Appendix 5. Reports on Needs Assessment Activities (continued)

Access Research

"Access" refers to the "timely use of health services to the best possible outcomes." (IOM, 1993).

Many factors can limit access. Past research has told us a great deal about financial barriers facing individuals and families insurance and income). From the Medical Expenditure Panel Survey (MEPS), we know that two of the primary barriers to receiving needed health care reported by American families are the inability to afford needed care and factors associated with health insurance, together accounting for nearly 80 percent of reported difficulties (Weinick, Zuvekas, and Drilea, 1997).

In a health care system dominated by employer-sponsored insurance, individual's access to such coverage, as well as factors associated with choice of coverage, premium costs to individuals, and benefit structure all have a considerable impact on financial access to care. Access to private, employer-sponsored health insurance has number of dimensions: whether employers offer plans to their employees, the choice of plans employees face and their premium cost to employees, the structure of benefits, and the process by which households and families make decisions concerning health insurance coverage. Research at AHRQ has answered a number of the policy-related questions and contributed to our collective knowledge about employer-based coverage.

However, we know much less about providers and health plans' financial incentives and disincentives to assure access, and much less about the critical non-financial barriers facing individuals, families, providers and plans.

In recent years leaders from within the field of health service research have stepped forward in acknowledging deficiencies in access research, and in particular how to measure access and gaps in our knowledge of barriers. In a 1998 issue of Health Affairs, Marc Berk and Claudia L. Schur lament that "despite more than sixty years of conducting major national health surveys, we have not reached consensus about the number of Americans who do not receive adequate health care or whether access is getting better or worse. Even the most thoughtful persons find it difficult to disentangle the research on access. Policymakers recognize that millions of Americans are unable to obtain adequate health care and that informed debate is necessary if viable solutions are to be found. But carrying on this debate is difficult when health services researchers cannot agree about key aspects of the problem.

Access is an important area for the Agency, targeted first among the triad of strategic goals related to 'access, cost and use' for comprehensive agenda development. Beginning in 1997, the Agency has undertaken a variety of needs assessment activities to build an access agenda and has begun exploring potential partnerships that might play a role in executing the agenda. In January 1997, the Agency convened a meeting of stakeholders to explore and discuss, among other things, emerging access issues associated with managed care.

In November 1998, the Agency sought input from its National Advisory Council (NAC) on the Agency's role in developing the scientific basis for policy decisions affecting access. A consensus has emerged that the policy issues affecting access have changed in the past decade—e.g., as the predominant form of insurance shifted from fee-for-service to managed care, and as Americans have become more culturally diverse—and that they vary from one market area to another. Council members supported Agency involvement in access and posed particular policy-relevant research for consideration on the access agenda (Minutes, November 1998 NAC meeting). Needs assessment related to access is ongoing and will continue into 2000 as the Agency prepares to execute its omnibus access agenda in 2001.

The Agency's access agenda-in-progress is organized into the following five domains:

1. Insurance coverage and other financial issues facing families.
2. Access for persons at high risk for utilization enrolled in managed care.
3. Access in rural areas.
4. Access among minorities.
5. Access among children.

For each domain, meetings with users—stakeholder meetings—important in helping AHRQ build its access agenda are identified, and key research findings informing future research-related needs are summarized. Draft Agency priorities, gleaned from these two sources, are then listed; these draft priorities for future work in each of the domains span research, data (including enhancements to MEPS, HCUP and CAHPS®), tools , measures, and extramural training needs.

Domain I: Insurance coverage and other financial issues facing families

1997-1999 stakeholder meetings: Association for Health Services Research meeting, June 1999. Presentations and discussion with researchers and policy-makers on: Border Crossings—Piercing the Boundary Between Private and Public Coverage; and on Assessing Health Plan Choice in the United States. American Enterprise Institute meeting, November 1999. Presentations and discussion with researchers and policy-makers on Employer Contributions and Health Insurance Premiums—Does Managed Competition Work?

Findings from intramural and/or extramural research related to future research needs: Although the primary source of private coverage in the United States remains the employment-based health insurance system, such coverage is not accessible to all American families. Research is required to assess both the equity and efficiency of the current employment-based system and to determine whether lack of coverage reflects issues of affordability, other non-price impediments to coverage, or both factors.

Current research needs include studies to determine how out-of-pocket premium costs, incentives associated with the tax subsidy for employer-based coverage, and employer premium contributions affect decisions to participate in health plans as well as the kinds of plans selected.

