Health Impact Assessment as a Tool for Population Health Promotion and Public Policy

Discussion and Analysis

Measuring the Impact of Public Policy-What Do We Need?

Health Indicators

In their review of healthy public policy, Pederson, Edwards, Kelner, Marshall, and Allison (1988) called for the establishment of health indicators that could be used to elucidate the relationship between healthy public policy and health status. They highlighted the importance of obvious indicators such as morbidity rates and disease-specific mortality rates, as well as positive measures of health and well-being (which are more difficult to quantify). Most important, Pederson et al. suggested that indicators of phenomena not always recognized as health-related, but which are causally linked to health, must be developed for evaluation efforts. To this end, the recommendations developed at the second international conference on health promotion, in Adelaide, South Australia ("The Adelaide Recommendations," 1988), called for the development of health information systems.

Relating this recommendation to health impact assessment, we would see the use of such a health information system as providing performance measures to be associated with health impact measures available from existing vital statistics systems, discharge and other medical care record or billing systems, pharmacy record systems, tumour registries, and other registry systems, and periodic population surveys. The indicators of causally important determinants of health can and should be used as benchmarks of progress toward improving health, but they cannot be taken as equivalent to health impact without stretching the definition of health beyond the credibility and tolerance of other sectors.

In 1984, an international conference, the "Beyond Health Care Conference," was held in Toronto on healthy public policy. Following the conference, a one-day workshop, "Healthy Toronto 2000" was held to examine how the broad themes of healthy public policy could be applied at the municipal level. So began the "Healthy Cities" movement (Manson-Singer, 1994; Rachlis & Kushner, 1989). In Canada, the project, named "the Canadian Healthy Communities Project, (CHCP)" aimed to ensure that health was a "primary factor in political, social and economic decision-making" at the municipal level (CHCP cited in Manson-Singer, 1994, p. 111).

The CHCP started out with great promise but had a short life. Manson-Singer (1994) explicated many reasons for its failure, including inexperience, lack of resources, definitional problems, competition with other social movements, and poor relationships with funders. The problem most relevant to our current discussion was the difficulty associated with arriving at a suitable definition of "healthy communities." This problem parallels our caveats of defining health too broadly. In defining the concept broadly, the steering committee of CHCP believed that they were being inclusive, but those at the municipal level were concerned that the Federal government was passing on its responsibility to the municipalities. Meanwhile, those in federal health funding positions had difficulty justifying or defending the broad definitions as having sufficient focus on health.

Rootman (1990) examined whether indicators were required within the healthy communities context. His observations seem relevant to the broader notion of health impact assessments of all public policy and the search for appropriate indicators. First, he asked, how do we know what indicators mean in the absence of a conceptual framework or theory to locate and explain them? Second, Rootman pointed to some methodological issues related to the identification and development of appropriate indicators including:

• indicators are not currently available for the positive dimensions of health (i.e., other than reductions in morbidity and mortality);
• it is difficult to develop "holistic" measures that combine objective and subjective dimensions of health;
• it is unclear how one would capture the contextual factors related to health status, and
• there are no indicators targeted at the community level; only indicators based on the aggregation of individual measures.

The Healthy Communities strategy is being played out in many communities in the province of British Columbia. Within the Healthy Communities context, B. C. communities create their own vision of a healthy community, develop community profiles that highlight both community needs and strengths, and forge coalitions and partnerships committed to improving the health of citizens. The B.C. Ministry of Health has encouraged communities to define healthy community measurement indicators as a means to define and track progress toward improved health status of people and communities. Using indicators to measure the health of communities is considered a means by which communities can influence decision makers and participate in program planning and policy development processes that ultimately impact them. Further description of the Healthy Communities Project in British Columbia is offered in Appendix A, Abstract 12.

