Abstract
At the outset it is useful to define what is meant by the term policy. This is not easy as the term has a certain conceptual fuzziness and has been used to describe government action, government programmes and/or the political process [1]. In this article the following definition is used: ‘public policy is deciding at any time and place what objectives and substantive measures should be chosen in order to deal with a particular problem, issue or innovation’ [2].
In free markets consumers and suppliers are to be left alone to interact and balance supply and demand for services. Much activity in society takes place in such markets, especially with increasing deregulation and globalization [3]. It is outside the scope of this paper to discuss market failure in health, however, it is generally accepted that governments need to intervene to provide certain services and regulate the market [4]. This intervention occurs in the form of specific policy action. Private sector organizations as well as governments have policies to guide their operations, however, this paper limits its discussion to government policy.
In Australia, responsibility for the health system is shared between the Commonwealth, State and Territory governments and the private sector. Section 51 of the Constitution does not specifically allocate a role for the Commonwealth in health apart from quarantine powers to prevent disease entering the country. However, it has gradually assumed a prominent role following the centralization of income taxation collection which occurred during World War II. The fiscal dominance of the Commonwealth has resulted in it raising more revenue than it spends in directly delivering services, with the States spending more on their services than they raise in state taxes. This is referred to as vertical fiscal imbalance. It is therefore necessary for the Commonwealth to transfer money to the states and territories for areas such as health, allowing the Commonwealth to influence or even dictate the direction of reforms. The Australian Health Care Agreements are an example of such a transfer and understanding these arrangements is important when considering national health policy.
Commonwealth and State governments fund the public health system and contribute most of the funding for private non-inpatient services. With the recent decision to subsidize private health insurance, the Commonwealth government now has an even greater stake in the provision of care in what is supposed to be the private sector. In deciding how to spend this money, politicians are wary of the disproportionate influence of vested interests and know that modern health programmes can offer more than the health budget will be able to pay for.
This paper identifies from the public policy and health policy literature [3–8] a five-step process of policy development and implementation. Examples from health and mental health are used to highlight how issues reach the threshold for being considered for policy attention and the factors which will shape the policy and its implementation. Understanding this process should assist mental health researchers to use the outcomes of their research to exert greater influence on the policy process. Research is understood in this paper in the broad sense of the generation or identification of new knowledge, and is not limited to biomedical research.
Problem identification
A policy is usually developed in response to a real or perceived problem. The identification of this problem occurs within a larger social, cultural, historical and political environment. The intent of action in any of these broad areas may be unrelated to health but still have an impact on it. For example, a political decision to address a macroeconomic problem, for example to reduce a budget deficit, will impact on the spending departments (e.g. health and social security). In the Australian States and Territories, health and education account for nearly half of the government outlays and will be confronted with budget cuts to reign in the deficit.
Given that hundreds of variables impact on the health of a population and the functioning of any health system, how does one or more of these become identified as needing attention? Roberts and colleagues [8] highlight two mechanisms which determine this selective attention. First, cultural norms and social attitudes provide a set of filters that selectively focus or divert public attention. Some aspects of the functioning of the health sector are taboo and not to be discussed in public. This was the case with the plight of mentally ill individuals in large psychiatric hospitals until the 1960s. A more recent example is doctor-assisted suicide of patients with terminal illness. With a change in societal attitudes, previously neglected areas such as these can become a topical social issue and a problem needing attention.
The second mechanism relates to the role of issue entrepreneurs. These individuals or groups act on issues which emerge from the filters to articulate (or create the perception of) a public problem. The entrepreneurs who take up these issues may be politicians, bureaucrats or particular interest groups within the community. All seek to put a particular issue on the public's radar screen. Sudden infant death syndrome is an example of success in doing this. The motives of issue entrepreneurs can be varied. Some act on the basis of their conviction of the need to address a particular area. A coalition with such convictions, including the Australian National Association for Mental Health and the Royal Australian and New Zealand College of Psychiatrists, acted as issue entrepreneurs between 1984 and 1989 to build the momentum for the National Mental Health Strategy.
Motives of political players can be similar but can also be political in the electoral sense. This may not be recognized by others involved in the debate. The willingness to adopt and fund policies is influenced by the electoral cycle, with unpopular policies avoided, and potentially popular policies introduced, in an election year. Political and bureaucratic competition can emerge between individual politicians or branches of government. The competition between government departments and between States and the Commonwealth often reflect political rivalry when different political parties hold government in different jurisdictions. Alternatively it can reflect the long-standing Commonwealth/State power struggle for influence in the health system.
