Comments on the Committee's Task

It was suggested that the public health practices framework that has been applied to local health departments may prove useful to the committee in looking at the performance of a broader range of public and private entities that play a role in protecting and improving a community's health. The set of 20 practice indicators that has emerged from the University of Illinois at Chicago–University of North Carolina collaboration may prove helpful as well. An additional suggestion was to move ahead using available knowledge and resources rather than waiting for a more complete understanding of the relationship between public health practices and health outcomes. Performance monitoring will be one way to obtain information needed to learn more about the effectiveness of some practices for improving health outcomes.

Discussion by the committee pointed to the importance of state infrastructure for local health department performance and the need to be able to assess state as well as local capacity and performance. In addition, it was emphasized that differences among states in the nature of local health departments can affect which functions can be conducted at the local level and, therefore, their apparent level of “effectiveness. ” In Massachusetts, for example, which has many small community health departments but no county health departments, leadership for essential public health practices often rests at the state level.

MONITORING AND IMPROVING COMMUNITY HEALTH: A WASHINGTON STATE CASE STUDY

Understanding the political, economic, and social systems that influence the determinants of health will be crucial for the committee as it considers sets of indicators for performance monitoring. Together, several workshop presentations offered a case study to explore how one state has been preparing its health system for a role in monitoring community health.

Washington State was chosen for this case study because of the substantial change in medical care and public health systems during a major health reform initiative. In April 1993, the state legislature passed the Health Services Act, a landmark health reform measure, to provide universal access to health insurance for all residents through a managed competition approach funded by an employer mandate, individual contributions, and state-subsidized insurance premiums. The legislation called for caps on insurance premiums, insurance reform, a statewide health services information system, quality improvement, and an oversight health care commission.

The Health Services Act also initiated the Public Health Improvement Plan, which serves as a biennial blueprint for the future public health system. The plan emphasizes the core functions of public health and population-based



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary Comments on the Committee's Task It was suggested that the public health practices framework that has been applied to local health departments may prove useful to the committee in looking at the performance of a broader range of public and private entities that play a role in protecting and improving a community's health. The set of 20 practice indicators that has emerged from the University of Illinois at Chicago–University of North Carolina collaboration may prove helpful as well. An additional suggestion was to move ahead using available knowledge and resources rather than waiting for a more complete understanding of the relationship between public health practices and health outcomes. Performance monitoring will be one way to obtain information needed to learn more about the effectiveness of some practices for improving health outcomes. Discussion by the committee pointed to the importance of state infrastructure for local health department performance and the need to be able to assess state as well as local capacity and performance. In addition, it was emphasized that differences among states in the nature of local health departments can affect which functions can be conducted at the local level and, therefore, their apparent level of “effectiveness. ” In Massachusetts, for example, which has many small community health departments but no county health departments, leadership for essential public health practices often rests at the state level. MONITORING AND IMPROVING COMMUNITY HEALTH: A WASHINGTON STATE CASE STUDY Understanding the political, economic, and social systems that influence the determinants of health will be crucial for the committee as it considers sets of indicators for performance monitoring. Together, several workshop presentations offered a case study to explore how one state has been preparing its health system for a role in monitoring community health. Washington State was chosen for this case study because of the substantial change in medical care and public health systems during a major health reform initiative. In April 1993, the state legislature passed the Health Services Act, a landmark health reform measure, to provide universal access to health insurance for all residents through a managed competition approach funded by an employer mandate, individual contributions, and state-subsidized insurance premiums. The legislation called for caps on insurance premiums, insurance reform, a statewide health services information system, quality improvement, and an oversight health care commission. The Health Services Act also initiated the Public Health Improvement Plan, which serves as a biennial blueprint for the future public health system. The plan emphasizes the core functions of public health and population-based