Discussion related to the identification of appropriate indicators has occurred, most often, within the context of the Healthy Cities movement. However, even there, committed individuals seem to have failed to establish an agreed upon set of indicators. The World Health Organization Healthy Cities Project of the WHO Regional Office for Europe sponsored a meeting in March 1987, in Barcelona, to achieve a consensus on what could and should be measured (O'Neill, 1990). Criteria agreed upon for the selection of indicators included that they must be:

• relatively simple to collect and use;
• sensitive to short-term change;
• capable of analysis at the small-area level;
• related to health, "Health for All," health promotion, and the Healthy Cities Project;
• able to carry social and political "punch";
• limited to approximately 30 in number;
• concerned with all aspects of city life; and be
• both subjective and objective (WHO Healthy Cities Project, 1988).

In response to the Barcelona Workshop, A Guide to Assessing Healthy Cities (WHO Healthy Cities Project, 1988) was developed. However, as O'Neill (1990) pointed out, the plan could not be operationalized because of its massive size; it included everything from knowing the geography of the city, including its topography, climate, natural resources, biological ecosystem, and "urban form," to its history, demography, political structure (including its jurisdiction and governance), economy, social issues, influence of religion and the churches, and a "general sense" of the city. This appears to be one more instance of mixing performance indicators and background data with the need for health impact indicators. At a subsequent invitational conference on Evaluation of Healthy Cities in Maastricht, The Netherlands, none of the reports presented offered concrete indicators for health impact assessment (de Leeuw & O'Neill, 1992).

The Healthy Cities and Shires Project in Australia is an exemplar of the WHO's Healthy Cities Project as a means to achieve "Health for All by The Year 2000." Under the Healthy Cities and Shires Project, municipal governments develop a vision for their city, identify needs, set priorities, define measurable health goals and targets, and monitor and evaluate progress toward stated goals. In 1993, the Healthy Cities and Shires Project was piloted in three cities in the state of Queensland. With only a few years of operation, it is difficult to assess the Australian experience at this time. Further description of the Healthy Cities and Shires Project in Australia is presented in Appendix A, Abstract #11.

It now appears that the WHO European Healthy Cities Project has modified its approach to indicators, admitting the difficulty associated with a standardized approach. Work conducted in other parts of the world, and reviewed by O'Neill and Cardinal (1992) and Cardinal and O'Neill (1992), seems to have resulted in the same conclusion. There is no simple and uniform way to assess healthy cities.

Recently, the mayors of Canada's fourteen largest municipalities recommended that a quality-of-life index be developed to assess the effects of federal and provincial spending cuts on the "livability" of cities. The proposed index would consider such factors as the proportion of people living under the poverty line, food bank usage, the numbers on welfare, and the types of community services offered (Munro, 1996). This mix of impact indicators does not include morbidity, disability, or mortality indicators, but it would amount to a combination of determinants and consequences of changes in health.

Determinants of Health

As described earlier, the Lalonde (1974) report introduced the Health Field Concept of Laframboise (1973) in the first major policy document seeking to reorient the emphasis of government health policy from medical care to three other major determinants of health. These were identified then as lifestyle, environment, and human biology. The Canadian Institute for Advanced Research (CIAR, 1991; Evans & Stoddard, 1990, 1994), building on the work of others who provided evidence for the importance of the other determinants of health relative to medical care (e.g., Dutton, 1986; Levine & Lillienfeld, 1987; Marmot, 1986; McKeown, 1979; McKinlay, McKinlay, & Beagehole, 1989; Townshend & Davidson, 1982), separated environment into social environment and physical environment. It also separated human biology into genetic endowment as a primary determinant and biological response to the environmental and genetic determinants as part of individual response along with behavioural response as a secondary or more proximal determinant of health. They further separated health into health and function, disease as immediate effects of these determinants, and "well-being" as a longer-term or secondary effect.

The significance of this formulation has been substantial in the reorganization and reorientation of Canadian health bureaucracies and priorities in health policy and program development. It has resulted, among other effects, in a renaming of federal and provincial units in health ministries, and a refocussing of the Ministers of Health on population health issues associated with these determinants (Federal, Provincial and Territorial Advisory Committee on Population Health, 1994).

