson correlation indicated by (r); Categorical variables by Fisher's exact or Chi-square tests; ADI = Area Deprivation Index; Sources = amount of time able to live at current address and standard of living when all sources of family income are lost.Financial assets = checking and savings accounts, and any stocks and bonds; Net worth = Financial assets minus any debt (credit card debt, unpaid loans including car loans, home mortgage).
References
- 1.Fries J., Rose M., Krishnan E. The PROMIS of better outcome assessment: responsiveness, floor and ceiling effects, and Internet administration. J Rheumatol. 2011;38(8):1759–1764. doi: 10.3899/jrheum.110402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cella D., Riley W., Stone A. The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008. J Clin Epidemiol. 2010;63(11):1179–1194. doi: 10.1016/j.jclinepi.2010.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Rose M., Bjorner J.B., Becker J., Fries J.F., Ware J.E. Evaluation of a preliminary physical function item bank supported the expected advantages of the Patient-Reported Outcomes Measurement Information System (PROMIS) J Clin Epidemiol. 2008;61(1):17–33. doi: 10.1016/j.jclinepi.2006.06.025. [DOI] [PubMed] [Google Scholar]
- 4.Eisenberg L. Disease and illness distinctions between professional and popular ideas of sickness. Cult Med Psychiatr. 1977;1(1):9–23. doi: 10.1007/BF00114808. [DOI] [PubMed] [Google Scholar]
- 5.Urwin M., Symmons D., Allison T. Estimating the burden of musculoskeletal disorders in the community: the comparative prevalence of symptoms at different anatomical sites, and the relation to social deprivation. Ann. Rheum. Dis. 1998;57(11):649–655. doi: 10.1136/ard.57.11.649. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kraus M.W., Adler N., Chen T.W.D. Is the association of subjective SES and self-rated health confounded by negative mood? An experimental approach. Health Psychol. 2013;32(2):138. doi: 10.1037/a0027343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Adler N.E., Epel E.S., Castellazzo G., Ickovics J.R. Relationship of subjective and objective social status with psychological and physiological functioning: preliminary data in healthy, White women. Health Psychol. 2000;19(6):586. doi: 10.1037//0278-6133.19.6.586. [DOI] [PubMed] [Google Scholar]
- 8.Singh-Manoux A., Marmot M.G., Adler N.E. Does subjective social status predict health and change in health status better than objective status? Psychosom Med. 2005;67(6):855–861. doi: 10.1097/01.psy.0000188434.52941.a0. [DOI] [PubMed] [Google Scholar]
- 9.Operario D., Adler N.E., Williams D.R. Subjective social status: reliability and predictive utility for global health. Psychol Health. 2004;19(2):237–246. [Google Scholar]
- 10.Ostrove J.M., Adler N.E., Kuppermann M., Washington A.E. Objective and subjective assessments of socioeconomic status and their relationship to self-rated health in an ethnically diverse sample of pregnant women. Health Psychol. 2000;19(6):613. doi: 10.1037//0278-6133.19.6.613. [DOI] [PubMed] [Google Scholar]
- 11.Singh-Manoux A., Adler N.E., Marmot M.G. Subjective social status: its determinants and its association with measures of ill-health in the Whitehall II study. Soc Sci Med. 2003;56(6):1321–1333. doi: 10.1016/s0277-9536(02)00131-4. [DOI] [PubMed] [Google Scholar]
- 12.Jackman M.R. The subjective meaning ofSocial class identification inthe United States. Publ Opin Q. 1979;43(4):443–462. [Google Scholar]
- 13.Aittomäki A., Martikainen P., Laaksonen M., Lahelma E., Rahkonen O. The associations of household wealth and income with self-rated health–A study on economic advantage in middle-aged Finnish men and women. Soc Sci Med. 2010;71(5):1018–1026. doi: 10.1016/j.socscimed.2010.05.040. [DOI] [PubMed] [Google Scholar]
- 14.Hu P., Adler N.E., Goldman N., Weinstein M., Seeman T.E. Relationship between subjective social status and measures of health in older Taiwanese persons. J Am Geriatr Soc. 2005;53(3):483–488. doi: 10.1111/j.1532-5415.2005.53169.x. [DOI] [PubMed] [Google Scholar]
- 15.Keddie A.M., Peek M.K., Markides K.S. Variation in the associations of education, occupation, income, and assets with functional limitations in older Mexican Americans. Ann Epidemiol. 2005;15(8):579–589. doi: 10.1016/j.annepidem.2004.08.012. [DOI] [PubMed] [Google Scholar]