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary prevention rather than acute clinical care for individuals. The first version of the plan, published in November 1994, articulated how the public health system would assure accountability for its contribution to health status improvement through a set of system capacity standards and health status outcomes (Washington State Department of Health, 1994). In addition, the health care industry was rapidly preparing for anticipated changes in the delivery of and reimbursement for personal health care services. The changes in the state's health system were slowed considerably in the spring of 1995 when the legislature repealed large sections of the 1993 Health Services Act. The new legislation contained some insurance reforms, portions of the previous health data system, and quality improvement initiatives. It increased state-sponsored health care for low income individuals and families and left intact the public health system reforms reflected in the Public Health Improvement Plan. Despite the legislative changes, some activities already underway emphasized the need to develop partnerships among public, private, and academic health systems to improve community health status. The growing emphasis on cost containment, managed care, and quality also made a focus on prevention crucial. At the workshop, representatives of the public health, academic, and private health care systems in Washington State described some of the partnership activities underway and work being done on health status monitoring. Public Health in Washington State4 The Washington State Health Department functions in a state with a strong populist tradition combined with acceptance of an active role for state government. Washington's population of about 5 million is served by 33 local health jurisdictions. With an area of roughly 71,300 square miles and 50 percent of the population living in the Puget Sound area, large portions of the state remain sparsely populated. Washington's local health jurisdictions are independent of the state health department and offer relatively few personal health care services. In the late 1980s, the state reestablished a Department of Health separate from the combined Department of Social and Health Services that had been established several years earlier. The broad perspective of public health and the concerns of local health departments had not fit well with the more targeted responsibilities of the various social service programs. Several state actions leading up to this departmental reorganization set a pattern for a broad systems approach to health issues. The role of the state board of health was specifically 4   This section is based on a presentation by Kristine Gebbie.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary reaffirmed, along with its connection to local health departments and requirements for consultation across a broad range of state agencies (e.g., agriculture, labor and industry, and the state insurance commissioner). The state legislature also chose to respond to the AIDS epidemic primarily through local health departments with requirements for regional planning. Efforts to reduce low birth weight provided additional funds to areas with poor outcomes and required that they develop plans for responding to the problem. Some of these plans brought in services from private providers. A program to develop a state trauma system led not only to collaboration among a variety of providers at local, regional, and state levels but also to work on data systems. The Department of Health is a principal link between the state government and local health departments, but it also has a broader perspective on health that includes working with other state agencies (e.g., education) and with the private sector to provide for the health of the public. The department has devoted attention to developing improved data systems, which includes working with the private sector. It has also been given responsibility for monitoring and reporting on state needs for health professionals. Factors considered include numbers of providers, their geographic distribution, and the mix of specialties. State professional schools are expected to have plans for responding to identified needs. Two elements of the Washington State experience were considered especially important for the IOM committee's work. The first was the early and continuing emphasis on a systems approach that required consultation and collaboration across organizational boundaries. It was observed that voluntary efforts of this sort can be difficult to sustain. Second, the success of many of the state's activities relied on the difficult process of developing a “shared vision” among many groups for criteria for good health in the state. That vision becomes the basis for assessing public health performance and outcomes. Linking Academic Health Centers and Local Health Departments5 The School of Public Health at the University of Washington, established in 1970, grew out of the medical school's department of preventive medicine. It originally emphasized academic training for physicians and other doctoral-level professionals rather than training for midlevel public health practitioners. The school receives limited state support and relies heavily on federal research support and other grants and contracts to fund salaries and other activities. In the 1980s, CDC encouraged greater attention to public health practice issues, and a 1990 grant from the Health Resources and Services Administration (HRSA) supported the creation of a Center for Public Health Practice at the 5   This section is based on a presentation by James Gale.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary School of Public Health. The Center's goals were to provide continuing education to public health practitioners, to place students in practicum situations, and to form linkages between the school and public health practice settings. Many faculty members and graduates began working with local health departments. An assessment of training needs in health departments identified interest in epidemiology, public health law, management, administration, and leadership. The Center responded, using HRSA grant funds, with a two-week summer institute with 30–60 participants. In seeking support to continue the program, the Center found that health departments wanted broader staff access to more practical training in the essentials of assessment techniques, analysis and presentation of data, and community organizing. In response the Center, in collaboration with the state Department of Health, the local public health community, and several university programs, has developed a series of day-long training modules that are being offered via satellite in seven locations throughout the state. About 1,100 people will have been able to participate. Other collaborations are also proving beneficial. A tenured faculty member is serving as health officer for a largely rural county in central Washington. This arrangement provides a training site for students in nursing, preventive medicine, and public health, and establishes a direct link for the faculty with local health officers and the Washington State Association of Local Public Health Officials. In addition, ties with the state Department of Health have been strengthened by cross appointments. For example, the state health officer serves as assistant dean