Research is also required to determine the nature of employee preferences for coverage, specifically, the kinds of coverage that are valued and the willingness of employees in different economic circumstances to trade wage income for health insurance benefits. Such decisionmaking becomes increasingly complex among households with two working spouses, when employers provide several plans to choose from, and when the benefits and premium contribution schemes differ among offered health plans.

Research is also required to assess the willingness of employers to offer health benefits, to describe the kinds of health plans made available, and to examine whether the provision of plans with differing benefits, out-of-pocket costs, and contribution schemes is associated with adverse selection by persons with particular kinds of health problems. At the same time, it will be important to assess whether the provision of health plan choice leads to greater consumer satisfaction with their coverage and to examine the implications of expanded plan choice and the types of plans selected on the use of health care services and quality of care received.

Given policy interest in the effectiveness of managed competition, research is required to assess whether expanded health plan choice and existing premium contribution arrangements effectively promote competition among health plans and contribute to lower health plan costs. Finally, research is required to examine the extent to which current health plans do protect consumers from the risk of excessive health care expenditures and whether efforts to improve employee access to coverage in the small group insurance market have been successful. (From research by Monheit and Vistnes, 1999; Schone and Cooper, 1999; Vistnes, Cooper, and Vistnes, 1999; Zuvekas, Banthin, and Selden, 1998; Monheit, Schone, and Taylor, 1999; Cooper and Schone, 1997; and Monheit and Schone, 1999.)

There is considerable research interest in assessing whether efforts to expand public insurance coverage to specific population groups have been successful in reducing uninsured rates, improving access to and use of health services, and in improving health outcomes. Research is required to assess whether specific kinds of outreach programs associated with expanded coverage have been successful in raising enrollment rates and the extent to which expansion of public coverage to groups with incomes above the poverty level will result in a crowding out of other sources of insurance coverage. Moreover, given efforts to enroll the Medicaid and Medicare populations in managed care plans, research is needed to assess the stability of enrollment in such plans, whether such plans engage in favorable enrollee selection, the impact of such plans on the costs of care relative to traditional Medicare and Medicaid plans, and whether such plans yield disparities in treatment and health outcomes compared to enrollees in public programs reimbursed via fee-for-service. Finally, there is continuing interest in assessing the role that supplemental Medigap coverage plays on the expenditures of this population and whether enrollment in such plans is governed by adverse selection. (From research by Selden, Banthin, and Cohen, 1998; Selden, Banthin, and Cohen, 1999; and Vistnes and Banthin, 1997.)

Recommendations for future research, data, measures, tools, and training:

• Research on the role that out-of-pocket health insurance costs have on the access to coverage of low-income workers.
• Research on how health insurance benefit structure affects access to care for both mental and physical health coverage.
• Research on processes by which households make their health insurance coverage decisions for both private and public plans, including the choices they face, the information available to them as they make their decisions, and how they evaluate their options.
• Research on the role that health insurance coverage plays in obtaining needed health care services, including the financial and non-financial (e.g., gatekeepers) effects that plans may have.
• Evaluation of outreach efforts for expanding public health insurance enrollment for children.

Domain II: Access for groups at increased risk for underservice in managed care plans

1997-1999 stakeholder meetings:

• Expert Meeting on Markets and Managed Care, January 1997.
• Medpac session on access for Medicare beneficiaries, December 1999
• In addition, AHRQ will be sponsoring a February 2000 stakeholders' meeting on HIV research.

Findings from intramural and/or extramural research related to future research needs. Nicole Lurie (HSR, December 1997) reviewed the study of access in managed care environments:

"Concerns about the profit and cost-cutting motives of managed care organizations abound, and are combined with increasingly frequent anecdotes and projections of decreased access to care. The debate is emotional and highly visible, yet the data are scant. Prior research on access to care can only partially illuminate the issues at hand. [Research is needed] to sort out which system characteristics, financial characteristics, for profit status, and so forth-seem to matter, not only for maintaining access, but for expanding it."

Lurie identified a need for research on managed care issues related to special populations, such as low-income and Medicaid enrollees, individuals needing mental health care, children with special health care needs.

In a companion HSR article, Edward Wagner focus on another special population—the chronically ill—and asserts may have the most to gain or lose by managed care.

"...We are experiencing unprecedented, unevaluated tinkering with basic care models...Patients with major chronic illnesses and disabilities are most at risk if this tinkering disrupts critical health care relationships or reduces access to needed services."

