Ethnicity and Knee Osteoarthritis: A comparative analysis of pain, function, and pain catastrophizing between South Florida Hispanic and Non-Hispanic Adults

doi:10.21203/rs.3.rs-4400220/v1

Download PDF

Research Article

Ethnicity and Knee Osteoarthritis: A comparative analysis of pain, function, and pain catastrophizing between South Florida Hispanic and Non-Hispanic Adults

https://doi.org/10.21203/rs.3.rs-4400220/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objective:

Race is associated with reporting of pain, coping mechanisms, and disease severity in patients with knee osteoarthritis (KOA). However, few studies have evaluated its importance in ethnicity, particularly the Hispanic population. We compare pain perception (VAS), function (WOMAC), and pain catastrophizing (PCS) between Hispanic (HP) and non-Hispanic (NHWP) patients stratified by socio-economic status (SES) and K-L grade.

Methods:

A cross-sectional study of patients from a tertiary care clinic between July 2021 and December 2022 was performed. Patients with knee pain, radiographs, and doctor-diagnosis of KOA completed questionnaires in English or Spanish. Descriptive statistics characterized demographic differences between NHWP and HP in VAS, PCS, and WOMAC. Two one-way analyses of variance evaluated the effect of both ethnicity and sex, with subgroup analyses stratifying by K-L grade. Multivariate general linear models assessed primary outcomes while controlling for confounders.

Results:

HP exhibited higher VAS, PCS, and WOMAC scores compared to NHWP. PCS was higher in HP (p = 0.004, mean = 8.89) than NHWP (mean = 4.58), as was VAS (p < 0.001, mean = 4.28 vs. 2.74) and WOMAC (p = 0.029, mean = 27.86 vs. 21.58). These differences remained when controlled for SES and K-L grade. Stratifying by sex and comparing primary outcomes between HP and NHWP, male HP had greater VAS (p = .021, mean = 3.83 vs. 2.42) and PCS (p = .008, mean = 8.83 vs. 3.35), while female HP had greater VAS (p = .019, mean = 4.62 vs. 3.08) and nonsignificantly greater PCS (p = .164, mean = 8.94 vs. 5.92).

Conclusion:

HP with KOA reported greater pain intensity, functional limitation, and PCS compared with NHWP, even after adjusting for SES and K-L grade.

Orthopedics

Sports Medicine and Kinesiology

Osteoarthritis

pain

knee

ethnicity

Osteoarthritis (OA) is the most common form of arthritis worldwide, with a higher incidence in high-income countries and older individuals. In the United States alone, over 32.5 million individuals have been diagnosed with OA, 3 million of which are Hispanic, an ethnic minority [1–4]. Knee OA (KOA) is the leading cause of lower extremity disability, and as of 2010, was ranked the 11th highest contributor to global disability [5]. Studies including diverse populations are needed to more fully understand the total burden of OA in order to better address unmet needs of both the patient and the medical community serving them.

The diagnosis of OA is usually based on clinical assessment and confirmed by routine plain radiographs[6]. Patients typically report activity-related joint pain that is associated with morning stiffness of less than 30 minutes duration [6, 7]. However, this presentation is not consistent across all patient populations. Like other chronic diseases, the presentation and subsequent morbidity of KOA demonstrates considerable variability across different races and, more recently, ethnicities [8–12]. For example, the Johnston County OA project demonstrated statistically significant differences in both pain severity and radiographic severity (Kellgren and Lawrence [K-L]) scores between African Americans and Caucasians with symptomatic KOA at initial presentation [13, 14]. Other investigators have corroborated this finding by reporting greater pain severity among African Americans compared to non-Hispanic White persons (NHWP) with hip and knee OA [15–21]. Both studies utilized race, an observable characteristic with social implications, as the independent variable between groups. Recently, Hoffman et al. denoted that there are no biological differences in pain experience based on race alone yet there are racial biases in the treatment and assessment of pain [22].

Ethnicity, on the other hand, may or may not carry a different race; however, the term attempts to capture the culture of people in a given geographic region, including their language, heritage, religion and customs [15]. Hollingshead et al. discussed how ethnicity is involved in the processing and reporting of pain [23, 24]. Moreover, a recent study observed that Hispanic persons (HP) report greater pain sensitivity and less pain tolerance than NHWP [23]. Furthermore, a study by Wang studied four different ethnicities (East Asian, African American, Hispanic and Caucasian) and found differences in pain sensitivity between ethnicity after capsaicin administration [25, 26]. Finally, Levy et al. highlighted differences in symptom severity on presentation for a well-known chronic illness, alcoholic liver disease, between Caucasian and Hispanics [27].

Although current data is limited, in addition to studies indicating an association between pain experience, disease severity, and ethnicity, there are findings specific in Hispanics with chronic pain regarding pain intensity and coping mechanisms. Studies utilizing population-based data from the third National Health and Nutrition Examination Survey (NHANES-III) demonstrated that HP with self- reported arthritis are twice as likely to self-report symptoms of activity limitation when compared to NHWP [2]. HPs with an assigned diagnosis of arthritis also tend to report greater pain scores as measured by the Visual Analog Scale (VAS) [15]. Furthermore, when compared to NHWP, HPs have a greater propensity to utilize pain-catastrophizing coping mechanisms to address any physical manifestation of chronic pain [15, 24]. Previous studies have highlighted the psychosocial aspect of arthritic pain and how implications in socioeconomic, psychological, and environmental factors have a substantial influence on an individual’s perception of their self-efficacy and quality of life [26–28].