for public health practice. In another arrangement, a joint search was conducted by the Department of Health and the School of Public Health to select a single individual to serve as state epidemiologist and to hold a full faculty appointment in the Department of Epidemiology. Some of these collaborations also assist the university by adding faculty with state-funded salaries. The links with public health practitioners are adding new perspectives to the content of the academic program in the School of Public Health. Overall, however, faculty participation in practice-oriented activities remains limited because of competing priorities, including the need to consider the typically research-oriented requirements for tenure. Partners for community-based projects have been found elsewhere in the university in the Department of Family Medicine and the School of Nursing. In addition, the School of Social Work has been involved in a community project on violence prevention. Workshop discussion suggested that local and state health departments would benefit from easier access to the technical expertise of academic health centers, particularly in analytic areas such as biostatistics.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary Private Sector Participation in Community Health Activities: The Role of Group Health Cooperative of Puget Sound6 Group Health Cooperative of Puget Sound is the nation's largest consumer-governed nonprofit health maintenance organization (HMO). It was established in 1947 and now has about 540,000 enrollees, most of whom live in the Puget Sound area. Recently, as Group Health has expanded its geographic scope, it has begun adding physician networks to its original organization as a staff model HMO. Group Health has an active program of performance monitoring that serves a variety of purposes: HEDIS reporting to employers on health plan performance; accreditation requirements; data for an internal quality initiative; feedback to individual clinicians; marketing aimed at increasing enrollment; and reporting to the Cooperative 's consumer governance structure. Performance measures, which include clinical and consumer satisfaction elements, are developed based on sources such as the Cooperative's strategic plan and periodic vision statements. In the area of prevention and health promotion, 11 priorities have been specified.7 Four years ago, Group Health adopted a vision statement that calls for delivery of quality health care to the entire community, not just its enrolled population. At the same time, a set of community service principles were adopted. These steps gave formal recognition to and expanded on the work that Group Health had already been doing in the community and in the areas of health promotion and disease prevention. Attention is currently focused on four areas: childhood immunization, the reduction of infant mortality, health care for homeless families, and the reduction and prevention of interpersonal violence. Sources of support include revenue from premiums, grants, volunteer services, and community partnerships. Although the organization has a tradition of working with the community, the increasing competition in the health care market is creating a need to demonstrate more clearly that the community-based activities bring a benefit to the Cooperative. Several factors were considered in undertaking community programs. Improved community health is expected to lead to improved health for members. In some cases, members—or future members—may benefit as part of the target audience for specific programs. Visible community-based programs may also help Group Health successfully compete for and retain contracts such as those with large employers and for services to Medicaid and Medicare populations. Some important health problems such as violence and alcohol abuse require community-based programs because they are not easily addressed 6   This section is based on a presentation by Bill Beery. 7   The 11 prevention priorities are tobacco use, alcohol consumption and abuse, depression, cancer, high-fat diet, inactivity, diabetes, immunization and infectious disease, HIV/AIDS, heart disease, and injuries.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary in a medical care framework. These programs also provide opportunities for community service that many staff members value. They help fulfill the obligation associated with nonprofit status to provide benefit to the community. In addition, public relations benefits are expected. The community-based programs are also expected to foster innovative approaches to meeting members' needs and to improve the organization 's ability to respond successfully to a more diverse membership and workforce. It was noted that altruism continues to be an important motivation but is not sufficient to outweigh long-term financial considerations. Evaluating the impact of Group Health's programs on community health and assessing the costs and benefits for the organization will be challenging. Group Health must address the competition for resources within the organization that creates pressures to focus exclusively on members' health care needs. There is as yet no consensus on the extent to which a health care organization 's responsibilities extend into areas often considered “public health.” Group Health is committed to efforts on behalf of the community's health but also believes that it will be difficult to sustain that commitment unless other healthcare organizations accept a similar responsibility, including public reporting on the extent to which their efforts are meeting expectations. It was suggested that it may prove harder for less established health plans or plans in highly competitive markets to make the investment in community programs. Communities can, however, promote the expectation that health plans will participate in such programs. Performance monitoring presents an opportunity to assess individual healthcare organizations and also to encourage efforts such as health promotion rather than care for preventable illness. It may be possible as well to monitor the extent to which health plans develop partnerships with local health departments and other community groups. Such partnerships can be viewed as a component of a community's capacity to meet the health needs of its population. It was suggested that some organizations (e.g., health plans, employers) might find it easier to see their connection to the concept of community service monitoring than to public health performance monitoring. From Group Health's perspective, the committee's vision statement sets out an appropriate challenge for all health plans. Performance monitoring will, however, require developing explicit expectations in several areas: community responsibilities of medical care organizations; monitoring and reporting by those organizations; encouraging and facilitating public–private partnerships; defining contributions to community capacity-building; conducting community programs in collaboration with the community; determining how to monitor the extent to which true partnerships exist; and establishing the appropriate role of the consumer (individuals, groups, or the community) in developing and monitoring community health efforts.