Abstract
PURPOSE Public health insurance programs have expanded coverage for the poor, and family physicians provide essential services to these vulnerable populations. Despite these efforts, many Americans do not have access to basic medical care. This study was designed to identify barriers faced by low-income parents when accessing health care for their children and how insurance status affects their reporting of these barriers.
METHODS A mixed methods analysis was undertaken using 722 responses to an open-ended question on a health care access survey instrument that asked low-income Oregon families, “Is there anything else you would like to tell us?” Themes were identified using immersion/crystallization techniques. Pertinent demographic attributes were used to conduct matrix coded queries.
RESULTS Families reported 3 major barriers: lack of insurance coverage, poor access to services, and unaffordable costs. Disproportionate reporting of these themes was most notable based on insurance status. A higher percentage of uninsured parents (87%) reported experiencing difficulties obtaining insurance coverage compared with 40% of those with insurance. Few of the uninsured expressed concerns about access to services or health care costs (19%). Access concerns were the most common among publicly insured families, and costs were more often mentioned by families with private insurance. Families made a clear distinction between insurance and access, and having one or both elements did not assure care. Our analyses uncovered a 3-part typology of barriers to health care for low-income families.
CONCLUSIONS Barriers to health care can be insurmountable for low-income families, even those with insurance coverage. Patients who do not seek care in a family medicine clinic are not necessarily getting their care elsewhere.
INTRODUCTION
Children from lower socioeconomic backgrounds have poorer health outcomes.1,2 These health disparities are due, in part, to barriers in accessing medical care and utilizing primary care services.3–5 Recent expansions in insurance coverage have improved access to health care for this population,3,6–8 and the presence of family physicians in underserved communities has made primary care services more widely available.9 Yet, even with the services of family physicians and expanded health insurance, children from low-income families are not guaranteed access to health care services.5,10–15 Among the poor, who visits a family physician and who does not? What barriers persist? Family physicians caring for vulnerable populations must understand differences in access to care and utilization of services in their communities.
Recent efforts to better understand these disparities have queried physicians and patients about differing practices based on a patient’s health insurance status and type of insurance.12–14,16 Other analyses have used vital statistics and Medicaid data to study utilization and coverage patterns.6,17 Secondary analyses of administrative data are often several steps removed from the real-life experiences of vulnerable families, and studies conducted in physicians’ offices miss the invisible families who are unable to visit health care facilities. To our knowledge, limited information has been collected directly from families living in poverty about factors affecting access to medical care and how insurance status affects their situations. This study was designed to add richness and depth to the current research by directly capturing the experiences of low-income families as they navigate the health care system and to determine how insurance coverage affects their concerns.
In collaboration with state policy makers, we designed a cross-sectional survey to collect statewide primary data. More than 25% of the survey respondents provided additional written comments to a concluding open-ended survey question that asked, “Is there anything else you would like to tell us?” We report findings from a mixed methods analysis of this subset of respondents.
METHODS
Study Population and Data Collection
The study population included all Oregon families enrolled in the federal food stamp program at the end of January 2005 with children who were also presumed eligible for publicly funded health insurance. Both programs require a household income of less than 185% of the federal poverty level and proof of the child’s US citizenship. A random sample of families was drawn with purposeful oversampling to ensure adequate representation from rural areas and uninsured families. More details about this sampling procedure are reported elsewhere.18
The return-mail survey questionnaire was developed to allow parents the opportunity to report about various health-related issues for 1 randomly selected focal child. Survey questions were grouped into 4 major sections: child’s health insurance status, child’s access to various health care services, child’s demographic information, and family information (primarily demographics and parental insurance information). The questionnaire contained 62 questions with multiple-choice response options, and 1 final open-ended question that asked, “Is there anything else you would like to tell us?” These responses were used for our mixed methods analysis.
For validity testing of the self-administered questionnaire, cognitive interviews were conducted during a pilot test phase with a small sample of low-income parents that were representative of the study population. Surveys were translated into Spanish and Russian (the 2 most common non-English languages among this population), and then independently back-translated to ensure fidelity of translation. The questionnaire was written at a fifth-grade reading level. All aspects of the study protocol were approved by the Oregon Health and Science University Institutional Review Board (OHSU eIRB# 1717).