Another significance of the CIAR formulation of population health determinants with more direct relevance to health impact assessment is that the model and its documentation clearly disaggregates the concept of health. In search of more practical alternatives to "the all-encompassing definition of the WHO, almost a Platonic ideal of 'the good' " (Evans & Stoddard, 1990, p. 3), they separate biological responses, which could be equated with physiological "risk factors," from functional limitations and health as capacity, from illness or disease, and from well-being. They conclude that:

There are no sharply drawn boundaries between the various concepts of health in such a continuum; but that does not prevent us from recognizing their differences. Different concepts are neither right nor wrong, they simply have different purposes and fields of application. Whatever the level of definition of health being employed, however, it is important to distinguish this from the question of the determinants of (that definition of) health. (Evans & Stoddard, 1990, p. 4)

These important distinctions by the progenitors of the population health model driving the current interest in health impact assessment lead us to conclude that health impact must be assessed on the basis of measurable outcomes on this continuum of health and cannot be equated with impact on the determinants of health, or the presumed consequences of health (e.g., well-being, medical care).

Assessing the Policy Environment

In addition to the development of indicators to assess the health impact of policies, Pederson et al. (1988) called for the development of specific indicators to measure the very presence of healthy public policy. In this way, researchers could track the various phases of policy-making including:

• the identification of social organizations and institutions engaged in policy-making;
• the identification of policy-making processes and outcomes;
• a description of policy directions; and
• an analysis of past, present, and future trends.

The suggestion of Pederson et al. (1988) is consistent with the report prepared by Kickbusch, Macdonald, and King (1988) for the Adelaide Conference on Healthy Public Policy wherein it is recommended that, to facilitate healthy public policy, analyses of physical, social, and economic factors, political priorities and political commitment are necessary.

Again, we must agree with the importance of these other analyses as indicators of determinants and performance, but they cannot stand alone as health impact assessment.

Health Objectives, Goals and Targets as a Strategy for Health Impact Assessment

One approach that may stave off the pitfalls associated with the adoption of expanded definitions of health and the confusion of determinants and outcomes, described above, is to articulate a health strategy, or more specifically, to specify health goals and objectives for the nation and/or provinces. In so doing, a framework is provided by which relevant indicators or outcomes can be identified for the health impact assessment process. While general policy directions are important, the setting of measurable objectives, with deadlines, is an important motivator for action. Such objectives obviate the need to assess the effectiveness of previous actions and the feasibility of future proposals (Asvall, 1988). In the absence of achievable objectives and targets for health agreed upon as policy, the relative merits of other proposed policies as to their impact on health can be argued endlessly. Goals and objectives provide the essential yardstick for assessment.

Although individual provinces have developed, or are in the process of developing, health goals, the development of national goals has been elusive. Pinder (1994) provided evidence that, on several occasions, and from different stakeholders, calls have been made for national goal setting; goal setting was advocated in the Lalonde Report (Lalonde, 1974), by the Ad Hoc Committee on National Health Strategies (Canada, 1982), the Canadian Public Health Association (1984, 1987, 1992), and Deputy Minister of Health, Dr. Maureen Law (1989a, 1989b).

At the national level, objective setting has been limited to select priority areas such as tobacco demand reduction, the drug use strategy, children's health, and injury control. For example, those concerned with injury control held a symposium in May 1991 to establish a national strategy and to attempt to establish national objectives focusing on injury prevention, based on the U.S. framework outlined in Healthy People: Health Objectives for the Nation (i.e., Year 2000 Injury Control Objectives for Canada Symposium, May 20-22, 1991, Edmonton, Alberta). The symposium brought together representatives of government, public interest groups, professional associations, academia, standard setting organizations, workers, employers, injury survivors, and the general public who developed injury control objectives for Canada in four settings: home and community, occupational health and safety, sport and recreation, and transportation. A fifth group was convened to examine the feasibility of setting objectives related to violent and abusive behaviour (Saunders & Stewart, 1991). Setting injury control objectives is one of the few examples of a Canadian approach to setting targets by health priority area. This initiative, referred to as A Safer Canada: Year 2000 Injury Control Objectives for Canada, is described in greater detail in Appendix A, Abstract #15.