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary The discussion highlighted the concern that performance measures will focus attention on some tasks to the exclusion of others. Because this is often true, it was emphasized that the selection of measures is a critical process that must be carefully and openly considered so that, as much as possible, the implications of those selections are understood. Workshop comments also reflected concern that most healthcare organizations view measurement tasks very narrowly. Group Health has taken a broader approach with, for example, an information system that not only provides data on immunizations administered but also provides tools for reminding providers and parents that immunizations are needed. There is concern, however, that it may be difficult to continue taking that approach if competing groups do not make a similar effort. Addressing Environmental Health Issues8 Although there are important links between environmental health issues and public health, some members of environmental health agencies have seen themselves as having separate responsibilities from what are perceived as personal health issues in public health. This has encouraged an organizational distance between the two fields that has left many environmental health professionals out of the discussions about new approaches to public health practice and without an appreciation of their role in public health systems. In Washington State, efforts are being made through means such as the state's Public Health Improvement Plan to bring environmental health clearly into the realm of public health. It was emphasized that environmental health needs to develop new approaches to data collection and analysis that demonstrate more clearly the links between environmental health functions and good health outcomes. In Seattle–King County as in most other communities, the focus has been on process measures (e.g., numbers of restaurant inspections conducted or number of water samples tested) without knowing to what extent those activities contribute to disease prevention. The current pressures to reduce the size of government and to reduce regulatory burdens are increasing the need to demonstrate the benefit of the activities an agency conducts. There is also concern that the rapid development of new information systems is not being guided by an understanding of what data will be of greatest use. The committee's vision statement was endorsed with comments on some issues that it raises. One concern is ensuring that staff members are given adequate training to understand the purposes of performance monitoring and the information it requires. Planning for information for the public, including non- 8   This section is based on a presentation by Carl Osaki.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary English speaking groups, is also important. A related issue is the extent to which the community can or should participate in selecting performance measures to be used. A further concern is how to integrate or reconcile data collection and information systems for health-related performance monitoring with similar activities initiated by other community agencies. Environmental health directors from across Washington State have begun working together to try to address concerns such as these. The need for clear and comparable definitions of terms, across communities and across disciplines, has emerged as a central issue. In addition, the group is trying to address how to collect information that can be useful in efforts to improve community health status. They are also looking at differences in the “customers” for information on personal health programs and on environmental health services. One product from the group is an environmental health addendum to the APEXPH materials for community assessments (Washington State Department of Health, 1993). The state's environmental health directors are also trying to formulate a new model to use in thinking about the links between a community's health and environmental factors and what performance standards might be applied. Their initial discussions led to a “reactive” model based on the interactions among environmental hazards, populations at risk, and health outcomes. Overlap among the three signals a health problem that an intervention might be expected to reduce or eliminate. They also have considered a new approach that looks at health outcomes at the intersection of a population at risk, unsafe behavior, and an environmental hazard.9 Each component of this model seems to have measurable elements that might guide preventive efforts to avert their intersection and an adverse health outcome. Understanding the relationships among these components will be critical. At the workshop it was suggested that geographic mapping of those measures might be a way to identify a convergence of factors and likely adverse outcomes. A particular focus of discussion was the term “unsafe behavior,” which many thought could be misunderstood as referring only to the behavior of the population at risk rather than to actions by a broad range of parties. In Washington State's outbreak of food poisoning from Escherichia coli contamination, the unsafe behavior was improper cooking of hamburgers not the actions of those who became ill by eating the hamburgers. 9   Since the workshop, there has been further refinement of the model, influenced in part by comments offered at the workshop. It bases an assessment of health status on the interaction of environmental hazard(s), population(s) at risk, and public health protection factors. The model demonstrates the need to consider more factors than the regulation of hazards and to take a broad community perspective. The revised model has been endorsed by directors of environmental health in Washington State's local health departments and by the state Department of Health.