Analysis
We confirmed that our subsample of respondents to the open-ended question had demographic characteristics that were similar to those of all survey respondents and to the total eligible survey population. We also conducted bivariate and multivariate analyses to confirm that the experiences of our subsample in accessing care were similar to previous reports about the strong association between insurance status and different utilization of primary care services.3,5,8 Covariates were selected based on the conceptual model of Aday and Andersen and the work of others on predictors of access to care.8,19,20 These analyses were conducted using SPSS 14.0 software (SPSS Inc, Chicago, Illinois) with the complex samples module to account for the complex survey design and to ensure proper weighting back to the Oregon food stamp population.
After the preliminary quantitative review, our qualitative analysis team began the process of identifying major themes. The team included a family physician researcher (J.E.D.) and a health services investigator skilled in quantitative and qualitative research (P.A.C.). For further diversity in analysis, we included a medical student jointly enrolled in a public health master’s program (A.B.) and a research associate from our rural practice-based research network (H.A.). Feedback throughout the process was also received from 2 authors with expertise in evidence-based state policy development and implementation (L.K. and C.E.).
Each team member independently read all written comments and grouped them into categories. We then met to discuss the items and agree upon a common cataloguing of themes. Once consensus was reached, we organized the categories into a codebook of tree nodes using NVivo qualitative software 7.0.21 Tree nodes are used to represent categories that are catalogued in a hierarchical structure, moving from a general category at the top (parent nodes) to more specific categories (child nodes). We repeated our individual reviews with codebook guidance and met regularly to conduct a series of immersion/crystallization cycles.22 During these meetings, specific categories were grouped into more general categories, and the codebook of tree nodes was revised to reflect the multiple interpretations of all team members. The use of NVivo also facilitated line-by-line coding of each text entry for further review. After multiple reviews of all 722 responses to the open-ended survey question, we had reached saturation on 3 dominant themes, each with several subthemes.
For further in-depth analysis to determine whether experiences with each of these themes varied among subgroups, we imported several pertinent demographic attributes from the SPSS data set into NVivo and conducted univariate matrix-coded queries. The quantitative variables used in the matrix coding queries were chosen for 1 of 2 reasons: (1) relevance to the themes (measures of insurance status, access to and utilization of health care services); and (2) demographic predictors of access to care (age, ethnicity, and household income).
RESULTS
Demographics and Different Patterns of Utilization
Completed questionnaires were received from 2,681 of 8,636 (31%) eligible households. Among the survey respondents, a sub-sample of 722 wrote responses to the open-ended question, “Is there anything else you would like to tell us?” Although this subsample was demographically similar to the overall study population, it had a slightly higher percentage of uninsured children, uninsured parents, and children with a gap in coverage, when compared with all survey respondents (Table 1⇓).
Comparison of Respondent Characteristics to Overall Sample Population
Among this group, after adjusting for age, race/ethnicity, parental insurance status, region of residence, and household income, children with health insurance (private or public) were more likely to have a usual source of care and to have seen a doctor in the past year (Table 2⇓). These findings confirm that the low-income families in our subsample had utilization patterns based on insurance status similar to those previously reported.23 The associations between parental insurance status and children’s utilization of primary care services were not significant after adjusting for children’s insurance status.
Insurance Status and Child’s Utilization of Primary Care Services
Dominant Themes and Different Patterns of Experience
In the qualitative analysis, 3 major themes emerged as factors affecting access to and utilization of health care services for children. First and most commonly reported, parents were concerned about getting and keeping health insurance coverage for themselves and their children. Parents repeatedly stated that they need insurance coverage not only for their children, but also for themselves so that they can be healthy enough to care for their children. They also expressed frustrations about not being able to meet all the restrictive criteria for continuous Medicaid enrollment. Second, gaining access to services and finding providers was a big challenge. Parents described feeling unwelcome at medical practices and traveling long distances to seek care. Finally, unmet health care needs were attributed to the high costs of medical care (Table 3⇓). These responses largely focused on unaffordable private insurance premiums and a hesitancy to seek care because of the high deductibles and co-payments.
