Socioeconomic Status and Utilization of Health Care Services in Canada and the United States: Findings From a Binational Health Survey

• Published in: Medical care 2009-11, Vol.47 (11), p.1136-1146
• Main Authors: Blackwell, Debra L., Martinez, Michael E., Gentleman, Jane F., Sanmartin, Claudia, Berthelot, Jean-Marie
• Format: Article
• Language: English
• Subjects: Adolescent; Adult; Adults; Age Factors; Aged; Canada; Chronic diseases; Discriminant analysis; Female; Health care surveys; Health care utilization; Health insurance; Health Services - utilization; Health Status; Health surveys; Healthcare Disparities; Hospital utilization rate; Hospitalization; Hospitalization - statistics & numerical data; Humans; Insurance coverage; Insurance Coverage - statistics & numerical data; Insurance, Health - statistics & numerical data; Male; Middle Aged; Office Visits - statistics & numerical data; Patient care; Physicians; Physicians, Family - statistics & numerical data; Regression analysis; Sex Factors; Socioeconomic Factors; United States; Young Adult
• ISSN: 0025-7079; 1537-1948
• DOI: 10.1097/MLR.0b013e3181adcbe9

Objectives: Building on Andersen's behavioral model for the utilization of health care services, we examined factors associated with utilization of physician and hospital services among adults in Canada and the United States, with a focus on socioeconomic status (enabling resources in Andersen's framerork). Methods: Using the 2002–2003 Joint Canada/United States Survey of Health, we conducted country-specific multivariate logistic regressions predicting doctor contacts/visits and overnight hospitalizations in the past year, controlling for predisposing characteristics, enabling resources, and several factors representing perceived need for health care. All analyses were appropriately weighted to yield nationally representative results. Results: Several measures of socioeconomic status—having a regular medical doctor, education, and, in the US income and insurance coverage—were associated with doctor contacts or visits in both countries, along with various predisposing and need factors. However, these same measures were not associated with hospitalizations in either country. Instead, only the individual's predisposing characteristics (eg, age and sex) and his/her need for health care predicted utilization of hospital services in Canada and the United States. Insurance coverage status in the United States became a significant predictor of hospitalizations when count data were analyzed via Poisson regression. Conclusions: Given our particular outcome measures, adults in Canada and the United States exhibited similar patterns of hospital utilization, and socioeconomic status played no explanatory role. However, relative to Canadian adults, we found disparities in doctor contacts among US adults—between those with more income and those with less, between those with health insurance and those without—after adjusting for health care needs and predisposing characteristics.