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary Developing useful performance standards for environmental health will require further work to ensure that the actions being measured have an impact on health. An important aim is to focus attention on issues beyond environmental hazards. For example, in Seattle– King County educational efforts are directed at the 10–20 percent of food workers who fail a mandatory food workers knowledge test. In discussion it was noted that although this may reduce some health risks, it is probably not a sufficient response to avoid all food handling problems. The workshop participants were very interested in the new model and saw how its concepts might apply beyond the traditional bounds of environmental health. It was emphasized that community-based assessments need approaches to performance monitoring that can link all of the determinants of a community's health including the environmental and personal health care perspectives. A Voluntary Approach to Developing Public–Private Health Information Systems10 As part of its 1993 health care reform program, Washington State mandated creation of a statewide health data system that would be used to monitor and evaluate the effectiveness of reform efforts. Subsequently, the state has stepped back from the original mandates to adopt a more voluntary approach to health care reform. Health data system issues are now being addressed through a public–private partnership facilitated by the state. It was noted that the change has overcome some of the resistance to a state-run data system and may promote progress. The state is serving as a convener and manager of the process but has no role in directing what will be done. A variety of participants who would have little other opportunity to work together on these issues have accepted the state's role. Work started under the Health Services Information System, part of the 1993 health reform program, has provided a valuable foundation for much of the public–private collaboration now underway. In the revisions to the state's health reform program, the state Department of Health retained responsibility for considering data standards (as well as its traditional responsibility for population-based data collection systems such as vital statistics and reportable diseases). With this responsibility linked to quality improvement efforts at the state and local levels, there is an opportunity to look broadly at the whole health system in the state. That approach should also avoid the emphasis on technology and engineering issues that characterizes some of the work on data standards. Several groups have been formed to address specific areas such as patient care, health status, and community assessment. 10   This section is based on a presentation by Elizabeth Ward.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary Members are drawn from local health departments, health plans, employers, purchasers, and providers. The process is moving ahead as each group comes forward with a proposed set of standards. Rather than waiting for all of the groups to complete their work, proposals are being tested as they become available. Collaboration among the groups working on data on patient care, community assessment, health status, and enrollee measures has produced a proposal for a “clinical outcome measure amended HEDIS” strategy, referred to as COMAH. It combines clinical outcome measures with the more process-oriented HEDIS measures; only a few measures in this set will be community-based. Collecting data for the COMAH measures will be tested in a variety of settings including physicians' offices. Other groups will be producing proposals that focus more on community health measures than on the clinically-oriented COMAH measures. It appears that communication between participants with a personal health care perspective and those with a public health perspective has made health care providers more interested in having access to population-based data as well as clinical data. It was suggested that, for community-oriented health plans such as Group Health Cooperative of Puget Sound, Washington's collaborative approach to data systems development may offer a way to receive public recognition for some of their activities and to have an influence on what other health plans are willing to do. The voluntary collaboration among the various groups with an interest in health data is not intended to produce a statewide data system. There are expectations, however, that common interests will eventually lead to the equivalent of a state system. At present there are no mechanisms for aggregating data across providers, but some groups are hoping to be able to take that step in the future. A discussion of data entry issues pointed to the risk that this disaggregated approach may create a substantial data entry burden unless data systems can be coordinated. One goal might be developing ways to make routinely collected data items (e.g., birth weight) available to many users so that they need not be reentered in various data systems. It was suggested at the workshop that Washington's experience with a voluntary public–private approach to developing data systems may be a useful reality check for the IOM committee on the kinds of information health plans and others are willing to share. The COMAH measures illustrate the selection of indicators that reflect the interests of the participating groups rather than externally established or coordinated criteria for an appropriate or complete set of indicators. In the current voluntary and cooperative environment, the Department of Health is relying on the Public Health Improvement Plan to represent the broad perspective on information needed to protect the health of the public.