A secondary analysis was performed using data from the Osteoarthritis Initiative (OAI). The OAI is a publicly and privately funded multicenter, longitudinal study to examine the onset and progression of knee OA. A total of 4,796 participants in the OAI study were recruited from four clinical sites in the United States (US) (Baltimore, Maryland; Pittsburgh, Pennsylvania; Pawtucket, Rhode Island; and Columbus, Ohio) between February 2004 and May 2006. An overview of the OAI is provided elsewhere . The Institutional Review Board of the University of California (Coordinating Center), San Francisco has approved the OAI protocol. Each participant was instructed about the objectives and procedures of the study by the investigator at each site. All the participants were provided with informed consent forms, which they signed before enrollment in the OAI study.
In the present cross-sectional study, data of 4,674 participants aged 45–79 years with or at high risk for knee OA were included. Based on the racial and ethnic background, all participants were classified as Caucasians and American minorities, such as African Americans, Asians, and other non-whites. Participants (n = 122) who had no pain, aching, or stiffness in either knee in the past year; no radiographic finding of OA; and no eligibility risk factor of OA were excluded from the analysis.
Access to HC was defined by asking the self-reported question: “Where do you usually go for health care or advice about your health care?” Answers included “private doctor,” “public clinic,” “health maintenance organization (HMO),” “hospital clinic,” “emergency room,” and “other.” Similar questions were posed in the 2001–2003 Medicare Current Beneficiaries Survey and the 2002 Area Resource File . The HCC statuses of participants were determined by asking the self-reported question: “Do you currently have any kind of health care coverage?” The answers to this question included private health insurance (such as Blue Cross), prepaid plans (such as those of Health Maintenance Organizations [HMOs]), and Preferred Provider Organizations (PPOs) or any government-sponsored plans such as Medicare, Medicaid, or Veterans’ Affairs (VA) coverage. Participants’ insurance statuses were determined by asking the self-reported question: “Do you have any health insurance plan that pays for all or part of the cost of prescription medicines?” Similar questions were posed in a previous study . Based on the presence or absence of HCC and insurance, access to HC has been classified into four levels: 1) presence of both HCC and insurance, 2) presence of HCC and absence of insurance, 3) the absence of HCC and presence of insurance, and 4) absence of both HCC and insurance.
Participants’ sociodemographic and smoking status, as well as body mass index (BMI), were collected. Based on self-reports, participants’ gender (male or female), household composition (living alone or with others), and income per year in United States dollars (<50,000 or ≥50,000) were also recorded. Education was categorized into three levels: primary school or less, college graduate, and graduate or beyond. Age was used as continuous to show the age difference and also stratified by a five-year gap (<50, 50–54, 55–59, 60–64, and ≥65) to show the presence or absence of HCC and insurance between both groups. Data on marital (married and unmarried/divorced/widow) and smoking (none- and current or former smoker) status were included. BMI was calculated using participants’ measured weight in kilograms divided by height in square meters. According to the World Health Organization , BMI was classified into three levels: normal weight (18.5—24.9), overweight (25-29.9), and obese (30 or above).
The sample characteristics were described using means and standard deviation (SD) for the continuous variables and frequencies for the categorical variables. The significant difference between Caucasians and American minorities was determined by utilizing a t-test for the continuous variables and a chi-square test for the categorical variables. For Caucasians and American minorities, the distribution of sociodemographic data by age group, gender, educational level, income per year, household composition, marital status, smoking status, and BMI according to the presence and absence of HCC and insurance were provided in the count (percentages). The significance between the sociodemographic, HCC, and insurance statuses was examined using a chi-square test in both groups. A multinomial regression model was used to identify the factors behind insufficient access to HC in Caucasians and American minorities. All the analyses were performed using the Statistical Analysis Software (SAS) for Windows version 9.4 (SAS Institute, Inc., Cary, NC, US).