Another notable example of national goal setting in Canada is the National Strategy to Reduce Tobacco Use. The goals of the national strategy are to assist nonsmokers to remain smoke-free (prevention), to help those who want to quit to do so (cessation), and to protect the health and rights of a nonsmoker (protection). Seven strategic directions are outlined to meet these goals including legislation, access to information, availability of services and programs, support for citizen action, message promotion, research and intersectoral policy coordination. The tobacco reduction strategy is a collaborative initiative between the federal and provincial/territorial governments and various national health organizations including the Canadian Council on Smoking and Health, the Lung Association, the Heart and Stroke Foundation and the Canadian Public Health Association. Please refer to Appendix A, Abstract # 14 of this report for further description of the National Strategy to Reduce Tobacco Use.

Similarly, national goals have been developed for cardiovascular disease in Canada. The Canadian Heart Health Initiative (CHHI) links the federal health authority with all ten provincial/territorial health ministries and with communities where demonstration projects are conducted. Today, a national network of heart health initiatives exists and is characterized by shared responsibility among government jurisdictions. The national jurisdiction is responsible for technical support, the establishment of a national heart health risk factor database, and the provision of matching research funds for surveys and demonstration projects. Provincial governments are responsible for conducting baseline surveys of risk factors, developing action plans, and implementing and evaluating community-based demonstration projects. Further description of the Canadian Heart Health Initiative is provided in Appendix A, Abstract #16.

The management-by-objectives approach to planning has been practiced with increasing regularity across nations and through World Health Organization auspices and encouragement. The essential logic of this approach is that goals, objectives and targets can be specified with the levels of achievement and the dates they are to be attained projected from a current or recent starting point. Such goals, objectives, and targets provide a clear road map of where the policies and programs should be pointed, what pace they should be progressing in their impact, and what health outcomes they can be expected to achieve. A health objective takes the form of a single sentence that states (1) who, usually stated as a population group, (2) will achieve how much, usually stated as a morbidity or mortality rate or percentage target, (3) of what change, expressed as the health problem or need, (4) by when, usually expressed as a year within a ten-year time frame.

The process of developing health goals, objectives and targets will account for the extent of their support from various levels of government, various sectors and various organizations. The wider the participation in developing and ratifying the objectives, the greater can be the acceptance and active dedication of resources to their achievement by potential organizational partners at all levels in all relevant sectors.

U.S. Healthy People Initiative

The most systematic and sustained process of health impact assessment by a government in guiding health promotion and disease prevention policy has been the planning-by-objectives process that the U.S. Public Health Service has led for the past sixteen years. As a contribution to American national and state health policy and programs, the documentation of the sixteen years of effort is monumental among federal government efforts. Its contribution to the changes in health status of Americans can be debated because so much else has happened in these sixteen years, but progress is measurable and undeniable in the recently issued Healthy People 2000: Midcourse Review and 95 Revisions (U.S. Department of Health and Human Services, 1996). Setting goals, objectives, and targets as a strategy for health impact assessment is the cornerstone of the Healthy People 2000 initiative.

How much the planning, policy-making and program and data development efforts can take credit for much of the progress in reduced morbidity, disability and mortality, how much of the progress can be attributed to improved planning and policies at the national level, how much to the secondary influence on improved planning and policy at the state and local levels remain equivocal. These questions will be debated for decades to come. At the very least, the debates will have the history documented by the midcourse reviews of 1985 and 1995, as touchstones. They are unsurpassed in rich detail on the historical linkages between health promotion and disease prevention policies, goals, objectives, programs, and services, and their consequent reductions in population risk, environmental improvements, and population health outcomes.

The U. S. health promotion and disease prevention initiative influenced directly or provided a model for state and local levels, and to other countries. The diffusion effect from national to state (Centers for Disease Control, 1990) and local (American Public Health Association, 1991) levels in the United States stands out in some of the volumes documenting the Healthy People initiative. One of the objectives for the year 2000, for example, is "to increase to at least 90 percent the proportion of people who are served by a local health department that is effectively carrying out the core functions of public health." The bottom-up influence from local and state constituencies in formu-lating the Healthy People 2000 objectives was impressive. The ripple effect outward to other countries counts as an additional international contribution.