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary Community Health Assessment in Seattle–King County11 The experiences of Seattle–King County health department's community health assessment program provide a variety of valuable insights into the process of monitoring performance to improve the community's health. The program provides information on the community 's major health problems and strengths, on perceptions about health-related concerns, and on health services. The results guide and motivate the development of policies and interventions, including action by community groups. Responding to the community's consistent interest in learning about its positive resources helps build community support for the assessment process. It was emphasized that the assessment program is a process, not just publishing a report. A critical part of the process is reviewing data with the community and with decisionmakers to understand points such as whether the results seem reasonable, whether there are gaps between findings and perceptions, and whether there are concerns that have not been included. The review also identifies areas of special interest to the community and generates guidance on how to treat sensitive issues. In Seattle–King County, health department reports are released only after this kind of consultation. Assessment Domains The domains covered by the assessment process include health status, risk factors and other determinants of health, and health interventions. Morbidity and mortality data for injuries and a variety of illnesses and health conditions are a basic health status component. Of considerable interest, but only recently addressed in Seattle, are measures of functional status and quality of life. Behavioral risk factors are included in the assessment process. Good data are needed on social factors affecting health, but standard measures have not been developed. Variables such as income, education, and race, for example, often serve as proxies. Other factors for which better measures are needed include stress, social support, and community values. In monitoring community health interventions, information is needed on the nature of the interventions and the targeted recipients (e.g., specific individuals, subpopulations, or the entire community). It is also important to know who is providing interventions. This might include public health agencies, office-based providers, health plans, hospitals, employers, schools, and so on. For many interventions, however, monitoring is hindered by the limited evidence on the their effectiveness. Other assessment areas include measures of various 11   This section is based on a presentation by James Krieger.

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary dimensions of access to services (e.g., utilization, continuity of care, financial barriers) and measures of satisfaction with services. Data Sources In Seattle–King County, data for assessments come from a variety of sources, including vital statistics, reporting programs for specific conditions (e.g., some infectious diseases, AIDS), hospital discharges, environmental inspections, and crime reports. No data are routinely available, however, on outpatient or emergency department care. Use of cancer registries is beginning. Other registries (e.g., for immunization) would be valuable but few exist, and it was suggested that Washington 's current voluntary approach to data system development might slow their creation. Data on some environmental factors can be difficult to obtain with sufficient geographic detail for analyses at a county or subcounty level. Local surveys are used, but resources are not available to cover all areas of interest. State and national surveys, which include some of these topics, are not large enough to generate local estimates. Expanding the size of samples in state surveys, oversampling of some populations, or collecting additional information on residence are ways to make state data more useful to local areas. For example, adding zip codes to the Washington State behavioral risk factor survey has now made it possible to look specifically at data for King County. There is also a project to conduct a local survey on behavioral risk factors. In addition, the county has initiated a minority health survey and is collecting information on discrimination in receipt of health services. The health department is gathering information in other ways as well. Special, detailed studies, such as hospital record reviews, are used to address some health problems. Focus groups and selective interviews are providing qualitative information to complement some quantitative data. Market research data may be able to provide useful “lifestyle ” data. Comments during the workshop suggested the need for data on the various policies, public and private, that affect health. Also raised, was the possibility that data on health plan enrollees or from programs like Medicaid could be adapted to provide meaningful information about the larger community. The value of combining data from multiple sources and making the best use of available data sources (which can avoid the cost of original data collection) was emphasized in the discussion. Some recurring technical concerns related to the lack of clear standards for data and analysis were noted. Problem areas include classifying race and social status. Inconsistent age-adjustment practices in national data complicate comparisons across data sources and with King County data. Also mentioned

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary were differences in the diagnostic coding used to define diseases or conditions and in practices followed in statistical testing. Populations of Interest The health department has defined its population of interest on the basis of residence. Analyses of subgroups within the population are frequently valuable and require information on characteristics such as age, gender, race and ethnicity, educational attainment, occupation, and income. Some analyses might require information on employers or health plan membership. Many analyses are based on the county 's 21 health planning areas. Experience has shown that these more “local” data generate greater interest and impact than countywide data. Determining the size of the denominator population, and relevant subpopulations, is an important and challenging task at the community level. In Seattle–King County, census data are supplemented with intercensal estimates from state and local agencies and from commercial vendors. A model is available to generate population estimates down to the census tract level. It was suggested that health departments might benefit from access to additional expertise in making these kinds of estimates. Selecting Indicators The specific indicators used in the Seattle–King County assessment process reflect input from sources such as the statewide Community Data Task Force, various constituencies within the county, and the county health department itself. Ongoing consultation with interested groups contributes to continuing evolution of the list of indicators. Factors considered in selecting indicators include the incidence or prevalence and the severity of a condition. The perceived importance in the community and likely community response are also considered. Other factors include the cost and availability of data and consistency with indicators used elsewhere (e.g., Healthy People 2000 [USDHHS, 1991]). Indicators for conditions for which change is difficult or unlikely were considered less helpful. Using Assessment Data In Seattle–King County, information generated by the health assessment program is being used in establishing priorities for health interventions and in allocating state funds from the Public Health Improvement Plan. The assessments are also providing information needed to target interventions and are guiding program and policy development. For example, data on youth