One maladaptive strategy utilized more commonly in Hispanics for coping with chronic pain is pain catastrophizing. This is a tendency to develop a belief that the pain will not resolve (Helplessness), a constant repetitive dwelling on the negative aspects of the pain (Rumination) and exaggerating/ falsely intensifying the perceived pain stimulus (magnification). Greater pain catastrophizing is associated with greater pain and functional limitation. Additionally, greater pain catastrophizing has been independently associated with lower socio-economic status [9, 29–32]. Green et al. expanded on this claim and underlined that neighborhood socioeconomic status mediated chronic pain experience in younger White and Black individuals [33]. They argued neighborhood socioeconomic status (NSES) mediated the pain experience differences identified between races. Therefore, to explore if differences in pain experiences existed between ethnicity, NSES or a surrogate measure may be noteworthy [31, 33].

The role of socioeconomic factors and variation in values, attitudes, and beliefs regarding medical treatments are not well understood. To our knowledge, there are limited data investigating HP with KOA. One recent study has shown that HP reported higher pain and symptom scores despite fewer changes on plain radiographs compared with NHWP with KOA [34]. The purpose of the present study was to determine if there are differences in pain catastrophizing, functional limitation, and pain severity between HP and NHWP with KOA after adjusting for socioeconomic status (SES) and radiographic disease severity. Investigating a convenience sample of patients from southern Florida, a region somewhat unique with its high prevalence of HP, our hypothesis is that there are differences in reporting of pain catastrophizing, pain severity, and functional limitations between the two ethnicities that will persist after controlling for covariates including SES and KL grade.

Study population and Data Source

All procedures were approved by the University of Miami Institutional Review Board (IRB). Study participants were recruited between July 1st, 2021 and December 22nd, 2022 from our tertiary care medical center in Miami-Dade County, Florida via convenient sampling contingent on investigator (DQ, SH, JPC) availability and patient willingness. Patient eligibility was limited to those eligible to be seen at our clinic (valid insurance coverage, age greater than 18) presenting for the first time to our clinic with knee pain and/or stiffness and subsequently diagnosed with KOA by history, physical exam, and weight-bearing knee radiographs (i.e., AP, lateral and skyline Merchant views). Clinical OA was determined by a sports medicine attending (TMB) during the encounter. OA grade was determined by a blinded musculoskeletal radiologist (J Jose) using the KL grading system (4). Location of OA (tibiofemoral and/or patellofemoral) was also noted. Following patient consent, one of four interviewers (DQ, SH, JPC, EK) administered a series of standard questionnaires in either English or Spanish depending on patient preference. All required data were collected using REDCap, an electronic data capture tool hosted at the University of Miami. The data were handled by a single investigator (DQ), de-identified by random sequence software using a university- approved computer and stored on an encrypted hard drive.

Survey Instrument and Study Measures

We administered a 61-item study questionnaire during the clinical encounter to collect information about participants’ sociodemographic characteristics, including age, sex, occupation, smoking status, ethnicity, and body mass index. As part of the questionnaire, we issued the Charles comorbidity index (CCI) to quantify participants’ chronic comorbid conditions. Nineteen different conditions were assessed among study participants including, myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease (except hemiplegia), dementia, chronic pulmonary disease, connective tissue disease, ulcer disease, mild liver disease, diabetes (without complications), diabetes with end organ damage, hemiplegia, moderate or severe renal disease, solid tumor (metastatic and non-metastatic), leukemia, lymphoma, multiple myeloma, moderate or severe liver disease, and HIV infection. The presence of individual diseases was quantified using a weighted point system and added to produce a single value per participant [31].

Patient pain scores were quantified at the time of presentation to the clinic using a scale from 0 to 10 (visual analog scale - VAS). Before answering, the patient was told that 0 represented being pain-free and 10 represented the worse pain imaginable. We also administered a standardized pain catastrophizing scale (PCS) questionnaire. This scale was developed to assess three components of catastrophizing: rumination, magnification, and helplessness [32]. The scale included 13 validated questions that required responses on a point scale of 0 to 4, and the scores were summed to produce a single value representing the total pain catastrophizing exhibited by each participant [32]. Participant functional limitation was assessed using the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) [8].

Spanish Validation and Testing Administration

Investigator (DQ) is qualified per IRB requirements to gather consent and subsequently administered the 61-item questionnaire in Spanish to study participants. All questionnaires CCI, PCS,VAS and WOMAC were previously validated in Spanish [35–39].

Geocoding Socioeconomic Status

We replicated the approach used by Feldman et al. and utilized the Geographic Information System (GIS) to geocode individual home addresses [29]. For each patient’s individual address, we obtained a Federal Information Processing Standard (FIPS) code and linked it to U.S. Census and American Community Survey data at the block group level. The block group is the smallest geographic unit for which the variables are published. For each block, we collected the Neighborhood Socioeconomic Status (NSES) Index, a previously validated measure for socioeconomic status [40]. This measure includes unemployment (percentage of persons ≥ 16 years of age in the labor force who are unemployed and actively seeking work), income (percentage of persons below the federally defined poverty line and median household income), wealth (median value of owner-occupied homes), education (percentage of persons aged > 25 with less than a 12th-grade education and percentage of persons aged > 25 years with at least four years of college) and crowding (percentage of households containing ≥ 1 person per room). SES index scores were reported on a scale of 0 to 100. NSES index values for each identified block pertaining to a single ethnicity were averaged and reported as a mean and standard deviation.

