Healthcare services preferences: Dimensions of uninsured consumers priorities and choices

Background: Access to health care continues to pose problems for a significant portion of Americans. Over the last decade, the number of uninsured Americans has increased from approximately 39 million to 46 million (U.S. Department of Health and Human Services, 2011). Communities have developed local solutions to address the health care needs of uninsured populations. One such organization is the Sedgwick County Health Access Project in south central Kansas. This is a collaborative partnership between public and private sectors to support planning efforts for an insurance vehicle to cover uninsured patients. Methods: A convenience sample survey that explored county residents' perceptions toward affordability and availability of healthcare services was conducted. One item asked respondents to rank order, by perceived need for their families and themselves, eight healthcare services (primary care, hospitalization, emergency care, specialty care, tests, mental health, pharmacy, or other services such as chiropractic, home health care, physical therapy, etc.). Of the 385 respondents (ages 18 to 64 years), only 96 uninsured respondents ranked the healthcare services. A hierarchical cluster analysis using Ward's method was performed to categorize respondents by their preferences. Additionally, multidimensional unfolding analysis, a type of multidimensional scaling analysis (MDS), was used to uncover additional dimensions associated with respondents' service priority. Results: Cluster analysis identified three groups with varying preferences for healthcare services. Service preferences in group 1 (primary care, lab tests and prescription medication support) suggested needs of an acute nature. Group 2 preferences (specialty care and hospitalization) suggested experience with and need for chronic disease care. Group 3 showed another pattern, a preference for emergency services only. No participants indicated a preference for mental health services. Multidimensional unfolding analysis further described these clusters by classifying participants along three dimensions that suggested higher versus lower perceived need for services. Dimension 1 reflected participants' likeliness to use the particular service. Dimension 2 differentiated between preferences for services that might be needed by people with chronic disease(s). Dimension 3 reflected perceptions of the cost associated with services. Discussion: Preferences for healthcare services may differ based upon previous experience with healthcare insurance. Psychometric analyses may uncover additional dimensions of preference that would not be revealed by standard statistical analyses. This study's outcome shows that cluster analysis and MDS reveal critical information when analyzing perception datasets. These techniques may be helpful in guiding policymakers and clinics that serve the uninsured population.