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary violence demonstrated that a proposed curfew would not affect the after-school and evening periods when most incidents occurred. The data generated by the county assessments are also used for comparisons with state and national data and with the Healthy People 2000 targets. For example, rates of hospitalization for asthma in county health planning areas are useful both as an environmental health indicator and as an indicator for preventable hospitalizations for an ambulatory care-sensitive condition. Asthma control programs have been developed in areas with high hospitalization rates. Illustrating the value of combining information from different sources, it was noted at the workshop that other data show that one of these high hospitalization areas also has especially high levels of woodburning stove use. An analysis of trends in births, pregnancies, and abortions illustrated the importance of being able to look at differences in the patterns in various population groups. An analysis by income, however, required an ecological approach using median income by census tract. Supplementing the quantitative data with qualitative information provided valuable insights. Interviews and meetings with young people, community groups, and health care providers gave additional information on attitudes toward pregnancy and abortion, on how those attitudes were being shaped, and on misinformation about health risks and legal and financial barriers to services. It was noted that some of the assessments that health agencies would like to make are complicated by resistance within the community. For example, data on sexual activity among teenagers are valuable in addressing issues such as teenage pregnancy and controlling sexually transmitted diseases, but parental consent is now required in Washington State before asking students relevant questions. The state has chosen not to include that subject in its school survey. A suggestion at the workshop that health plans and other health care providers might be an alternative source of similar information brought the observation that sensitive information may be difficult to collect in that setting as well. Promoting Community Participation Seattle–King County has found that coalitions of community stakeholders (e.g., public health agencies, health plans, hospitals, providers, employers, etc.) need to be developed early in the assessment process. Such groups can provide valuable guidance on selecting indicators, interpreting assessment results, and understanding their policy implications. Public meetings and advisory groups that include community leaders can involve an even broader segment of the community in health assessment and planning. This kind of participation promotes greater “ownership ” of the process and the results. Facilitating access to assessment data has also increased support for these activities. Currently in

OCR for page 17
Using Performance Monitoring to Improve Community Health: Exploring the Issues: Workshop Summary Seattle–King County, data are available to a relatively limited technical audience, but there are hopes to provide broad community access. It was noted that health departments are generally a resource for essential technical and organizational services for community health assessment. They can provide the expertise and computing facilities needed to frame some indicators and to perform data management and analysis tasks. They can also help bring together the community stakeholders and help build coalitions. The discussion highlighted the different skills that technical and community organizing tasks require. It was suggested that schools of public health might make a valuable contribution by ensuring that students have training in both areas. The growing interest in the role that health plans play in meeting community health needs was noted. They are a source of information about their members and should be able to benefit from knowing more about the factors affecting health in the community. Their collaboration in community health assessment may encourage, and be encouraged by, the development of common goals for member and community health. It was suggested that establishing common community practices in technical areas such as data standards and data interchange might promote health plan participation. An assessment of health plan performance, which might use indicators such as those provided by HEDIS, is also seen as a necessary part of these activities. Observations to the Committee Final remarks in this presentation highlighted several issues in monitoring community health. It was emphasized that monitoring needs to be a dynamic process and that it should promote local involvement. Indicators used for monitoring need to focus, to the extent feasible, on risk factors with interventions known to be effective. Research is needed, however, to establish the effectiveness of a much broader range of interventions. Indicators should also address not only risk factors for poor health outcomes but also factors that protect against poor outcomes. In addition, indicators should be sensitive to changes in the organization and delivery of health care services. Some specific areas in which the committee might be helpful were noted: proposing indicators; encouraging the development of indicators for less developed domains such as environmental or social determinants of health; suggesting data standards for defining populations; providing guidance on effective presentation and dissemination of assessment results; and outlining effective processes for involving community stakeholders in monitoring. Part of the committee's contribution may be to assemble available expertise in these areas.