Clinical and Health Affairs
Access to Health Care
Differences between Insured and Uninsured Patients in South Minneapolis
By Richard Adair, M.D., Amy Greminger, M.D., and Bryan Post, M.D.
Abstract
This article reports the results of a survey of 575 patients who were seen at 2 clinics in south Minneapolis. About half had health insurance. We found a strong connection between not having insurance and going without medical care. We also found that when the uninsured do go for care, they often fail to take prescribed medications because of the cost. Our findings also contradicted some prevalent thinking about the uninsured: that they are unemployed, that they refuse employer-sponsored insurance, and that they can easily get care at free clinics.
Figuring out the best way to provide medical care for low-income persons is a widely debated topic. Employers and state government struggle to contain health care costs. At the same time, the number of uninsured Minnesotans is rising.1 Consequently, hospitals and clinics are finding themselves providing more and more uncompensated care.2
One reason why policymakers haven’t solved the problem of how to care for the uninsured is a lack of data about exactly who the uninsured are and whether lack of insurance is truly a barrier to care. Surveys show that most Americans believe the uninsured can obtain the care they need, are mostly unemployed, and belong to certain demographic groups.3 Some believe that many are immigrants or young people who elect to go without employer-sponsored insurance in order to save money.
Because we work in clinics where many low-income people receive care, we decided to ask our patients about their insurance status and their experiences in obtaining health care.
Methods
Patients younger than 65 years of age who registered for care at 2 south Minneapolis clinics during an 8-week period in February and March of 2005 were asked to complete a 1-page survey asking whether they had health insurance, whether they were currently employed, whether they had avoided going to the doctor, and whether they had skipped prescription medications by either not buying them or stretching their supply in the last year because of cost. Uninsured patients were asked more detailed questions about employment status. Participants completed the study in the clinics’ waiting rooms.
One of the clinics provided free care; the other was a fee-for-service internal medicine primary care clinic. Both serve inner-city, multicultural patient populations. Interpreters were available to help non-English-speaking patients complete the survey. Patients who reported having high-deductible insurance policies (deductibles >$500 per year) were classified as uninsured because they pay cash for most outpatient visits. We did not ask patients whether they had private insurance or government-subsidized insurance because both provide access to care.
Responses for insured and uninsured patients were compared using Fisher’s 2-tail exact test for categorical values and the chi-square whole-model test for continuous variables.
Results
♦ Access
Ninety-six percent (554 of 575) of the surveys were returned; about half came from each clinic. Of the respondents, 308 (56%) reported being uninsured and 246 (44%) insured. Of the uninsured patients, 82% reported avoiding going to the doctor during the last year because of cost, compared with 31% of the insured patients. Fifty-two percent reported not taking prescription medicines in the last year because of the cost, compared with 28% of insured patients (Table).
♦ Employment
Seventy-four percent of the uninsured patients reported they were employed. Eighteen percent were full-time students. The uninsured patients, including students, worked a mean of 31 hours per week. The most common occupations were food service, retail, industry/assembly, self-employment, construction/maintenance/cleaning, and office administration (Figure). Eighty-three percent reported that health insurance was not available through their employer. Of the 17% whose employers did offer health insurance, all said they declined it because they could not afford it. Only 20% of uninsured patients said that insurance was available at their previous job. The median length of time without health insurance was 2 years.
♦ Demographics
Uninsured patients were younger (median age, 29 years) than insured patients (median age, 43 years), more likely to be women (68%); more likely to have been born in the United States (89%); and somewhat more likely to live in the suburbs (29%) (Table).
Of the uninsured patients, 68% reported their heritage as European, 12% as African, 5% as Asian, 3% as Latin, and 1% as Native American. Twelve percent did not specify their ethnic heritage. Of the insured patients, 44% reported their heritage as European, 29% as African, 3% as Asian, 3% as Latin, and 2% as Native American. Twenty percent did not provide the information. Although the uninsured group included more patients with European ancestry and the insured group had more patients of minority ancestries, we did not attempt a statistical analysis of this data because of uncertainty about the patients who did not answer the question.
Discussion
The uninsured patients in our study reported going without medical care at a very high rate (82%). When they do go to the doctor, most do not take prescribed medications because of the cost.
A large majority of uninsured patients in our study work long hours. Few have the opportunity to purchase insurance though their employer.
Lack of health insurance crosses geographic and cultural lines, encompassing people who live in cities and suburbs, who are immigrants, and who are native-born Americans. In the clinics in our study, patients without insurance are more likely than those with insurance to be young, female, and born in the United States.
Undoubtedly, studying a different population would have produced somewhat different results. Surveying patients in the culturally diverse neighborhoods where the 2 clinics are located allowed us to find enough insured and uninsured patients to make statistically valid comparisons. At the same time, certain factors may limit our ability to draw general conclusions. For instance, many of the immigrants surveyed were from Somalia and have high rates of employment and good English skills. We also encountered a surprising number of suburban patients drawn to an inner-city free-care clinic. Also, we only surveyed patients who actually came to the clinic, as opposed to those who may have had reason to see a doctor but did not seek help. Those individuals may have been more likely to be uninsured, have poor English skills, be older, etc.
Despite those limitations, we believe our results contradict some prevalent thinking: that most uninsured people are unemployed, that many refuse employer-sponsored insurance, and that they can easily get care through free clinics. Our findings also cast doubt on the notion that uninsured people are mainly immigrants, members of minority ethnic groups, or inner-city residents. The truth is, not having health insurance affects all of us.
Most important, our findings document that many Minnesotans are going without health care and provide fresh data that may be useful for those considering what to do about this growing problem. MM
The authors thank Terry Rosborough, M.D., for statistical analysis and Kelsey Gilmet for analysis of patient employment.
Richard Adair is a clinical professor of medicine at the University of Minnesota. Amy Greminger and Bryan Post are residents in internal medicine. All work at Abbott Northwestern Hospital in Minneapolis.
REFERENCES
1. Minnesota Department of Health. Characteristics and trends among Minnesota’s uninsured population. Available at: www.health.state.mn.us/ divs/hpsc/hep/publications/coverage/uninsure.pdf. Accessed Feb. 22, 2006.
2. Physicians’ plan for a healthy Minnesota. The MMA proposal for health care reform. The report of the Minnesota Medical Association Health Care Reform Task Force. Minn Med. 2005;88(3 Suppl):1-41.
3. Blendon RJ, Young JT, DesRoches CM. The uninsured, the working uninsured, and the public. Health Aff. 1999;18(6):203-11.