Data Analysis

Continuous values were reported as the mean and standard deviation for each sample. For descriptive data defined by dichotomous selection, the total percentage within a category was reported.

Comparisons

Sample means were compared between groups via Student’s T-test for comparison of means. Comparisons of the study parameter that failed F-testing required a Mann-Whitney test for averages of unequally distributed sample values. Statistical comparisons of survey data were performed with GraphPad Prism Version 9.4.1 software (GraphPad Software, San Diego, California USA,).

Sub-group Analysis

Descriptive statistics were generated for the data sample with means, medians, standard deviations, and percentiles computed for continuous variables, and counts, means, and medians computed for categorical variables. The three primary dependent variables evaluated included VAS (1–10 scale), PCS, and WOMAC. K-L grade was evaluated as both an independent variable, and as a secondary dependent variable given specific secondary independent variables (Age, BMI, etc.) may influence disease severity radiographically.

ANOVA Analysis

Two one-way analyses of variance (ANOVA) were performed to evaluate the effect of both dichotomous independent variables: Ethnicity (NHWP or HP) and Sex (M or F). WOMAC, PCS, VAS, and K-L grade were analyzed as dependent variables. One-way ANOVAs were also performed comparing WOMAC, PCS, and VAS of ethnicities while stratifying by K-L grade and Sex. A one-way ANOVA was performed comparing the effect of K-L grade on WOMAC, PCS, and VAS, while stratifying by Ethnicity. One-way ANOVAs were also performed to compare means of Age, Height, CCI, and SES between NHWP and HP.

Nonparametric Tests

Nonparametric analyses including median tests and Mann-Whitney tests were used for comparisons of variables with significant outliers and unequally distributed sample values.

Linear Regressions

Linear regressions were performed to evaluate the effects of all continuous or scale independent variables (SES, CCI, Height, BMI, and Age) on each of the three primary dependent variables, as well as K-L grade. Observed cumulative probability was plotted against expected cumulative probability. Standardized residuals were plotted in histogram form to validate sample distribution, and standardized predicted residuals were plotted against actual standardized residuals to check validity of each regression model.

Multivariate Analysis

Multivariate general linear models were computed for primary independent variables and their interaction between each other and various secondary independent variables, including key continuous variables. Selected variables were based on the results of one-way ANOVAs and linear regressions, which demonstrated isolated effects of particular independent variables and therefore had to be cross-analyzed with primary independent variables to see interaction and confounding effects. Variables found to have potential confounding effects on primary independent variables, such as socioeconomic status and K-L grade, were then tested in multivariate general linear models as covariates both separately and in the same model. Post-hoc Tukey and Bonferroni tests were computed for analyses that contained 3 or more subgroups, which for this sample included only K-L grade. Similarly, univariate general linear models were computed for primary independent variables and secondary independent variables found to have statistically significant effects on VAS, PCS, or WOMAC.

Graphs were created to demonstrate both the effects of primary independent variables, as well as primary independent variables stratified by other variables shown to have significant or non-significant effect on VAS, PCS, or WOMAC. Scatterplots were generated to demonstrate correlation and R² values of linear regressions. Mean and median plots were generated, depending on the scale of the dependent variable and the vulnerability of the variable to have significant outliers, such as PCS. All graphs were generated using IBM SPSS statistics version 29.0.2.0 or Microsoft Excel version 16.83.

All statistical analysis was conducted using IBM SPSS Statistics version 29.0.2.0. Statistical significance was considered alpha values < 0.05 and within a 95% confidence interval.

Outcomes by Ethnicity

VAS, PCS, and WOMAC were greater in HP when compared to NHWP (Table 1, Fig. 1). The mean PCS score was greater in HP (p = 0.004) at 8.89 (STE = 0.814, CI = 95% [7.28–10.50]) while NHWP had a mean PCS of 4.58 (STE = 0.862, CI = 95% [2.85–6.31]). For VAS, the mean was again greater in HP (p < 0.001) at 4.28 (STE = 0.236, CI = 95% [3.81–4.74]) while NHWP had a mean VAS rating of 2.74 (STE = 0.313, CI = 95% [2.11–3.37]). The mean WOMAC in HP was greater than NHWP (p = 0.029) with an average of 27.86 (STE = 1.536, CI = 95% [24.82–30.89]) in HP and 21.58 (STE = 1.960, CI = 95% [17.64–25.52]) in NHWP. K-L grade demonstrated no difference between HP and NHWP (p = 0.581) with mean grades of 1.66 (STE = 0.096, CI = 95% [17.64–25.52]) and 1.76 (STE = 1.960, CI = 95% [17.64–25.52]) respectively. When stratifying by ethnicity, K-L grade showed no significant effect on VAS, PCS, or WOMAC in NHWP. However, increased K-L grade showed a significant increase in VAS, PCS, and WOMAC in HP (Table 2). Population sizes for NWHP at each K-L grade ranged from n = 5 to n = 17, while in HP they ranged from n = 14 to n = 52 (Table 3).