Recommendations for future research, data, measures, tools and training:

• Expansion of access measures for managed care enrollees, especially those special populations at risk of underservice.
• Research on access for special populations at increased risk for underservice, such as the chronically ill, the disabled, the elderly and those with costly conditions (e.g., HIV infection, mental illness, substance abuse).
• HIV database (DCC).
• Adding the coding of 'Medicaid HMO' as a criteria in selecting additional State HCUP partners.

Domain III: Access in rural areas

1997-1999 stakeholder meetings:

• Quarterly meetings throughout 1998-99 of the Capital Area Rural Health Roundtable.
• National Rural Health Association Annual Conference in May, 1999.
• Director's meeting of Rural Health Research Centers (sponsored by HRSA) in October, 1999.
• State Rural Health Association meeting, October 1999.

In addition, AHRQ will be co-sponsoring with the Office of Rural Health Policy, HRSA, a national invitational conference in Washington, D.C., in January 2000. The purpose is to develop a research agenda on rural health for the next ten years. This agenda will include a focus on access issues.

Findings from intramural and/or extramural research related to future research needs:

Traditional indicators of access to care—proportion of population with a usual source of care, and proportion of population with a hospital admission—appear to be less useful today in the study of rural areas as a result of significant losses of medical providers in these areas (making a local source of care less likely even for the well insured) and the rapid closing of a number of rural hospitals (Hayward, 1991; Ricketts, 1999). Much research examining access issues in rural areas has therefore focused instead on differences in health status and utilization of available health services in urban vs rural populations. MEPS data indicate that rural Americans report more chronic conditions and describe themselves as being in poorer health than their urban counterparts (Weinick, Zuvekas, Drilea, 1997). In addition, studies of the effect of state policies on childhood immunization rates indicate that heavy reliance on public sector programs does not alone ensure timely receipt of vaccines by rural residents. Buffering the effect of poverty on receipt of immunization by children living in rural areas appears to require a collaboration between the public and private sectors (Mayer et al, 1999).

Evaluations of the effect of local managed care penetration on access in rural areas have yielded less than encouraging results. A ten state study of Medicaid managed care in rural areas (supported by AHRQ) has questioned the benefits that these programs produce for rural beneficiaries and has suggested that the effort required to implement the program in rural areas may have actually increased the overall costs of care without truly having an impact on access (Felt-Lisk et al, 1999). In addition, a number of states have developed demonstration programs on methods of establishing managed care programs in rural areas. The overall effect of these programs on access to care remains unclear.

Published research has yet to address adequately the following issues: improving transportation access to health care services for rural community members; organizing emergency medical services to ensure regional access; understanding the consequences of telemedicine technology adoption and diffusion in rural regions; coordinating rural case management services across health and social services providers; improving the availability of mental health and substance abuse treatment services in rural areas; and improving the ability of rural communities to recruit and retain health care professionals.

Recommendations for future research, data, measures, tools and training:

• Expansion of access measures to address issues specific to rural areas, including transportation access and penetration of managed care.
• Development of a rural HCUP hospital database.
• Development of a rural HCUP emergency department database.
• Development of capacity of primary care practice-based research networks (PBRNs) to conduct research on underserved populations living in rural areas (1999 RFA on PBRNs).
• Research on recruitment and retention of health care professionals for rural communities (one R03—Pathman—already funded); and outcomes and costs associated with adoption of telemedicine technology by rural providers and patients.
• Research on market impacts on rural populations (1999 markets RFA).

Domain IV: Minority population's access

1997-1999 stakeholder meetings:

• Kaiser Family Foundation's conference on racial and ethnic disparities, October 1999 (co-funded by AHCPR).
• HCUP partners meeting, May 1999
• AHRQ expert meeting on Future Directions for Health Services Research Regarding Racial and Ethnic Minority Populations, May 13, 1999.

Findings from intramural and/or extramural research related to future research needs:

Racial and ethnic health disparities have been documented at every point in the health care system—in the incidence of disease, in access to and utilization and quality of health services, and in health outcomes. With minority Americans expected to comprise over 40 percent of the U.S. population by 2035, and 47 percent by 2050 (U.S. Bureau of the Census, 1996), addressing these disparities has become an increasingly visible public policy goal (From U.S. Department of Health and Human Services [HHS], 1998; Agency for Health Care Policy and Research [AHCPR], 1999; U.S. HHS, 1999.)

