Search

BACKGROUND Although there is increasing knowledge regarding the association between generational status on diabetes risk, less is known about the effects of generational status on diabetes management among Mexicans. OBJECTIVES We test whether generational status is associated with variations in diabetes processes of care among Mexican adults to optimize disease management. RESEARCH DESIGN Weighted multivariate logistic regression was used to test the association between generational status on the quality of diabetes care processes and health care utilization adjusting for socioeconomic/demographic factors. Four generational cohorts were constructed: first generation immigrant; second generation; generation 2.5; and third generation. SUBJECTS We conducted analyses of 3072 self-identified foreign and US-born Mexican adults with diabetes who participated in the 2005, 2007, 2009, and 2011/2012 California Health Interview Surveys. MEASURES Three diabetes process outcomes were measured, including receipt of at least 1 eye examination, 1 foot examination, or 1 hemoglobin A1C test (HbA1c) in the past year. One secondary prevention outcome was measured, receipt of an annual influenza vaccine. Finally, we measured whether an individual with diabetes received an annual doctor visit. RESULTS A gradient of progressively decreased odds of receiving an eye examination and HbA1c test were observed and significant for most of the higher generations. Higher generations also had significantly decreased odds of receiving an annual flu vaccine. CONCLUSIONS Data collection on generational granularity and a specific focus on mixed acculturation status can vitally inform health system improvements in diabetes care among Mexicans in the United States.

Background: Few studies have explored health care market structure and colorectal cancer (CRC) screening test use, and little is known whether market factors contribute to racial/ethnic screening disparities. Objective: We investigated whether HMO market level factors, controlling for individual covariates, differentially impact Asian American and Pacific Islander (AAPI) subjects' access to CRC screening compared with white subjects Research Design and Methods: We used random intercept hierarchical models to predict CRC test use. Individual-level survey data was linked to market data by metropolitan statistical areas from InterStudy. Subjects: Insured first-degree relatives, ages 40-80, of a random sample of colorectal cancer cases identified from the California Cancer Registry 515 white subjects and 396 AAPI subjects residing in 36 metropolitan statistical areas (MSAs) Measures: Dependent variables were receipt of (1) annual fecal occult blood test only; (2) sigmoidoscopy in the past 5 years; (3) colonoscopy in the past 10 years; and (4) any of these tests over the recommended time interval. Market characteristics were HMO penetration, HMO competition, and proportion of staff/group/network HMOs. Findings: Market characteristics were as important as individual-level characteristics for AAPI but not for white subjects. Among AAPI subjects, a 10% increase in the percent of group/staff/network model HMO was associated with a reduction in colonoscopy use (28.9% to 20.5%) and in receipt of any of the CRC tests (53.2% to 45 4%). Conclusions: The prevailing organizational structure of a health care market confers a penalty on access to CRC test use among high-risk AAPI subjects but not among high-risk white subjects. Identifying the differential effect of market structure on race/ethnicity can potentially reduce the cancer burden among disadvantaged racial groups.

Background: This study examines the factors associated with discontinuous health insurance coverage without periods of uninsurance during the past year (ie, switching coverage), and whether it has a detrimental effect on basic access to care. Research Design: We analyze the 2003 California Health Interview Survey samples of adults, ages 19-64 (n = 32,850) and children, ages 0-18 (n = 13,062), using weighted bivariate and multivariate analyses. We stratified the population first by age (modeling adults separately from children) and then by health status (modeling the full population and the population reporting fair or poor health separately). Results: Income, race/ethnicity, age, gender, and rural status were significant factors associated with switching coverage. Adults who switched insurance had significantly reduced odds of having a usual source of care [odds ratio (OR) = 0.63, P < 0.001)] compared with those with continuous coverage. In addition, adults and children who switched coverage were significantly more likely to report delaying care because of cost or insurance issues (adults: OR = 1.65, P < 0.001; children: OR = 2.00, P < 0.001). Children in fair or poor health who switched insurance coverage had much higher odds of reporting a delay in care (OR = 5.48, P < 0.001). Conclusions: Children and adults had disruptions in their basic access to health care when they experienced discontinuous insurance. These findings highlight the advantages of retention of enrollees as one means of promoting access to health care, in the short term, and the benefit of a continuous national health insurance program in the long term.

Objective: We sought to determine whether primary language use, measured by language of interview, is associated with disparities in cervical cancer screening. Data sources: We undertook a secondary data analysis of a pooled sample of the 2001 and 2003 California Health Interview Surveys. The surveys were conducted in English, Spanish, Cantonese, Mandarin, Korean, and Vietnamese. Study Design: The study was a cross-sectional analysis of 3-year Pap test use among women ages 18 to 64, with no reported cervical cancer diagnosis or hysterectomy (n = 38,931). In addition to language of interview, other factors studied included race/ethnicity, marital status, income, educational attainment, years lived in the United States, insurance status, usual source of care, smoking status, area of residence, and self-rated health status. Data Collection/Extraction Methods: We fit weighted multivariate logit models predicting 3-year Pap test use as a function of language of interview, adjusting for the effects of specified covariates. Principal Findings: Compared with the referent English interview group, women who interviewed in Spanish were 1.65 times more likely to receive a Pap test in the past 3 years. In contrast, we observed a significantly reduced risk of screening among women who interviewed in Vietnamese (odds ratio [OR] 0.67; confidence interval [CI] 0.48–0.93), Cantonese (OR 0.44; 95% CI 0.30–0.66), Mandarin (OR 0.48; 95% CI 0.33–0.72), and Korean (OR 0.62; 0.40–0.98). Conclusions: Improved language access could reduce cancer screening disparities, especially in the Asian immigrant community.