In terms of secondary demographic characteristics, height (STE = .577, CI = 95% [66.56–68.88], p = .006), Age (STE = 1.521, CI = 95% [57.12–63.24], p = .028), and SES (STE = 2.33, CI = 95% [59.09–68.45], p < .001) in NHWP were all significantly greater when compared to Height (STE = .328, CI = 95% [65.28–66.57]), Age (STE = .920, CI = 95% [54.36-58.00]), and SES (STE = 1.11, CI = 95% [53.03–57.39]) in HP (Table 1). Otherwise, BMI was less (p = .048) in NHWP (STE = .783, CI = 95% [26.25–29.40]) when compared to HP (STE = .533, CI = 95% [28.80-30.91]). There was no difference in CCI (p = .450) between NWHP (STE = .218, CI = 95% [1.50–2.38]) and HP (STE = .137, CI = 95% [1.47–2.01]). Subjects with tricompartmental disease had significantly lower VAS than Patellofemoral (p = .002) or Patellofemoral + Medial (p = .001) locations. Controlling for location, VAS (p = .005), PCS (p = .030), and WOMAC (p = .033) remained greater in HP compared to NHWP. Ethnicity*K-L grade yields no difference in WOMAC (p = 0.345), PCS (p = 0.292), or VAS (p = 0.322). Stratifying primary outcomes with ethnicity with SES analyzed as a covariate demonstrated a similar difference in PCS (p = 0.008) and VAS (p = 0.003) as reported previously, however in this condition WOMAC is no longer different between NHWP and HP (p = 0.066). When K-L grade is tested as a covariate, WOMAC (p = 0.016), PCS (p = 0.003), and VAS (p < 0.001) remain significantly greater in HP. A model with both SES and K-L grade as covariates maintains and strengthens the statistical difference between NHWP and HP in WOMAC (p = 0.033), PCS (p = 0.005), and VAS (p < 0.001) when compared to solely SES as a covariate.

Outcomes by Sex

VAS (p = .031) and WOMAC (p = .008) was greater in females when compared to males. Mean VAS in females was 4.27 (STE = .277, CI = 95% [3.73–4.82]) while in males it was 3.42 (STE = .278, CI = 95% [2.86–3.97]). Mean WOMAC in females was 29.28 (STE = 1.694, CI = 95% [25.92–32.64]) and in males was 22.63 (STE = 1.824, CI = 95% [19.00-26.25]). Neither PCS (p = .436 ) nor K-L grade (p = .737) was significantly different between sexes. However, mean PCS was greater in females (x̄=8.25, STE = .906, CI = 95% [6.46–10.05]) when compared to males (x̄=7.22, STE = .955, CI = 95% [5.33–9.12]). Analysis of means by Sex*Ethnicity yields no significant difference between WOMAC (.886), PCS (.393), nor VAS (p = .930).

Stratifying by sex and comparing means of primary outcomes was performed (Table 4). In males, HP (x̄=3.83, STE = .352, CI = 95% [3.13–4.53]) had greater mean VAS (p = .021) than NHWP (x̄=2.42, STE = .360, CI = 95% [1.68–3.16]). Male HP (x̄=8.83, STE = 1.260, CI = 95% [6.31–11.34]) had greater PCS (p = .008) than NHWP (x̄=3.35, STE = .776, CI = 95% [1.75–4.95]) (Fig. 5). Male HP (x̄=24.38, STE = 2.265, CI = 95% [19.85–28.91]) had nonsignificantly greater WOMAC (p = .136) than NHWP (x̄=18.38, STE = 2.870, CI = 95% [12.47–24.30]). Female HP (x̄=4.62, STE = .315, CI = 95% [4.00-5.25]) had greater mean VAS (p = .019) than female NHWP (x̄=3.08, STE = .521, CI = 95% [2.01–4.16]). In females, there was no significant difference between HP and NHWP in PCS (p = .164) or WOMAC (p = .177); the mean and median for both were not significantly greater in HP females compared to NHWP females. Median comparisons yield the same results, in which median VAS in females are different between HP and NHWP while median PCS and WOMAC scores in HP females do not significantly differ from NHWP.

Outcomes by SES, CCI, Height, BMI, Age, and KOA Location

Comparison of socioeconomic status (SES) by group ethnicity demonstrated a statistically significant difference between the normalized socio-economic status index of HP and NHWP (p = < 0.0001) non-Hispanics more commonly resided in more affluent communities, had greater income and reported greater values in the remaining 8 parameters accounted for in the NSES index. On average, non-Hispanics on average had an NSES value of 66.9 +/- 12.4 while Hispanics resided in areas with an average NSES value of 55.5 +/- 13.6 (Fig. 2).

Despite these findings, neither SES (p = 0.091), CCI (p = 0.052), Height(p = 0.173), BMI(p = 0.474), or Age (p = 0.537), had any significant relationship with VAS. BMI demonstrated a direct linear relationship with PCS (p = 0.012), with a R² value of 0.038 and an unstandardized beta value of 0.276. SES (p = 0.676), CCI (p = 0.849), Height(p = 0.228), and Age (p = 0.977), showed no relationship with PCS. BMI demonstrated a positive relationship with WOMAC as well (p < 0.001), with a R² value of 0.058 and an unstandardized beta value of 0.676. None of SES (p = 0.382), CCI (p = 0.895), Height(p = 0.0.59), nor Age (p = 0.111), had a relationship with WOMAC. Both Age (p < 0.001, R² = 0.207, B = 0.045) and BMI (p < 0.001, R² = 0.052, B = 0.045) were positively associated with K-L grade. R² value 0.207 for Age, and 0.052 for BMI. SES demonstrated a negative association with VAS (p = 0.028, R² = 0.025, B= -0.03), but no significant relationship with PCS (p = 0.190), WOMAC (p = 0.116), or K-L grade (p = 0.353). When SES is modeled with VAS and stratified by ethnicity, the significant linear relationship no longer exists (Table 5).