One historical turning point in the management-by-objectives process was the point at which the objectives gained teeth as a force driving the planning by agencies of the Public Health Service. After a two-year start in the final stages of the Carter Administration, Dr. Edward Brandt, the Assistant Secretary who arrived with the Reagan Administration in 1981 to replace Dr. Julius Richmond as Assistant Secretary of Health, saw fit to retain the objectives as policy in the Republican administration that inherited them. Dr. Brandt added the powerful policy inducement of requiring the agencies of the Public Health Service to justify their next fiscal year budgets based on their contributions to achieving the objectives. We note this historical event with particular emphasis because it was clearly a decision that resulted in far greater influence of the outcomes-oriented, objectives-based planning in federal government agencies than they had initially. Agencies could no longer pay lip service to the objectives and then continue to pursue their former priorities and conventions. Now they had to show clearly how each of the activities or programs for which they were requesting budgetary support would contribute to accomplishing one or more of the objec-tives in disease prevention or health promotion (Green, 1996).

The success of the U. S. Healthy People initiative suggests some lessons for the content of the documents that accompany a federal initiative in health impact assessment and policy. It suggests even more for the process of their development. Central to the whole experience in the U.S. have been the consensus building effort, the wide-ranging consultation procedures, the coalition-building activities, and following each of these, a willingness to revise and improve the objectives. The subtitle of the latest volume, Midcourse Review and 1995 Revisions, expresses the spirit and the substantive essence of the ongoing needs-assessment and planning-by-objectives process: it is self-correcting and constantly improving in its targeting of efforts where they are needed most, and in demonstrating health impact. The U.S. Healthy People Initiative is summarized in Appendix A, Abstract #1.

Other Countries' and World Health Organization Initiatives in Goals and Targets for Health

The World Health Organization adopted a policy of "Health for All by the Year 2000" in 1981 including the setting of goals and targets (World Health Organization, 1980), two years following the first U. S. Surgeon General's Report on Health Promotion and Disease Prevention. WHO's European Regional Office published its first set of Targets for Health for All in 1985, updated in 1991 based on broad consultation among European member states (World Health Organization Regional Office for Europe, 1985, 1991).

National goal setting for health has become a feature of the United Kingdom. The Health of the Nation initiative in England selects five health priority areas for action, sets national objectives and targets in the priority areas, outlines the actions needed to achieve the targets, proposes implementation strategies, and offers a framework for ongoing monitoring and review. In setting health targets an estimate of future trends is made and then combined with an assessment of the potential impact of interventions, programs and policies on the health of the population. Consequences for health are integrated into the policy formulation process. National goals and targets in England have been translated into regional and local goals, targets and action plans. Similar goal setting initiatives have been completed or are underway in Scotland, Northern Ireland and Wales. Additional information on the Health of the Nation strategy in England is provided in Appendix A, Abstract #2.

Australia published its first national goals and targets for population health in 1988 (Health Targets and Implementation Committee, 1988). Similar to the U.S. experience, goal setting at a national level in Australia influenced the development of health promotion goals and objectives at the state level. Healthy Victorians 2000 is an outgrowth of the Australian national strategy for health. Healthy Victorians 2000 is a joint initiative between the Victorian Health Promotion Foundation and the Victorian Department of Health and Community Services. The central aim of the initiative is to develop health promotion goals and targets for the population of the state of Victoria. Formulating health goals and targets is expected to increase health system accountability and to improve the ability of decision makers to assess the impact of programs upon the health of its people. To date, four of seven Australian States and Territories have instituted health promotion foundations whose central function is to set state goals and targets to monitor and assess the impact of programs on health. The Healthy Victorians 2000 initiative is presented in greater detail in Appendix A, Abstract #8.

The history and development of further goals and targets for the year 2000 and beyond in Australia is outlined by Nutbeam and his colleagues at the University of Sydney, giving due credit to the model provided by the U. S. objectives (Nutbeam, Wise, Bauman, Harris, & Leeder, 1993).