Themes and Subthemes of Barriers to Care Among Low-Income Families
Among several demographic characteristics and other family circumstances, insurance status was the factor associated with the most disparate reporting of the 3 themes (Table 4⇓). More than 87% of uninsured parents commented on difficulties obtaining insurance coverage compared with approximately 40% of those who were insured. Only 14% of parents with uninsured children wrote about access concerns in their responses to the open-ended question, whereas more than 25% with publicly insured children and 20% with privately insured children were concerned about access to health care services. More than 30% of privately insured parents and those whose children had private coverage mentioned costs compared with less than 20% in the other insurance groups (Table 4⇓). In summary, obtaining and maintaining insurance was the most important theme among all families. Comparing families in all insurance groups, insurance coverage issues were more often reported by families with uninsured parents or uninsured children. Access concerns were mentioned most often among those with public health insurance, whereas privately insured families more commonly mentioned unaffordable medical costs.
Frequency of Families Reporting Each Major Theme by Demographic Characteristics and Other Pertinent Factors
The disproportionate reporting of themes based on insurance status illustrates different patterns of experience among this population of low-income families. Parents made clear distinctions between insurance and access, and there appeared to be a hierarchical order for obtaining both. Insurance coverage was the primary concern; access and costs were secondary. Families without insurance were most focused on obtaining insurance and tended not to write as much about access or cost. Families with insurance were worried about whether they could use the insurance. These access concerns had 2 major subthemes: clinician acceptance of insurance and insurance coverage of services at a level that makes them affordable. Cost was less a concern to this group because without access, the service was unobtainable. For example, as shown in Table 4⇑, a smaller percentage of parents with children who had not visited the doctor in the past year reported cost concerns. For those with access, cost played an important role. Although there is a hierarchical order to the themes, all 3 themes—insurance, access, and cost—are interrelated. Families can achieve both emotional and financial security when all 3 are balanced. If insurance coverage is unstable, access and cost are also jeopardized. Alternatively, once insurance is solidly in place, it still takes continued effort to achieve optimal access at an affordable cost.
DISCUSSION
These study findings are in agreement with previous findings about the importance of stable insurance coverage for the entire family in gaining access to care.24–34 Our study advances this literature by creating a more in-depth understanding of the hierarchy of barriers faced by low-income families and the interactions that exist between insurance, access, and cost. For parents surmounting the insurance barrier was most important, but then access became a bigger issue. Families that got over the insurance and access humps were still struggling to get care as a result of unaffordable costs, such as co-pays for office visits, insurance deductibles, and prescription drug costs. This hierarchical model emphasizes that there is no single struggle in obtaining health care—there are 3 successive barriers to care.
Interestingly, we found a wide variation in concerns among parents depending on insurance status. Because fewer uninsured families were accessing primary care, they were not experiencing the barriers to access and the costs associated with care. For those who had secured public health insurance for the entire family, the major challenge became access. Perhaps costs were not mentioned as a major problem by these respondents, because if the insurance plan did not have provisions for access or if providers did not accept the coverage, there were no options for obtaining care. In this case, cost was not a barrier because the children went without the services, which might explain why a smaller percentage of parents with children who had not visited a doctor in the past year reported cost concerns. For those with private insurance, access was better but at higher, and often unaffordable, costs. The motto for these families might be: “We have insurance and we have a family physician, but we cannot afford to get health care.”
In the examination of the interrelationships between insurance, access, and cost, our in-depth analyses uncovered a 3-part typology of health care access barriers. First, many families without insurance are invisible to health care providers. These families no longer call the clinic for fear of hearing the opening question: “What is your insurance?”23 They feel intimidated and helpless because their infrequent interactions with the health care system have resulted in denied care or unaffordable medical charges. Occasionally these families make an appointment for someone in the family who has insurance, then seek care for the uninsured family members during the single visit. The second group, often with public coverage, spends much of its time searching for clinicians and facilities that accept public insurance. Family physicians are more likely to care for these patients compared with other primary care physicians,9 but these are the patients who cannot get into the specialists’ offices.12–14 Finally, there is a growing number of low- and middle-income families with private health insurance who gain access to most services, but the high deductibles and co-pays prevent them from getting necessary care. These are the patients who prefer to call clinicians for advice rather than be seen, and they often have difficulty filling expensive prescriptions and rely more heavily on pharmaceutical samples.