In the HP group, there was a comparable prevalence of tricompartmental, medial, and bicompartmental (medial and patellofemoral) OA with percentages of 37%, 31%, and 25% respectively. For the NHWP group, 64% of subjects’ radiographs demonstrated tricompartmental OA, 24% depicted medial compartment OA and 7% were bicompartmental (medial and patellofemoral) (Table 6).

In the present study, we recruited three times as many HP compared to NHWP with KOA. Our sample accurately reflects the ethnicity distribution of people in southeastern Florida and builds upon recent studies that, for the first time to our knowledge, begin to elucidate the burden of KOA in HP [34, 41]. In the current study, HP reported greater VAS, PCS and WOMAC scores compared to NHWP. Perhaps more importantly, these differences persisted after controlling for socioeconomic status and KOA severity, further highlighting the clinical significance of ethnicity in caring for patients with KOA.

Previous studies have demonstrated that Non-Hispanic Blacks report greater values for PCS, VAS, and WOMAC in comparison to NHWP with KOA, similar to our findings for HP [42–46]. These observations are consistent with the findings from 3 other groups [22, 47, 48]. The Dunlop et al. study noted non-Hispanic black and Hispanic older adults reported having symptomatic arthritis at a substantially higher frequency than did non-Hispanic whites [49]. Hollingshead underlined the prevalence of pain catastrophizing in Hispanic adults and reviewed that there are specific cultural values such as religiosity and spirituality that promote an external locus of control for pain experience [23]. Fabian et al. explored these cultural differences and found that catastrophizing varied by ethnicity, with African Americans reporting greater catastrophizing than Asian/Pacific Islanders and Caucasians [50]. The authors also found that situational catastrophizing significantly mediated pain intensity [50]. Similarly, Bishop et al. associated intensification of pain and a greater level of pain catastrophizing for individuals with chronic low back pain [38]. Dagenias et al. conducted a systematic review for acute lower back pain and arrived at the same conclusion; PCS and pain intensity were positively associated [51]. Furthermore, Miller et al. noted that pain catastrophizing is positively associated with pain experience and functional limitation [52]. Moreover, this association is influenced by ethnicity even after accounting for social economic status (SES) and disease severity [52].

To evaluate SES as a possible mediator of pain, pain catastrophizing, and functional limitation in patients with KOA, we utilized GIS software to create a heatmap of patient residence and overlayed this information with census block data. Interestingly we identified that Hispanics tended to be more concentrated in lower SES communities that were situated more inland within Miami-Dade County, Fla. Non-Hispanic participants tended to reside closer to the county coastline and tended to have greater SES status. Yet, after pairing individuals based on NSES value, Hispanics reported greater pain, pain catastrophizing, and functional limitations. These differences can likely be explained in part by two prior investigations using the Catastrophizing subscale of the CSQ* that found individuals of Hispanic ethnicity with chronic pain used catastrophizing as a coping method more than NHWP [44, 52]. Moreover, our study did not analyze the relationships between NSES and PCSs within group therefore a statement of SES and PCS correlation would be inappropriate. Nevertheless, work by Feldman and others have noted that lower SES is usually associated with greater PCS scores possibly due to reduced access to healthcare, inadequate treatment, or specific occupational hazards [29].

As K-L grade increased, a worsening VAS, PCS, and WOMAC trend was seen in both NHWP and HP, though this trend is only significant for each outcome in HP. Interestingly, Hispanic subjects had a greater prevalence of OA in the medial and patellofemoral compartments when compared with the NHWP. We propose that this difference in anatomic distribution among ethnic groups may help to account for the differences in VAS [44]. Further study is needed however to confirm our theory and how that may affect treatment.

As with any study, there are limitations. First, our data collection was carried out at a tertiary clinic, making our patient population biased to those who can access our healthcare center including availability, transportation, and insurance. Second, when utilizing NSES data we inferred individual SES based on neighborhood values which are an approximation and do not report the individual’s level of education, type of occupation nor individual disposable income among other characteristics. Accordingly, this measurement may be subject to error. In the same vein, our analysis was restricted to Miami-Dade County and failed to include 9 individual NHWP who lived in Broward County and 1 individual who lived in Colorado. Given the limited sample size in the sub analysis (n = 40), the inclusion of these individuals would likely have influenced study results. Moreover, we used a cross-sectional study design where our evaluation of pain was at single time point that is likely not representative of the individual patient’s overall pain experience. Our study however has several strengths. Participant sampling was conducted in Miami-Dade County, Florida, with a large diaspora of the Hispanic population providing greater access to this minority population and to our knowledge the largest sample size to date of this minority group. Our data collection tools have been validated for Hispanic persons and were administered in a culturally competent manner. Finally, our GIS application for NSES has been previously utilized as a measure for SES in various populations, including Hispanic persons.

Disparities in the treatment of patients with OA involving different racial, ethnic, and socioeconomic groups are well described and continue to rise [53]. We examined the pain experience of HP and NHWP presenting with symptomatic KOA and demonstrated that differences between the 2 groups are significant for VAS, PCS and WOMAC. More importantly, after adjusting for socioeconomic status and K-L grade, these differences persisted. This observation questions the effect of both K-L grade and NSES on pain, function, and pain catastrophizing coping for individuals with KOA. Given that the treatment of individuals with OA is based largely on patient reported symptoms, our findings suggest including a PCS inventory that provides information to the expected severity pain experience may be of value in the assessment and treatment of these patients.