Some argue that elimination of financial differences in access would profoundly reduce health disparities (Andrulis, 1998; Schur, Albers and Berk, 1995). Clearly there are disparities in health insurance coverage. MEPS has found that Hispanics and blacks were much more likely than whites to be uninsured (Kass, Weinick, and Monheit, 1999). Nevertheless, disparities have been documented within health systems that provide equal financial benefits to all covered individuals—such as the Veterans Health Administration, Medicare, or single health plans (Oddone et al., 1999; Whittle et al., 1993; Goldberg et al., 1992; Robbins, Whittemore, and Van Den Eeden, 1998; Carlisle, Leake, and Shapiro, 1997; Peterson et al., 1994; Ayanian et al., 1993). This indicates that linguistic and cultural barriers as well as financial ones make it hard for individuals from racial and ethnic minority groups to obtain appropriate health care.

Members of different racial and ethnic groups vary in language, cultural beliefs, health behaviors, and health preferences, and therefore the delivery of health care must be altered if it is to meet their differing needs. While acknowledging that everyone would benefit from health care systems adopting more patient-centered approaches to delivering care (Delbanco, 1992; Silberman, 1992), particular attention should be paid to minority populations because that is where the gap between the prevailing health system and the needs of patients is greatest.

While there is a large literature on racial and ethnic disparities in health care (Mayberry et al., 1999), it has focused on rigorously documenting the disparities, an important contribution of outcomes and effectiveness research and health services research. Nevertheless, these insights have infrequently led to significant improvements in racial and ethnic disparities, in part, because the causes of and contributing factors to these inequalities are inadequately understood

Cultural competency (the ability of health care systems and their clinicians to deliver to diverse populations appropriate services that lead to good outcomes) has the potential to help improve access by minority populations and decrease disparities. However, the research on cultural competency has not linked cultural competency with outcomes that could be expected to follow from cultural competency activities. The application of cultural competency to organized health systems such as managed care organizations has not received needed scholarly attention. Lack of cultural competency measures hinder both the testing of cultural competency's theoretical premises and health systems' ability to monitor and improve their own performance. (Excerpts from Cultural Competency: A Tool for Health Systems to Reduce Racial and Ethnic Health Disparities, submitted for publication December 1999, by Brach and Fraser, and the AHRQ RFA "Understanding And Eliminating Minority Health Disparities.")

Recommendations for future research, data, measures, tools and training:

• Minority HCUP database.
• Development of HCUP emergency department data.
• Research on market impacts on minority populations (1999 markets RFA).
• Intramural chartbook analyzing hospitalizations among minorities.
• Working with NCHS to have race/ethnicity added to the HIPAA standard for the institutional claim.
• In HCUP data, the need to preserve rich, state-specific data on race/ethnicity (i.e., in certain states) while at the same time adding a recoded variable that conforms to HIPAA standards.
• A research agenda that balances short-term and long-term needs, and balances descriptive research with evaluative approaches, and includes: cultural competence, institutional racism, and the patient/provider relationship; understanding the causes of health disparities and identifying strategies to eliminate them (1999 disparities RFA); the influence of managed care on minority health; rapidly changing nature of the health care marketplace and the means by which populations and their needs are identified; community and population characteristics must be examined in order to identify the needs of specific populations; effect of violence and substance abuse on care provided to minority populations.
• Identify innovative strategies for involving communities of color in research.
• Increase the participation in health services research of minority researchers, minority institutions, and institutions that serve minority populations and increase capacity to study minority health issues among non-minority researchers.

Domain V: Children's access to insurance and to care.

1997-1999 stakeholder meetings:

• AHSR's Improving Quality of Health Care for Children: An Agenda for Research, May 1997, which was co-funded by AHRQ.
• MEPS and Child Health Analyses Expert Meeting, February 1997.
• Learning from CHIP I, March 1998.
• November 1998 NAC meeting discussion.
• Learning from CHIP II, June 98.
• AAP meeting, May 98 [see cite below].
• ULP dialogues with states, June 1998, August 1998.
• Coordinating committee for CHIP RFA, November 1999.
• HCUP state partners' meeting, May 1999.
• Expert Meeting on Quality Improvement for Children and Adolescents, April 1999.