To achieve a national (as opposed to a Commonwealth) policy requires at a minimum the formal agreement of the Commonwealth, State and Territory governments. Prior to the adoption of the National Mental Health Policy [9] the Commonwealth considered mental health to be the responsibility of State and Territory governments and was not supportive of a national policy. A critical piece of research helped change this. A report prepared in 1988 for the Commonwealth by Peter Eisen and Kevin Wolfenden demonstrated that Commonwealth expenditure for people with mentally illness was far more than the total cost of all State mental health expenditure. This was confirmed by the National Health Strategy [10] which showed the extent of the Commonwealth expenditure on people with mental illness was $2.58 billion in 1991/1992 (e.g. on sickness benefits, pensions, the Medicare Benefits and Pharmaceutical Benefits Schemes). The combined expenditure of the public mental health services run by States and Territories was only $871 million in that year.
The research went further and highlighted the inconsistencies in the existing social policy framework as it applied to mental health and psychiatric disability services. For example, the Commonwealth spent $1.45 billion in 1991/1992 on income security for people with mental illness and psychiatric disability, but at the same time excluded them from the programmes designed to decrease dependence on welfare payments and help disabled people back into the workforce (e.g. in the Commonwealth Rehabilitation Service). Research such as this became tools for the issue entrepreneurs and helped establish the need for a national approach to mental health in which the Commonwealth became a key player in terms of funding and national leadership.
The interaction of social filters and issue entrepreneurship tends to produce certain patterns in the process of problem definition. An issue/attention cycle develops in which issues fluctuate as a matter of public attention. One reason for this lies in the incentives and behaviour of modern mass media. The revenues of print, radio and television media all depend on circulation or audience. Media outlets believe that readers, listeners and viewers consume news, in part, for its entertainment value. Stories become stale after a time and media outlets have incentives to move to new, interesting topics to attract or retain viewers and readers. Politicians have similar incentives. Voters tend to want new ideas and not the same old programme.
Media in market economies tends to focus on scandal and/or personalities, which at its worst is called tabloid journalism. There is also a herd pattern or pack mentality found among the media. Competitive media outlets feel compelled to cover a story simply because other outlets are doing so. This obviously reinforces the issue/attention cycle since massive coverage results in more impact but also makes the stories ‘wear out’ faster. Issue entrepreneurs have to be patient as their issue cycles in and out of public attention. However this attention cycle can also modify social attitudes and norms thereby influencing the social filters. In recent years there has been concern about the plight, and alternatively the dangerousness, of mentally ill people living in the community. In Australia this has not escalated to a sustained demand for reinstitutionalization.
What is the role for the scientific community in problem identification? The outcomes of scientific research, provided strategically and in a way that promotes ready consumption, can focus selective attention in problem definition with a considerable impact. It can determine whether the issue matters and what can be done to address the problem. The Global Burden of Disease study [11] is an example of how quantitative summary measures of the research can demonstrate whether an issue matters. Using the disability adjusted life year (DALY) mental disorders were found to be responsible for nearly 30 per cent of the non-fatal disease burden [12]. The controversial technical issues surrounding the DALY did not affect the impact of this data and the message received by policy makers is that mental disorders are much more burdensome than previously thought.
Likewise epidemiological studies in the area of mental health [13] have demonstrated that the majority of people with mental health problems do not access health practitioners. Seventy-five per cent of those who do, access general practitioners. This has focused attention on the need to increase access to services and, given the limited supply of providers, ensure the cost-effectiveness of those treatments being provided to individuals who access care.
There is a further role for scientific data. Issue entrepreneurs are usually seen as less than objective with regard to the issue they are advocating for. Governments and the media often wish to gain an opinion independent of the issue entrepreneur, and turn to scientific experts for an assessment of whether a problem exists, its nature and extent. Research is immeasurably strengthened by being seen as independent [14]. However the policy debate can and does proceed in a data vacuum and scientific information is only one input. But without objective information myth, prejudice and ideology will dominate the debate.