Authors’ contributions: DQ and TMB conceived the study, participated in its design, coordination, performed statistical analyses, and co-drafted the manuscript. EK, DQ, SH, OP, JPC, J Jahn, AJCM, and TMB collected field data, entered study data, and assisted in interpretation of study results. DQ and LMT performed the statistical analysis and statistical interpretation. DQ, EK, J Jose, J Jahn, LMT, OP, JPC, AJCM and TMB interpreted study results, and helped with the manuscript draft. All authors read, revised, and approved the final manuscript.

Acknowledgements: The authors would like to thank all the participants who took the time to participate in this study.

Funding: This research work was supported by the Soffer Family Foundation. The content is solely the responsibility of the authors and does not necessarily represent the official views of the University of Miami.

Institution and Ethics approval and informed consent: The study research protocol was reviewed and approved by the University Institutional Review Board (#20190389) at the University of Miami.

Statements and Declarations: The authors declare no potential conflicts of interest.

Disclaimers: The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of the University of Miami, Miller School of Medicine.

Lane, N.; Brandt, K.; Hawker, G.; Peeva, E.; Schreyer, E.; Tsuji, W.; Hochberg, M. OARSI-FDA initiative: defining the disease state of osteoarthritis. Osteoarthritis and cartilage 2011, 19, 478-482.
Reyes, C.; Leyland, K.M.; Peat, G.; Cooper, C.; Arden, N.K.; Prieto‐Alhambra, D. Association between overweight and obesity and risk of clinically diagnosed knee, hip, and hand osteoarthritis: a population‐based cohort study. Arthritis & Rheumatology 2016, 68, 1869-1875.
Leyland, K.M.; Gates, L.S.; Sanchez-Santos, M.T.; Nevitt, M.C.; Felson, D.; Jones, G.; Jordan, J.M.; Judge, A.; Prieto-Alhambra, D.; Yoshimura, N. Knee osteoarthritis and time-to all-cause mortality in six community-based cohorts: an international meta-analysis of individual participant-level data. Aging clinical and experimental research 2021, 33, 529-545.
Cisternas, M.G.; Murphy, L.; Sacks, J.J.; Solomon, D.H.; Pasta, D.J.; Helmick, C.G. Alternative methods for defining osteoarthritis and the impact on estimating prevalence in a US population‐based survey. Arthritis care & research 2016, 68, 574-580.
Cross, M.; Smith, E.; Hoy, D.; Nolte, S.; Ackerman, I.; Fransen, M.; Bridgett, L.; Williams, S.; Guillemin, F.; Hill, C.L. The global burden of hip and knee osteoarthritis: estimates from the global burden of disease 2010 study. Annals of the rheumatic diseases 2014, 73, 1323-1330.
Hootman, J.M.; Helmick, C.G.; Barbour, K.E.; Theis, K.A.; Boring, M.A. Updated projected prevalence of self‐reported doctor‐diagnosed arthritis and arthritis‐attributable activity limitation among US adults, 2015–2040. Arthritis & rheumatology 2016, 68, 1582-1587.
Barbour, K.E. Vital signs: prevalence of doctor-diagnosed arthritis and arthritis-attributable activity limitation—United States, 2013–2015. MMWR. Morbidity and mortality weekly report 2017, 66.
Bellamy, N.; Buchanan, W.W.; Goldsmith, C.H.; Campbell, J.; Stitt, L.W. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. The Journal of rheumatology 1988, 15, 1833-1840.
Huang, Y.-q.; Gou, R.; Diao, Y.-s.; Yin, Q.-h.; Fan, W.-x.; Liang, Y.-p.; Chen, Y.; Wu, M.; Zang, L.; Li, L. Charlson comorbidity index helps predict the risk of mortality for patients with type 2 diabetic nephropathy. Journal of Zhejiang University Science B 2014, 15, 58-66.
Bhavsar, N.A.; Gao, A.; Phelan, M.; Pagidipati, N.J.; Goldstein, B.A. Value of neighborhood socioeconomic status in predicting risk of outcomes in studies that use electronic health record data. JAMA network open 2018, 1, e182716-e182716.
Lackland, D.T. Racial differences in hypertension: implications for high blood pressure management. The American journal of the medical sciences 2014, 348, 135-138.
Chia, Y.C.; Beh, H.C.; Ng, C.J.; Teng, C.L.; Hanafi, N.S.; Choo, W.Y.; Ching, S.M. Ethnic differences in the prevalence of knee pain among adults of a community in a cross-sectional study. BMJ open 2016, 6, e011925.
Jordan, J.M.; Helmick, C.G.; Renner, J.B.; Luta, G.; Dragomir, A.D.; Woodard, J.; Fang, F.; Schwartz, T.A.; Abbate, L.M.; Callahan, L.F. Prevalence of knee symptoms and radiographic and symptomatic knee osteoarthritis in African Americans and Caucasians: the Johnston County Osteoarthritis Project. The Journal of rheumatology 2007, 34, 172-180.
Kohn, M.D.; Sassoon, A.A.; Fernando, N.D. Classifications in brief: Kellgren-Lawrence classification of osteoarthritis. Clinical Orthopaedics and Related Research® 2016, 474, 1886-1893.
