National Prescribing Service: creating an implementation arm for national medicines policy

Abstract

Medicines make an essential contribution to the health of the community, but rapidly rising drug budgets have caused governments to seek ways of ensuring this expenditure results in value for money. The National Prescribing Service was established against this background to implement a quality use of medicines service as part of the National Medicines Policy. A range of programmes that attempt to use evidence-based strategies to deliver evidence-based messages have been established. These use multifaceted interventions, such as newsletters, prescriber feedback, clinical audit and educational visiting, that are provided both centrally, through the national office, and locally, through Divisions of General Practice. The work is underpinned by an evaluation strategy that incorporates strong qualitative elements as well as an emphasis on time-series analyses for changes in drug utilization. Some 80% of Australian general practitioners have voluntarily participated in activities such as educational visiting and clinical audit within the National Prescribing Service programmes. New programmes for the community and consumers will be coordinated with the work that has become well established within general practice.

Background

The appropriate use of medicines has much to offer consumers and the health system overall. Recent studies have indicated that problems with medication use are responsible for 2–3% of hospital admissions and 400 000 general practice consultations annually involve adverse drug events [1]. Probably a more common problem is not getting the best possible outcome from medicine use because of issues such as drug selection, incorporation of lifestyle measures and adherence to therapy.

It is important therefore to assess critically and respond to evidence that medicines use varies significantly across communities, often due to variations in prescribing practice for reasons other than diagnosis and patient factors. The doctor's age, his or her previous experiences and reliance on drug company information have been shown to be important determinants of an individual's prescribing choices [2, 3].

The direct costs of medicines to government and the wider community are also significant. In 1998 the cost of medicines to the Australian government was £1028 million (AUS$2623 million) compared with £1864 million ($4755 million) in 2003. These costs are rising at a faster rate than the actual volume of pharmaceuticals prescribed, reflecting the availability and prescription of new, more expensive drugs. There are questions about whether this rate of growth is sustainable and if general access to subsidized essential medicines, through the Pharmaceutical Benefits Scheme, could be compromised.

Policy environment

Australia is one of few developed countries with a comprehensive national medicines policy [4]. Developed by the Australian Pharmaceutical Advisory Council, the policy has been formulated in partnership by government, health educators, healthcare professionals, the medicines industry, healthcare consumers and the media. There are four objectives or ‘pillars’ to the policy which are interrelated and codependent (Box 1).

Australia's National Medicines Policy [6].

Aim of the policy is to meet medication and related service needs so that both optimal health outcomes and economic objectives are achieved.

 Objectives (frequently described as the ‘pillars’) of the policy are:

1. Timely access to the medicines needed at a cost individuals and the community can afford.

2. Medicines meeting appropriate standards of quality, safety and efficacy.

3. Maintaining a responsible and viable medicines industry.

4. Quality use of medicines.

Quality use of medicines (QUM) is a core element of this policy and is defined as medicine use that is judicious, appropriate, safe and effective. This section of the policy seeks to create cooperation between all groups that influence medication use, promoting the consumer as an active and central participant in the healthcare team.

In the mid 1990s, it was recognized that although there was research and seeding projects to demonstrate interventions that improved medicine use, there was no mechanism for taking successful projects to full national implementation. This gave rise to the notion of a national service that could do so [5]. An independently coordinated consultation with stakeholders, prior to its establishment, identified priorities for the service (Box 2). The consultation included written submissions, public meetings and key informant interviews with doctors, pharmacists, medical and pharmacy organizations, consumer organizations, academics and relevant government committees.

Priorities identified by Australian stakeholders for a National Prescribing Service.

Key roles

• Provide advice about quality prescribing and QUM

• Develop a national overview of quality prescribing policies, practices, initiatives and resources

• Implement best practice guidelines in prescribing

• Provide national leadership and coordination to ensure the availability of consistent, independent and objective medicines information for every prescriber.

Key strategies

1. Establish critical alliances to promote quality prescribing.

2. Provide advice to promote policy and systemic change.

3. Support medical educators to develop increased focus on quality prescribing within training programmes.

4. Promote collaboration between producers of independent medicines information.

5. Provide leadership to increase the level of computer-assisted prescribing in general practice.

6. Coordinate the development of a national medicines information phone line for both prescribers and consumers that is networked with existing medicine information services.