Development of a policy option
Once the problem has been identified there is a need to find a way to address it. On occasions the option proposed to fix the problem is predetermined. This occurs when a political party is elected with an explicit plank of its platform to take certain action in an area. However, usually any political party's pre-election mental health commitments are light on substance. Once elected it must find solutions to the problems confronting government.
How is a response in the form of policy developed? Governments often start with local or international experience in similar areas. When the problem is at the systems level, for example the need to increase equity or efficiency, health is considered an industry and the solutions which are applied to other industries are often applied to the health sector. In health and mental health services the implementation of routine outcome measures is considered as essential in order to provide customer feedback on services. The first National Mental Health Plan [15] identified consumer outcome measures for this purpose [16–18]. Competition among health service providers can also be promoted, as this is believed to bring greater efficiencies. The drive to achieve ‘value for money’, and not only cost containment, is an explicit importation of market concepts into the Second National Mental Health Plan [19].
On occasion a local or international scientific discovery provides an option so clearly revolutionary that the argument for adopting the findings of the research as policy is overwhelming. However, most research findings add incrementally to the knowledge base. Their application to policy and services is not always apparent. However, even when research repeatedly demonstrates that a particular action is beneficial, it can take many years for this to find its way into policy. It was not until 1970, 20 years after its efficacy was established that lithium was approved by the Federal Drug Administration in the USA [20]. Despite the well-established efficacy of psychoeducation and rehabilitation in improving outcomes in schizophrenia, these components of treatment are not yet widespread in Australia [21].
It is necessary to build the political will and institutional capacity to make a success of a particular option whether it is being ‘imported’ or developed locally. Political will depends, in part, on the likely public support for any particular option and the vulnerability to which politicians are exposed given the electoral cycle. Stakeholder analysis is used to determine the position of relevant groups and individuals both inside and outside government who are likely to influence the policy choice and the success of its implementation. Policy development threatens change and will therefore be resisted by groups with certain vested interests.
Stakeholder analysis combines two distinct groups of analysis within political science. One is interest group analysis and the second group is bureaucratic analysis. The first consists of understanding those groups in the community that are advocating for government action. The second is focused on competition among agencies and organizations within government. Stakeholder analysis allows the opportunity for research data to be collated and analysed to address the options being considered. It may even allow time for new research to be commissioned. When an option has to be chosen in a short time frame there is little or no time for stakeholder analysis and only a small window of opportunity (often a few months) for research to influence policy [22]. Here it is a case of advocating with information already available. Following the 1997 Port Arthur shootings there were inaccurate media reports concerning Martin Bryant's mental state. The immediate debate in the bureaucracy and media turned on the issue of whether the problem needing policy attention was that of dangerous mentally ill individuals in the community or gun control. Research data available at the right time helped swing the debate. Philip Alpers, a New Zealand gun policy researcher, had presented at a conference in Melbourne just before the Port Arthur tragedy. His paper reported that the majority of mass civil homicides are not committed by persons with known mental illness discharged from hospital and that most held a licence for their firearm [23]. Media coverage and government consideration of this data helped successfully turn the debate toward the option of stricter licensing of firearms.
Overriding the stakeholder analysis is the question of whether a given option adopted as a policy position is politically feasible. Science and politics can clash when the two come to irreconcilable conclusions. The debate about needle exchange programmes is an example [24]. The scientific data supported the use of such programmes but the political view was that they would be unpopular with the electorate. The likelihood of getting a policy adopted depends on the situation, skill and commitment of its advocates in the political environment.
Political decision
The adoption of a particular option as policy is a political decision. The factors bearing on this are the relative power of each player in the political game, the positions taken by them and the intensity of commitment for or against the policy. Within the political sphere, these players include not only the Health Minister, but also his or her staff, other key Ministers (especially the Minister for Finance and/or the Treasurer) and their staff and the Premier or Prime Minister and his or her social policy or health policy adviser [7]. Senior government bureaucrats in each of these departments are often exceptionally influential through the advice they give to the Minister's office. For each of these individuals and groups an analysis can be made of their interests, power and influence. Successfully negotiating a coalition of support usually involves bargaining and trade-offs. Throughout the process of negotiation, the content of a policy will be modified, as compromise is usually necessary to achieve consensus.