The situation is far more complex than assuming that expanding health insurance or increasing the primary care workforce will solve the problem. Whereas many primary care providers are already aware of the complexity of these issues,16 our study was designed to explore the issues from the patients’ perspectives. Understanding this hierarchy and these families’ varied experiences is essential for the redesign of primary care practices and training of future family physicians. The typology of access barriers displayed in Table 5⇓ may help further our understanding of these important issues. In all 3 scenarios, patients cannot get necessary care but for different reasons. Most alarming is that health care services are not obtainable for low-income families, even those with insurance coverage and access to primary care.
Barriers to Care Typology
Study Considerations
Interpreting the data reported here requires consideration of some important issues. Several factors about the sample population and fielding of the survey are described in detail elsewhere.18 Responses to the open-ended question may have been biased by the content of the survey questionnaire, which included specific queries about health insurance coverage for both children and parents, access to health care, and costs of health care. These questions did, however, provide several opportunities for respondents to report barriers and concerns about access to health care. It is telling that after completing the questionnaire, many parents were motivated to comment further. Finally, it is possible that we received responses to the open-ended question from only those families who encountered the most difficulties with the system, so the results may not be generalizable to all families. Our sub-sample, however, did have demographic characteristics similar to those of the original population.
Policy Implications
It is essential for policy makers to understand the barriers faced by low-income families when trying to access necessary medical care. Insurance does not guarantee access, and having access to primary care does not guarantee receipt of all necessary care. Clinicians must understand that many patients who are not coming to see them are not necessarily going somewhere else, such as a safety net clinic. Additionally, family physicians who are familiar with this complex situation can help patients navigate the system and can be advocates for vulnerable patients in policy discussions. As evidenced here, health insurance is an essential foundation for all families, but it does not solve all problems. Policy reforms need to address all 3 issues: insurance, access, and cost.
Footnotes
-
Conflicts of interest: none reported
-
Funding support: The study was funded by the Health Resources and Services Administration (HRSA). Jen DeVoe’s time on this project was supported by grants 5 F32 HS014645 and 1 K08 HS16181 from the Agency for Healthcare Research and Quality (AHRQ). Alia Baez’s time on this project was supported by a medical student research grant from the Oregon Academy of Family Physicians. Patricia Carney’s time was supported by the Oregon Health and Science University Department of Family Medicine Research Program.
- Received for publication December 18, 2006.
- Revision received July 29, 2007.
- Accepted for publication August 7, 2007.
- © 2007 Annals of Family Medicine, Inc.
REFERENCES
Jump to section
Related Articles
Cited By...
- Frequency and source of prescription eyewear insurance coverage in Ontario: a repeated population-based cross-sectional study using survey data
- Primary Care Access to New Patient Appointments for California Medicaid Enrollees: A Simulated Patient Study
- Multimorbidity and out-of-pocket expenditure on medicines: a systematic review
- Despite Boosting Childrens Coverage Rates To Historic Levels, Medicaid And CHIP Face An Uncertain Future
- Community of Solution for the U.S. Health Care System: Lessons from the U.S. Educational System
- Electronic Health Records vs Medicaid Claims: Completeness of Diabetes Preventive Care Data in Community Health Centers
- Randomized Controlled Trial of Mammography Intervention in Insured Very Low-Income Women
- Children's Receipt of Health Care Services and Family Health Insurance Patterns
- Socioeconomic and Racial Patterns of Colorectal Cancer Screening among Medicare Enrollees in 2000 to 2005
- The Problem of Fragmentation and the Need for Integrative Solutions
- In This Issue: Equity Global Theme Issue on Poverty and Human Development