- 16.Butler D.C., Petterson S., Phillips R.L., Bazemore A.W. Measures of social deprivation that predict health care access and need within a rational area of primary care service delivery. Health Serv Res. 2013;48(2pt1):539–559. doi: 10.1111/j.1475-6773.2012.01449.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.McEntegart A., Morrison E., Capell H.A. Effect of social deprivation on disease severity and outcome in patients with rheumatoid arthritis. Ann. Rheum. Dis. 1997;56(7):410–413. doi: 10.1136/ard.56.7.410. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fox K.R., Hillsdon M., Sharp D. Neighbourhood deprivation and physical activity in UK older adults. Health Place. 2011;17(2):633–640. doi: 10.1016/j.healthplace.2011.01.002. [DOI] [PubMed] [Google Scholar]
- 19.Lang I.A., Llewellyn D.J., Langa K.M., Wallace R.B., Melzer D. Neighbourhood deprivation and incident mobility disability in older adults. Age Ageing. 2008;37(4):403–410. doi: 10.1093/ageing/afn092. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Yen I.H., Kaplan G.A. Poverty area residence and changes in physical activity level: evidence from the Alameda County Study. Am J Publ Health. 1998;88(11):1709–1712. doi: 10.2105/ajph.88.11.1709. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Mehaffey J.H., Hawkins R.B., Charles E.J. Socioeconomic “distressed Communities index” improves surgical risk-adjustment. Annals of Surgery. 2020;271.3:470–474. doi: 10.1097/SLA.0000000000002997. [DOI] [PubMed] [Google Scholar]
- 22.MacArthur Research Network on SES & Health Sociodemographic questionnaire. 2008. https://macses.ucsf.edu/research/socialenviron/sociodemographic.php
- 23.Nielsen F., Roos J.M., Combs R.M. Clues of subjective social status among young adults. Soc Sci Res. 2015;52:370–388. doi: 10.1016/j.ssresearch.2015.02.006. [DOI] [PubMed] [Google Scholar]
- 24.University of Wisconsin School of Medicine and Public Health Area deprivation index. 2013. https://www.neighborhoodatlas.medicine.wisc.edu/
- 25.Döring A.C., Nota S.P., Hageman M.G., Ring D.C. Measurement of upper extremity disability using the patient-reported outcomes measurement information system. J Hand Surg. 2014;39(6):1160–1165. doi: 10.1016/j.jhsa.2014.03.013. [DOI] [PubMed] [Google Scholar]
- 26.Fries J., Rose M., Krishnan E. The PROMIS of better outcome assessment: responsiveness, floor and ceiling effects, and Internet administration. J Rheumatol. 2011;38(8):1759–1764. doi: 10.3899/jrheum.110402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Cella D., Riley W., Stone A. The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008. J Clin Epidemiol. 2010;63(11):1179–1194. doi: 10.1016/j.jclinepi.2010.04.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.World Population Review Austin population. 2018. http://worldpopulationreview.com/us-cities/austin/
- 29.Realpage, Inc . 2018. Texas North Caroline Dominate Fastest Growing Apartment Markets Lists.https://www.realpage.com/analytics/texas-north-carolina-dominate-fastest-growing-apartment-markets-list/ Accessed 11 January. [Google Scholar]
- 30.Dept of numbers Austin–round rock-san marcos Texas household income. 2018. https://www.deptofnumbers.com/income/texas/austin/
- 31.Austintexasgov Planning and zoning. 2018. http://www.austintexas.gov/page/demographic-maps
- 32.Quon E.C., McGrath J.J. Community, family, and subjective socioeconomic status: relative status and adolescent health. Health Psychol. 2015;34(6):591. doi: 10.1037/hea0000135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Noppert G.A., Brown C.S., Chanti-Ketterl M. The impact of multiple dimensions of socioeconomic status on physical functioning across the life course. Gerontol. Geriatr. Med. 2018;4 doi: 10.1177/2333721418794021. 2333721418794021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Beydoun M.A., Popkin B.M. The impact of socio-economic factors on functional status decline among community-dwelling older adults in China. Soc Sci Med. 2005;60(9):2045–2057. doi: 10.1016/j.socscimed.2004.08.063. [DOI] [PubMed] [Google Scholar]
- 35.Marsh J.D., Birmingham T.B., Giffin J.R. Cost-effectiveness analysis of arthroscopic surgery compared with non-operative management for osteoarthritis of the knee. BMJ open. 2016;6(1) doi: 10.1136/bmjopen-2015-009949. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ketola S., Lehtinen J.T., Arnala I. Arthroscopic decompression not recommended in the treatment of rotator cuff tendinopathy: a final review of a randomised controlled trial at a minimum follow-up of ten years. Bone Joint J. 2017;99(6):799–805. doi: 10.1302/0301-620X.99B6.BJJ-2016-0569.R1. [DOI] [PubMed] [Google Scholar]
- 37.Heerspink F.O.L., van Raay J.J., Koorevaar R.C. Comparing surgical repair with conservative treatment for degenerative rotator cuff tears: a randomized controlled trial. J Shoulder Elbow Surg. 2015;24(8):1274–1281. doi: 10.1016/j.jse.2015.05.040. [DOI] [PubMed] [Google Scholar]