7. Provide leadership to develop incentives for (and remove disincentives from) high-quality prescribing in fee-for-sevice medical practice.

8. Coordinate a national academic detailing programme.

9. Provide national coordination of the developmment of mechanisms for locally relevant prescribing feedback.

10. Support provision of independent information and education for both consumers and prescribers.

The National Prescribing Service came into being in mid 1998 as an independent public company that was funded almost exclusively by the Federal government. The model of a company, operating at ‘arm’s length' from government, was adopted after representations from health professional and consumer groups argued that a government-run agency would not have adequate credibility. The level of funding has increased from approximately £2 million (AUS$5 million) per year in the first 4 years of operation to £7.8 million (AUS$20 million) in 2004. The Board of Directors is drawn from 37 peak member organizations representing health professionals, academia, industry, government and consumers.

Implementation of the service

The National Prescribing Service designed its prescribing intervention programmes to be multifaceted, using evidence-based strategies to deliver evidence-based messages at national and local levels. Initial implementation focused on general practice because of their low level of engagement in previous quality use of medicines work. The organization had a small core of five employees in 1998–99 and this has increased to over 55 people as programmes have grown and funding increased.

Selection of topics

Therapeutic topics were selected according to information needs identified by general practitioners (GPs) and other predetermined criteria. The criteria include factors such as: a new guideline has been published, there is evidence of variability in prescribing, GPs have expressed a need for additional information or important new management options have recently been introduced. An expert advisory group determined the scope, objectives and messages for each therapeutic topic after reviewing published literature and prescribing data. The advisory group identified important changes to national guidelines that required explanation and areas in which the evidence was either uncertain or controversial.

In the first 5 years a range of therapeutic topics have been covered, including: Helicobacter pylori eradication, nonsteroidal anti-inflammatory agents, antibiotics in upper respiratory tract infections, benzodiazepines, new guidelines for hypertension, and appropriate management of depression, asthma, diabetes, heart failure and pain. Typically, four to six topics have been covered each year and of these, half were the subject of both national and local interventions. The other half were delivered only as printed resources although usually including prescription analysis and feedback.

The evidence to formulate and support key messages for each topic was drawn together by staff with critical appraisal skills from a range of sources including original research. However, it was also considered important to ensure that these messages were consistent with and referred readers to other national sources of independent information such as the Therapeutic Guidelines series and Australian Medicines Handbook.

Interventions

A significant body of evidence has accrued concerning the factors which influence prescribing behaviours, including a range of behavioural, social and political determinants. The literature also provided evidence of interventions which had been shown to change behaviour when implemented in a sustained fashion at a national or local level (Table 1).

Table 1.

Summary of evidence base for prescribing interventions

Educational visiting or outreach	Has consistently been shown to improve prescribing [7] and in one study was shown to (academic detailing) improve health outcomes [8].
Clinical audit	Both computer-assisted and manual clinical audit have been shown to improve prescribing behaviour [9,10].
Mailed feedback of prescribing data	Mailed feedback alone was not found to change behaviour [11] but feedback accompanied by specific recommendations was more successful [12].
Interactive peer group meetings	An education programme involving audit feedback in peer groups saw larger changes in prescribing for acute situations and smaller changes for chronic therapy [13].
Problem-based learning using case  scenarios	A study of problem-based, face-to-face, small-group education on drug treatment found sustained improvement in prescribing of antibiotics [14].
Opinion leaders	Local opinion leaders were shown to accelerate adoption of effective treatments in general practice where best practice was clearly defined by evidence [15].
Written material	Variable effect alone but is additive with other interventions [16].
Guidelines	Locally adapted and implemented guidelines have been shown to change the prescribing behaviour of doctors [17].
Computer-assisted prompts	Well-designed reminders and alert systems coupled with an electronic health record have been shown to reduce initiation of inappropriate medication [18–20].

The National Prescribing Service drew on this in designing its key interventions of: educational outreach visits (academic detailing); clinical and self-audits; prescriber feedback with clear education messages; peer group meetings; and case studies that facilitated problem-based learning (PBL) for individuals or groups. Opinion leaders were used to deliver and endorse key messages when possible.