Findings from intramural and/or extramural research related to future research needs:

Recent policy changes have attempted to increase children's health insurance coverage and access to care so that children can obtain health care that is appropriate to their developmental needs. In the last decade, the Medicaid program has been expanded and the State Children's Health Insurance Program has been initiated to decrease the proportion of children who are uninsured. However, MEPS data suggest that: 1) in spite of these initiatives, a substantial number of children in American remain uninsured; 2) Black and Hispanic children are at increased risk of adverse outcomes, including being uninsured and being in poor health; 3) having a working parent is not enough to ensure children's access to private health insurance; 4) the most common reason children do not get needed health care is because their families cannot afford it; and 5) public coverage is a critical factor in providing insurance for children with health problems. (Excerpts from Children's Health, 1996: MEPS Chartbook No. 1, by Weigers, Weinick, and Cohen, 1998)

The consequences of lacking insurance have been fairly well documented: children and adolescents are less likely to see a physician, be immunized, receive appropriate well-child care, receive timely treatment for acute health problems, and are more likely to have unmet health care needs (Weigers, Weinick, and Cohen, 1998; Newacheck et al., 1999). Yet the reasons why many children eligible for public insurance have not enrolled are poorly understood, and we are only beginning to understand other problems experienced by certain subpopulations of children in accessing health services. For example, researchers, including those supported by AHRQ, have documented how other factors, including legal, organizational, and interpersonal factors, present barriers to access and use for adolescents.

Studies of adult access to care cannot serve as a proxy for several reasons. First, adults and children have different epidemiology of health and health care (e.g., more need for disease prevention and health promotion services among children and adolescents; issues of transitions between child- and adult-oriented health care providers and settings for adolescents with chronic illnesses and disabilities). Second, children are dependent on adults, particularly parents, for access. Third, children are demographically different (high numbers of children and adolescents in poor families). Fourth, children's high rate of developmental change necessitates different levels and types of access. And fifth, children and adolescents are involved in a wide range of systems beyond mainstream health (e.g., school-based services, mental health and substance abuse clinics, reproductive health clinics).

Currently there is little understanding how these services do or do not interrelate with the mainstream health system to promote optimal health care. When asked about priority areas for children's health services research in November 1998, AHRQ's National Advisory Council advised that AHRQ expand beyond its traditional focus on mainstream health care delivery systems. Other stakeholders, responding to invitations to comment on AHRQ's overall Strategic Plan, recommended that AHRQ: 1) evaluate changes in the organization, delivery, and financing of children's health care services; 2) focus on access related to children's hospitals and hospital-based services; and 3) contribute to HHS strategic plan objectives to improve access to care for persons with specific needs, and to increase the availability of primary health care services.

Managed care's impact must also factor into any inquiry regarding children's access to health services. Children appear to be somewhat more likely to be covered by managed care plans than adults, with 22.2 percent of all children under age 20 versus 19.6 percent of adults (Leatherman, 1997). Managed care has many features that can promote excellent pediatric care and improved health status for children, e.g., it can focus on health promotion, a critical set of services for children to develop normally and achieve their full potential. On the other hand, as many have noted (Newachek et al., 1996; Newachek et al., 1994; Hughes et al., 1995; Fox et al., 1997a; Fox et al., 1997b), capitation and other cost controls have the potential to limit quality and promote under-service, and some of the strengths of managed care are less likely to apply to children. While theoretical speculation about the potential impact of managed care on children is abundant, efforts to test these theories empirically have not yielded any simple answers. Most of the research has focused on adults, and has contrasted fee-for-service with all forms of managed care as a whole. Five recent reviews of this literature conclude that the impact of managed care varies tremendously depending on the model of care, the population studied, and the methodology (Miller and Luft, 1997; Hellinger 1998; UHC Peterson; Reid et al., 1996; Cangialese et al., 1997).

The little we do know about the impact of managed care on adults is not easy to translate to children. The adult literature does, however, provide some hypotheses about where one might look in identifying issues for children. On the whole, the adult literature seems to show, though certainly not conclusively, that managed care is at least as good as fee-for-service for most of the people most of the time, but that sicker and poorer or otherwise vulnerable people may sometimes be the exception. And of course children are poorer, and have different epidemiologies and health needs.

Three recent papers have reviewed what is known about the particular impact of managed care on children (Leatherman and McCarthy, 1997; Szilagyi and Bergman, 1998; Bergman and Homer, 1998). One clear fact emerges: there is no simple answer to the question of whether managed care works for children. It all depends on what type of managed care is the focus of the study, what dimensions of care for children are examined, which population of children is included in the study, the study methods themselves, and when the research was conducted. (Excerpt from Children and Managed Care: What Research Can, Can't and Should Tell Us about Impact, MCRR Volume 56, by Simpson and Fraser.)

Recommendations for future research, data, measures, tools and training:

• Research on impact of specific features of managed care on children.
• Pediatric HCUP database.
• Research on why many children eligible for Medicaid and SCHIP do not become enrolled.
• Research on effectiveness and feasibility of interventions, including organizational change interventions, designed to improve adolescents' access to effective services.

Return to Contents
Proceed to Next Section