Campbell, C.M.; Edwards, R.R. Ethnic differences in pain and pain management. Pain management 2012, 2, 219-230.
Rahim-Williams, F.B.; Riley III, J.L.; Herrera, D.; Campbell, C.M.; Hastie, B.A.; Fillingim, R.B. Ethnic identity predicts experimental pain sensitivity in African Americans and Hispanics. Pain 2007, 129, 177-184.
Barger, S.D.; Donoho, C.J.; Wayment, H.A. The relative contributions of race/ethnicity, socioeconomic status, health, and social relationships to life satisfaction in the United States. Quality of Life Research 2009, 18, 179-189.
Glover, T.; Goodin, B.; Horgas, A.; Kindler, L.; King, C.; Sibille, K.; Peloquin, C.; Riley III, J.; Staud, R.; Bradley, L. Vitamin D, race, and experimental pain sensitivity in older adults with knee osteoarthritis. Arthritis & Rheumatism 2012, 64, 3926-3935.
Allen, K.D.; Helmick, C.G.; Schwartz, T.A.; DeVellis, R.F.; Renner, J.B.; Jordan, J.M. Racial differences in self-reported pain and function among individuals with radiographic hip and knee osteoarthritis: the Johnston County Osteoarthritis Project. Osteoarthritis and cartilage 2009, 17, 1132-1136.
Jones, A.C.; Kwoh, C.K.; Groeneveld, P.; Mor, M.; Geng, M.; Ibrahim, S.A. Investigating racial differences in coping with chronic osteoarthritis pain. Journal of cross-cultural gerontology 2008, 23, 339-347.
Creamer, P.; Lethbridge-Cejku, M.; Hochberg, M. Determinants of pain severity in knee osteoarthritis: effect of demographic and psychosocial variables using 3 pain measures. The Journal of rheumatology 1999, 26, 1785-1792.
Hoffman, K.M.; Trawalter, S.; Axt, J.R.; Oliver, M.N. Racial bias in pain assessment and treatment recommendations, and false beliefs about biological differences between blacks and whites. Proceedings of the National Academy of Sciences 2016, 113, 4296-4301.
Hollingshead, N.A.; Ashburn-Nardo, L.; Stewart, J.C.; Hirsh, A.T. The pain experience of Hispanic Americans: a critical literature review and conceptual model. The Journal of Pain 2016, 17, 513-528.
Rabbitts, J.A.; Holley, A.L.; Groenewald, C.B.; Palermo, T.M. Association between widespread pain scores and functional impairment and health-related quality of life in clinical samples of children. The Journal of Pain 2016, 17, 678-684.
Wang, H.; Papoiu, A.; Coghill, R.; Patel, T.; Wang, N.; Yosipovitch, G. Ethnic differences in pain, itch and thermal detection in response to topical capsaicin: African Americans display a notably limited hyperalgesia and neurogenic inflammation. British Journal of Dermatology 2010, 162, 1023-1029.
Feinstein, A.B.; Sturgeon, J.A.; Darnall, B.D.; Dunn, A.L.; Rico, T.; Kao, M.C.; Bhandari, R.P. The effect of pain catastrophizing on outcomes: a developmental perspective across children, adolescents, and young adults with chronic pain. The Journal of Pain 2017, 18, 144-154.
Levy, R.E.; Catana, A.M.; Durbin‐Johnson, B.; Halsted, C.H.; Medici, V. Ethnic differences in presentation and severity of alcoholic liver disease. Alcoholism: Clinical and Experimental Research 2015, 39, 566-574.
Huang, L.-J.; Wu, C.; Yeh, H.-H.E.; Huang, P.-S.; Yang, Y.-H.; Fang, Y.-C.; Wu, C.-T. Higher rumination tendency is associated with reduced positive effects of daily activity participation in people with depressive disorder. Occupational Therapy International 2022, 2022.
Feldman, C.H.; Dong, Y.; Katz, J.N.; Donnell-Fink, L.A.; Losina, E. Association between socioeconomic status and pain, function and pain catastrophizing at presentation for total knee arthroplasty. BMC musculoskeletal disorders 2015, 16, 1-10.
Sullivan, M.J.; Bishop, S.R.; Pivik, J. The pain catastrophizing scale: development and validation. Psychological assessment 1995, 7, 524.
Bonito, A.J.; Bann, C.; Eicheldinger, C.; Carpenter, L. Creation of new race-ethnicity codes and socioeconomic status (SES) indicators for Medicare beneficiaries. RTI International, Agency for Healthcare Research and Quality 2008.
Wolff, B.; Burns, J.W.; Quartana, P.J.; Lofland, K.; Bruehl, S.; Chung, O.Y. Pain catastrophizing, physiological indexes, and chronic pain severity: tests of mediation and moderation models. Journal of Behavioral Medicine 2008, 31, 105-114.
Green, C.R.; Hart-Johnson, T. The association between race and neighborhood socioeconomic status in younger Black and White adults with chronic pain. The Journal of Pain 2012, 13, 176-186.
Novin, S.; Alvarez, C.; Renner, J.B.; Golightly, Y.M.; Nelson, A.E. Features of knee and multijoint osteoarthritis by sex and race and ethnicity: a preliminary analysis in the Johnston County Health Study. The Journal of Rheumatology 2024, 51, 75-83.