A mix of interventions has been implemented for each topic both nationally and locally. Local implementation occurred through 116 Divisions of General Practice, geographical groupings with an average of 145 GPs, via facilitators. The facilitators, about half of whom were pharmacists and the remainder comprising doctors, nurses and other health professionals, were employed by the Divisions of General Practice. National Prescribing Service provided workshops on the therapeutic messages and a background précis of the evidence, programme materials such as detailing aids, one-on-one support, peer networks, professional development and debriefing opportunities for facilitators.

Hence, a typical programme topic would commence with distribution of NPS News, followed within 2–3 months by direct-mail to GPs and other medical specialists of prescription analysis and feedback (including educational messages) on the drug topic of interest. During this period a clinical audit would be made available for those GPs who wished to examine their own practice in more detail. The larger range of local interventions in Divisions of General Practice would commence at the same time as the national programme roll-out, but an individual Division may have chosen to run that programme at any time over the subsequent 12–18-month period.

After 5 years over 14 000 individual GPs had actively participated in one of these programmes, equating to about 80% of GPs and indicating a good spread in all parts of the country. All GPs and specialists also receive written material and prescription analysis and feedback several times each year.

Incentives

Incentives were linked to participation but not to prescribing performance or patient outcomes. They included continuing education points required by professional registration bodies and a financial incentive of approximately £390 (AUS$1000) for GPs who participated in a set mix of at least three activities annually.

Evaluation

The evaluation plan used the priorities of major stakeholders and their information needs for decision-making and policy development to arrive at relevant questions. Mixed methodologies were identified for process, impact and outcome assessments. In addition, progress against the national Quality Use of Medicines indicators has been measured to show contribution to National Medicines Policy.

In the early years it was important to have a strong component of formative evaluation to inform programme development and case studies, key informant interviews and other qualitative methods formed a major part of the evaluation effort. More recently it has been necessary to look at the impact of NPS programmes including influences on drug utilization, knowledge and attitudes. This has raised significant challenges, including:

• dependence on an incomplete prescription dataset;

• lack of a ‘control’ group;

• broad national implementation that is likely to lead to widespread small change rather than more easily observed concentrated change;

• programme delivery over an extended period to suit local circumstances, which meant there was no single point of intervention, rather a cumulative period of intervention;

• attribution of any change to the NPS programme given that it was occurring alongside other activities.

The approach to this has been to use prescribing and drug utilization data from a range of sources, both longitudinal and cross-sectional, to build up a picture of change over time. Time-series analyses underpin much of the evaluation but have not been sufficient alone to describe the impact of the programmes.

Other priority areas

An essential parallel programme to influence community attitudes and demands on medicines use also forms part of National Prescribing Service's core work. A community-based social marketing campaign was conducted each winter from 2001 to 2003, entitled, Common Colds Need Common Sense, alongside the antibiotic programme for doctors and pharmacists. In 2003, community work expanded to mass media and community development strategies to encourage use of information by consumers that will support appropriate use of medicines. This is a major new programme which contributed a quarter of the government funding received in 2004.

Consistent with priorities from the original consultation, a web-enabled prescribing curriculum, using problem-based learning principles, has been collaboratively developed with all Australian medical schools for senior medical students (http://nps.unisa.edu.au). It is currently in use in the majority of Australian medical schools to varying degrees.

Two telephone drug information services, staffed by trained pharmacists, have been established for health professionals and consumers. Clinical and information models for computer-assisted pharmaceutical decision support for doctors and pharmacists are being built and decision prompts have been incorporated into some prescribing software.

Drug use evaluations in hospitals have been funded as multicentre projects to promote transfer of information across primary care and hospital sectors and build capacity for delivery of effective interventions.

Conclusion

National Prescribing Service has 5 years' experience delivering quality use of medicines programmes at both national and local levels in a coordinated but flexible manner. Having a dedicated implementation arm for the national policy has meant that the focus on therapeutics has not been subverted by other needs. There has been time for programmes and the organization to gain credibility among users, especially GPs and pharmacists. Growth in the first 5 years has been substantial and a coordinated attempt to cross professional and community boundaries promises to yield more sustained effects in the future.