Reflecting on the Australian experience in setting "National Health Goals and Targets" (NHGT), Harris and Wise (1996), note the significant impact it has had on health policy by:

• focusing the attention of the health system on the outcomes which it achieves, rather than the services it provides;
• providing an information base against which it is possible to measure progress;
• highlighting the difference between the health outcome of different groups within the community; and
• providing a way of thinking about what may need to be done if the health of Australians is to be improved and health inequality reduced, including the integral importance of health literacy and skills and health promotion environments.

The National Goal and Target Setting strategy in Australia defines goals and targets for four national priority areas including heart disease, cancer, mental health and injury. Strategies to achieve targets have been developed at the state and territorial level and encompass all aspects of care including health promotion and prevention, early intervention, treatment, rehabilitation and extended care, and research. States and territories establish best practice and performance monitoring standards and systems which facilitate regular reporting to the national government. The goals and targets framework has been incorporated into the 1992 Medicare Agreement, which determines the health funding arrangements between the Commonwealth of Australia and the States and Territories and requires all states to participate in the goal setting process. Further description of the National Goal and Target Setting initiative in Australia is presented in Appendix A, Abstract #3.

The New Zealand experience presents a contrasting history in terms of the staying power of health policy and objectives promulgated at the national level (New Zealand Department of Health, 1989). The New Zealand Health Charter and model contract made Area Health Boards accountable to the national government for achievement of objectives to justify their budgets. The U. S. and Australian approaches have not been so directive to states and communities, giving them complete autonomy in deciding whether to adopt or adapt the national objectives in their own policies and plans. With the change in Government in 1990, New Zealand abandoned the goals and targets, whereas the process and the objectives themselves have survived several changes of government in the U. S. Four Presidents and six Secretaries of Health and Human Services have seen the U.S. health promotion and disease prevention objectives through their respective transitions without changing course. Although the White House made some notable modifications of the draft Healthy People 2000 objectives before they went to press (Green, 1992), the succession of governments has honoured the process and the goals and targets.

Conclusion

In Canada, health reform is occurring at an unprecedented rate. This reform is occurring within an evolving context shaped by macro social trends that include greater demand for community involvement, diminishing resources, an aging population, and recognition that health care delivery, alone, does not bring about health. We believe that such reform should not proceed in the absence of a conceptual or organizing framework that provides the requisite guideposts--population health goals. Such goals ought to be operationalized in concrete, measurable objectives.

Health impact assessment offers an innovative approach to ensuring that governments' program and policy initiatives align or are congruent with the agreed upon health goals. It suggests that policies and programs, regardless of the sectors from which they originate, should be assessed as to their influence on the health and quality of life of Canadians. Setting health objectives and targets, and conducting health impact assessments in relation to these goals and targets, will need to involve all sectors of government. The ideal role of the health sector is not only to act, but also to influence, enable, and mediate partnerships for intersectoral collaboration.

In the current climate of broadening definitions of health, the conceptual merging of health determinants and outcomes, intersectoral competition more than collaboration, and continuing debate over appropriate indicators of population health and an overarching policy framework, it is unlikely that the tools and strategies necessary for health impact assessment can be developed. The specification of goals and objectives with measurable targets can provide the requisite guideposts for health impact assessment. We believe that such an undertaking will set the stage and provide the necessary foundation for the next step in the health impact assessment process. Policy makers will then be required to elaborate and assess how proposed initiatives will contribute to progress toward the meeting of those goals, and how they will not hamper such progress.

Many, if not all, of the provinces have established goals, albeit some are not comprehensive and most have not set quantifiable targets. The federal government, despite repeated calls, has not set comprehensive national health goals to guide its policy cycle. Given this current situation, we offer several recommendations aimed at the Federal Department of Health, in keeping with the purpose of this study. We do not intend the use of federal health goals to be for purposes as broad as they have been used in other countries. Our purpose in these recommendations is for federal health goals and targets to be used exclusively for health impact assessment of federal programs and policies. These recommendations, for example, are not intended to suggest that the federal government should assume the leadership over the provinces in planning for and assessing the health of Canadians. This remains a shared provincial/territorial and federal responsibility, and many of our recommendations are appropriate considerations for the provincial/territorial governments. The federal government must, however, develop those mechanisms that will facilitate the assessment of the health impact of all federal initiatives.

[Return to Table of Contents]