Decisions on health policy are characterized by levels of complexity, which need to be specifically considered. Health is technically complex in both a clinical and systems sense. Service policy often needs to change multiple parts of the system at the same time, with failure in one area undermining the success of change in another. The failure of community accommodation for patients discharged from psychiatric hospitals (often the responsibility of the Housing Department) can seriously undermine a policy of expanded community based mental health care and the closure of hospital beds [25].
Another component of this complexity is the concentrated costs on groups renowned for being well organized and influential, for example the medical profession and the pharmaceutical industry. The beneficiaries (civil society and consumers) are less informed and often less organized. This can create significant political obstacles to reform. The development of national and State consumer advisory groups under the National Mental Health Strategy was in part to help address this imbalance [26]. The benefits of reform are often dispersed and this, combined with the lack of political influence of some recipients, makes it difficult to create significant political support for reform which will benefit them. The closure of a ward in a psychiatric hospital with the savings going to community-based services is likely to bring a much more vigorous response from the staff who are to be affected in the hospital than the potential beneficiaries of the community services.
To build support for the policy it is useful to align it with symbols which are seen as ideologically unchallengeable. This can generate wide public support. Community mental health care was aligned with ‘freedom’. Health is said to be a ‘human right’ and ‘equity’ is sought in access to health care. All align a health issue with the universally supported symbols of freedom, human rights and equality.
When an impasse is reached, reframing of the policy can often overcome differences between groups. The policy position of allowing patients to be involuntarily detained in private hospitals was initially seen as unsupportable because of the perception that the private sector would profit from patients being treated in hospital against their will. This was reframed to state that a person with mentally illness who had paid for their private health insurance should be allowed to remain with the psychiatrist (and hospital) of their choice even when (or especially when) their illness was at its worst. The policy became politically acceptable and was adopted. Perceptions of reform are matters of values as well as facts. Political decision making is about emotion as well as data.
Policy implementation
It often seems as if the hard work is over once a political decision has been made to adopt a particular policy. However in reality the hard work has only just begun. Many policies are developed and never implemented. Governments are not omnipotent and in reality have only four main levers available in formulating and implementing an option [5, 8]. These are the financing system which determines what resources are available, where these resources come from and who has access to them; the payment system which determines on what terms these resources are made available to individuals and organizations; the organization of the health system in both the distribution of services and how they respond to consumer demands; and the regulatory system which imposes a set of constraints on services, for example how providers are trained and recognized, how the medical and pharmaceutical benefit schemes operate and the regulations which cover the private health insurance and private hospital systems.
Assembling these levers into an implementation plan (in the case of Australia, the first and Second National Mental Health Plans) requires the ongoing support of the professional and community organizations which came together to ensure the policy was adopted in the first place. Alignment with other organizations consolidates support and enhances implementation. Such an alignment has occurred between governments and the Strategic Planning Group for Private Psychiatric Services. The private sector had not been involved in the early implementation of the National Mental Health Policy or Plans. The more broadly representative and independent Mental Health Council of Australia has strengthened the role of non-bureaucrats in the ownership and implementation of the Second National Mental Health Plan.
The financing lever was used to help drive the implementation of the National Mental Health Policy. The first National Mental Health Plan [15] detailed how the Policy would be implemented and initially came with $135 million (over 5 years) of specific Commonwealth funding. This was increased a year later as discussed below. The provision of this funding through Schedule F of the 1993–1998 Commonwealth/State Medicare Agreements, was made to States on the basis that they not only implement the Policy but maintain their own financial effort in mental health care. Importantly, there was a requirement on all parties to provide data on their progress in implementing the Policy. The data provided under the Medicare Agreements was drawn from a setof 49 national policy indicators adopted by the Australian Health Ministers Advisory Council (AHMAC) Mental Health Working Group in July 1993 [27]. The indicators were first collected for the 1992/1993 years, published in March 1994, and provided a baseline of activity for future years. Published annually since that year in a National Mental Health Report, the information became a public benchmark for measuring progress and was used by community groups and the media.
During the life of the first National Mental Health Plan the political party in power in all governments (with the exception of the Northern Territory) changed. However the national reform agenda continued essentially unchanged. In part this was because substantial public expectation for change had been created and progress toward meeting this was being reported on annually. Successful implementation was rewarded with positive community reaction. Poor progress was met with demands to do better. The other reason for sustained implementation was because the Commonwealth funding for reform was locked into the 5-year Commonwealth/State (Medicare) financing agreement. Once both major political parties had, either at State or Commonwealth levels, supported the National Mental Health Policy it was seen to enjoy bipartisan political support.
