Principles-based accreditation: the way forward?

Accreditation processes are widespread in health service delivery and health education, and are used for a variety of purposes.1 While the goals of accreditation processes vary, they typically encompass regulation (whether by governments or self-regulation), establishing barriers to entry (eg, access to funding programs, approval of medical schools, acceptance as a service provider) and providing evidence of quality assurance for individuals and the wider community (eg, accreditation of training practices and training posts). A recent example from health education is the national accreditation process introduced by General Practice Education and Training (GPET) to accredit training providers to deliver vocational education and training for general practitioners in Australia.2 Development of the GPET process prompted a detailed consideration of the conceptual basis for an accreditation system. GPET’s experience may help others who design or review accreditation processes, particularly in the way GPET drew on better established approaches to quality improvement in other sectors.

Designers of accreditation processes have many choices about the nature and conceptual basis of their processes. Paramount among these is whether the process is to be mainly or exclusively summative in nature, or formative as well (ie, generating improvements in the entity or process being accredited). A typical approach to summative accreditation involves developing (or adopting) a set of “standards”, and employing an auditing process to confirm that those standards have been met.1 The primary focus of this type of process is one of compliance. In this article, I refer to accreditation systems that are primarily concerned with ascertaining compliance as “rules-based” (RB) approaches. Certification against the International Organization for Standardization’s ISO 9001 (1994) quality assurance standard is an example,3 as are some other more specific accreditation systems that are used more widely in the health sector for accreditation of such facilities as laboratories and hospitals.1

An alternative to the RB approach to determining quality has emerged internationally in the form of a variety of “business excellence” models.4-7 These are normally used to improve the management and performance of organisations rather than for accreditation. They share a variety of common features, such as being based on a “framework” (rather than standards), being systemic in approach (rather than focusing on aspects of an organisation) and, particularly, by being based on a set of underlying principles. They owe much of their heritage to an approach to management often called “total quality management”.8 I refer to these types of models as “principles-based” (PB) approaches. They differ from RB models in being focused on continuing improvement and on change management rather than on compliance and, as such, appear to have played little part in influencing the design of compliance-oriented accreditation processes. One intrinsic feature of PB models that distinguishes them from RB models is that the amount of improvement and organisational change that might be sought through their application is open-ended.4

It is historically significant to note that sections of the wider “quality” community have, at times, been sharply divided over the merits of the two approaches.3 My intention here is not to advocate for PB approaches per se, but to encourage examination of their potential benefits for accreditation in the health care and health education sectors.

Some important characteristics that might be considered in designing an accreditation process are outlined in Box 1, in which I contrast the differences between RB and PB approaches, based on my survey of and experience with common PB and RB models. The GPET accreditation process2 was designed with reference to these characteristics. Despite PB approaches not being traditionally associated with accreditation, their intent was instructive, and resulted in the GPET process being strongly PB. Others who are operating in different contexts may come to a different view.

Empirical and scholarly evidence suggests that there are differences in the quality improvements that are likely to be achieved with PB and RB approaches (Box 2).4,9 Because PB approaches are systemic and more comprehensive in scope, the initial improvements are likely to be gained more slowly for any given level of resourcing. On the other hand, the response is likely to be enduring4 because the PB model is intrinsically open-ended. In contrast, the RB approach is likely to have more limited and realisable quality goals that may be reached (ie, complied with) more quickly, but once those goals are reached, the motivation for continuing improvement may be lost without additional intervention or a change in “standard”. In some cases there may even be a regression.

It is interesting to see that the World Federation for Medical Education in its “Global standards for quality improvement” has recognised the limitations of what might be regarded as an RB approach by introducing a duality of a “basic standard” and a “standard for quality improvement”.10 While this approach seeks to solve the problem of lack of motivation for continuing improvements that is an inherent risk in pure RB models, it does not necessarily incorporate the systemic focus and underlying principles that are intrinsic to PB approaches. Similarly, the ISO 9001 (1994) standard has been revised to focus more on improvement in the 2000 version,3 but also without explicit underlying principles to guide its application.

The benefits that GPET has identified through independent reviews of its PB process include the universal use of formative feedback to guide improvements by providers, satisfaction with the process reported by providers and peer reviewers alike, evidence that quality improvement is being incorporated as an integral part of providers’ management, and a climate of mutual respect surrounding accreditations. In adopting a PB approach that provides comprehensive formative feedback, GPET has recognised that the process requires more resources than an RB process, but this is seen as an investment in system-wide improvement across all providers, which is hoped to be enduring. Enduring improvements is a goal shared by other Australian accreditation systems, such as professional colleges, medical schools and intern programs, all of which demonstrate PB characteristics in this respect.

In the light of successful experiences with PB approaches to accreditation and the evidence of success with PB models in other contexts, I commend PB models as a framework for designing and reviewing accreditation processes in health care and health education.

2 Rates of quality improvement with principles-based versus rules-based approaches