Casas Duran, F.; Valduvieco, I.; Oses, G.; Cortés, K.; Barreto, T.; Muñoz-Guglielmetti, D.; Ferrer, F. Spanish validation of Charlson index applied to prostate cancer. Clinical and Translational Oncology 2020, 22, 1187-1192.
García-Campayo, J.; Rodero, B.; del Hoyo, Y.L.; Luciano, J.; Alda, M.; Gili, M. Validation of a Spanish language version of the pain self-perception scale in patients with fibromyalgia. BMC Musculoskeletal Disorders 2010, 11, 1-7.
Campayo, J.G.; Rodero, B.; Alda, M.; Sobradiel, N.; Montero, J.; Moreno, S. Validación de la versión española de la escala de la catastrofización ante el dolor (Pain Catastrophizing Scale) en la fibromialgia. Medicina Clinica 2008, 131, 487-492.
Bishop, M.D.; Lentz, T.A.; George, S.Z. Pathology and intervention in musculoskeletal rehabilitation, Chapter 12; DJ Magee, J.Z., WS Quillen, RC Manske, Ed.; Elsevier Health Sciences: 2015; Volume 3.
Escobar, A.; Quintana, J.; Bilbao, A.; Azkárate, J.; Güenaga, J. Validation of the Spanish version of the WOMAC questionnaire for patients with hip or knee osteoarthritis. Clinical rheumatology 2002, 21, 466-471.
Fullwood, D.; Gomez, R.; Huo, Z.; Cardoso, J.; Bartley, E.; Booker, S.; Powell-Roach, K.; Johnson, A.; Sibille, K.; Addison, A. Pain catastrophizing mediates ethnic/race group differences and pain-related outcomes in adults with or at risk for knee osteoarthritis. The Journal of Pain 2021, 22, 596-597.
Nemati, D.; Quintero, D.; Best, T.M.; Kaushal, N. Investigating the association between knee osteoarthritis symptoms with pain catastrophizing domains between Hispanics and non-Hispanic Whites. Rheumatology international 2023, 1-8.
Terry, E.L.; Fullwood, M.D.; Booker, S.Q.; Cardoso, J.S.; Sibille, K.T.; Glover, T.L.; Thompson, K.A.; Addison, A.S.; Goodin, B.R.; Staud, R.; et al. Everyday Discrimination in Adults with Knee Pain: The Role of Perceived Stress and Pain Catastrophizing. Journal of Pain Research 2020, 13, 883-895, doi:10.2147/JPR.S235632.
Son, K.M.; Hong, J.I.; Kim, D.-H.; Jang, D.-G.; Crema, M.D.; Kim, H.A. Absence of pain in subjects with advanced radiographic knee osteoarthritis. BMC musculoskeletal disorders 2020, 21, 1-9.
Cubukcu, D.; Sarsan, A.; Alkan, H. Relationships between pain, function and radiographic findings in osteoarthritis of the knee: a cross-sectional study. Arthritis 2012, 2012.
Wani, R.T. Socioeconomic status scales-modified Kuppuswamy and Udai Pareekh's scale updated for 2019. Journal of family medicine and primary care 2019, 8, 1846-1849.
Fullwood, D.; Gomez, R.N.; Huo, Z.; Cardoso, J.S.; Bartley, E.J.; Booker, S.Q.; Powell-Roach, K.L.; Johnson, A.J.; Sibille, K.T.; Addison, A.S. A mediation appraisal of catastrophizing, pain-related outcomes, and race in adults with knee osteoarthritis. The journal of pain 2021, 22, 1452-1466.
Gagnon, C.M.; Matsuura, J.T.; Smith, C.C.; Stanos, S.P. Ethnicity and interdisciplinary pain treatment. Pain Practice 2014, 14, 532-540.
Valencia, C.; Robinson, M.E.; George, S.Z. Socioeconomic status influences the relationship between fear-avoidance beliefs work and disability. Pain medicine 2011, 12, 328-336.
Dunlop, D.D.; Manheim, L.M.; Song, J.; Chang, R.W. Arthritis prevalence and activity limitations in older adults. Arthritis & Rheumatism: Official Journal of the American College of Rheumatology 2001, 44, 212-221.
Fabian, L.A.; McGuire, L.; Goodin, B.R.; Edwards, R.R. Ethnicity, catastrophizing, and qualities of the pain experience. Pain Medicine 2011, 12, 314-321.
Dagenais, S.; Haldeman, S. Evidence-based management of low back pain; Elsevier Health Sciences: 2011.
Miller, M.M.; Meints, S.M.; Hirsh, A.T. Catastrophizing, pain, and functional outcomes for children with chronic pain: a meta-analytic review. Pain 2018, 159, 2442-2460.
Faison, W.E.; Harrell, P.G.; Semel, D. Disparities across diverse populations in the health and treatment of patients with osteoarthritis. In Proceedings of the Healthcare, 2021; p. 1421.

Tables 1 to 6 are available in the Supplementary Files section.

The authors declare no competing interests.

Tables1to6.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Ethnicity and Knee Osteoarthritis: A comparative analysis of pain, function, and pain catastrophizing between South Florida Hispanic and Non-Hispanic Adults

Status:

Version 1

Abstract

Figures

INTRODUCTION

METHODS

Study population and Data Source

Survey Instrument and Study Measures

Spanish Validation and Testing Administration

Geocoding Socioeconomic Status

Data Analysis

Comparisons

Sub-group Analysis

ANOVA Analysis

Nonparametric Tests

Linear Regressions

Multivariate Analysis

RESULTS

Outcomes by Ethnicity

Outcomes by Sex

Outcomes by SES, CCI, Height, BMI, Age, and KOA Location

DISCUSSION

CONCLUSION

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 1