Key players within and outside government do not often remain in a position to see the policy implementation through to conclusion. Institutional structures such as the AHMAC National Mental Health Working Group met regularly even though the membership changed during the life of the policy. Communicating both the successes and failures of the policy implementation enhances credibility, institutionalizes the process and legitimizes action to modify the implementation [28].
Finally the implementation of policy is not linear and rarely occurs according to plan. Flexibility in adapting to emerging forces which can create barriers and opportunities is essential. The implementation of the National Mental Health Plan provided a classic example of how an unplanned enhancing of the process can occur. The negotiations on, and drafting of, the National Mental Health Policy commenced in 1989 and concluded with the adoption of the Policy by all Health Ministers in April 1992. A parallel process at the time was the Human Rights and Equal Opportunities Commission (HREOC) Inquiry into the Rights of Persons with Mental Illness carried out by Commissioner Brian Burdekinbetween 1991 and 1993 [29]. The 1992 Federal Budget, with Brian Howe as Health Minister, contained $135 million for the National Mental Health Plan. When the HREOC report was released in 1993 with extensive media coverage, the Federal Health Minister was Graeme Richardson. Senator Richardson announced additional Commonwealth funding in response to the deluge of adverse media coverage surrounding the issues Commissioner Burdekin raised. This amounted to a further $134 million over the remaining 4 years of the National Plan, effectively doubling the budget for the National Mental Health Strategy.
Evaluation
Evaluation is often the most neglected area of policy development and implementation. By the time an evaluation of a policy is due to be conducted, most of the government officials originally involved in its development will have departed. During the implementation, organizations who support the policy may have wilted (or grown), and those who oppose it may be stronger. The environment will be different. For an evaluation to have credibility it must be as transparent and independent as possible. It was difficult to find individuals within Australia who had not been part of or impacted upon by the National Mental Health Strategy. Several individuals from outside the country were recruited to help ensure objectivity and this also provided an international perspective on the strengths and weaknesses of the policy and its implementation [30].
Feeding the evaluation into a revision of the policy completes the cycle of policy development and implementation. While the National Mental Health Strategy demonstrated success in promoting health sector reform, an objective evaluation demonstrated insufficient attention to certain specific groups (for example, child and adolescent populations, aged care, forensic groups). Issues such as prevention, early intervention and primary health care were also identified as relatively neglected. These have been incorporated into the Second National Mental Health Plan [19].
However in revising the policy it is important to resist the temptation to neglect those areas that have been successful and let the pendulum swing too far in the direction of areas that have been less well addressed. In Australia we run the risk of this under the Second Plan if we neglect service reform for those with established mental illness while trying to enhance mental health promotion and illness prevention. While we broaden the scope we must ensure that scarce resources remain focused on key areas where the best outcomes can be achieved. If we spread our effort and resources too widely, effectiveness will be diluted with an eventual backlash from the core mental health constituency and a loss of credibility for the Policy.
Conclusion
Policy development, adoption and implementation is often seen as a political and bureaucratic exercise. However research can impact on all these levels by providing options which are scientifically validated and data which allow decisions to be made more on the basis of fact and less on the basis of political expediency and ideology. In doing this it is necessary for the information to be available and communicated to the right people at the right time. It must be communicated in a way which can be assimilated by individuals unfamiliar with the technical detail and yet able to understand the essence of the argument or the information. It is often necessary to reframe the information or reduce it to what may seem to the scientist to be overly simplistic or inexact. Scientists must be able to tolerate imperfection in the use of their data as the political process inevitably involves oversimplification, compromise and trade-off.
The debate about whether science and clinical practice can be more proactive is important [31]. In their recent book on clinical futures, Marinker and Peckham [32] argue that anticipated breakthroughs in science and technology, and their potential impact on prevention and treatment of disease, need to be incorporated into thinking about health policy. They rightly note that political consideration of economic and social futures dominate the health debate to the exclusion of what is coming down the health research pipeline. The likely scale of medical advances, including in psychiatry, presents opportunities for researchers and clinicians to play a much more proactive role in health policy. To do this they will have to better understand the complex environment of policy development and implementation and be prepared to enter an arena many see as unfamiliar and disturbingly irrational.