Best Practices - Concurrent Mental Health and Substance Use Disorders

The Approach to Defining Best Practice for this Project

Defining best practice guidelines is a burgeoning area within health care. Best practices have been defined as "systematically developed statements to assist practitioner and patient decisions about appropriate care for specific clinical circumstances".¹⁶¹ These guidelines are typically directed at the service delivery level; that is they offer guidance to clinicians, therapists, social workers, etc. on the most effective, or cost-effective, services that can be delivered directly to health care consumers. There are, however, some examples of guideline development that explicitly aim for recommendations of best practice for health service organizations (e.g., performance measurement mechanisms and processes), and systems of care (e.g., policy development, planning and funding mechanisms and processes).¹⁶²

Evidence-Based Versus Expert Consensus

There are two main models for developing best practices^{163, 164} - the scientific evidence-based model and the expert consensus model. In the scientific evidence-based model, best practice guidelines are extrapolated from a comprehensive literature review.

Several concerns have been raised about this strictly research-based approach.^165,166 These include:

• reliance on a body of literature that is subject to pervasive publication bias;
• lack of attention to multicultural issues;
• limitation on generalizability; and
• over reliance on the "gold-standard" research design, the randomized control experiment.

The expert-consensus model is typically conceived of as an adjunct to the scientific evidence-based model whereby the opinions of experts are used to fill the gaps in the scientific literature. The four main methods of consensus formation are:

• informal processes;
• formal consensus development conferences;
• nominal groups;
• Delphi methods.^163,164,167

Criticisms of this approach often centre on the fact that the process of selecting experts is often a reflection of professional hierarchies which may result in no more than a group guess.¹⁶⁸

It is often necessary to develop treatment recommendations on the basis of less than conclusive research literature, and to use expert opinion to advantage. Thus, a mixed methodology is often used that takes the best information from published research studies and combines some form of expert review and consensus development.¹⁶⁵ The specific features of the combined methodology is often driven by the available budget.

Approach

The guidelines developed in the present project are based on:

• an extensive review of relevant literature;
• a synthesis of this literature according to the professional judgment and experience of the study team, including what the literature does not tell us, and a summary of other issues that must be taken into account in extrapolating from the published studies;
• review and advice of an expert panel concerning our general approach and results of our knowledge synthesis;
• a review of preliminary recommendations by key stakeholders in the mental health and substance abuse systems as obtained by a key informant survey;
• focus groups with consumers who have experienced concurrent disorders in order to validate our synthesis of research and expert opinion with their needs and lived experience;
• further synthesis of the above by the research team and additional review and input from the expert panel;
• in addition to ongoing process management the draft report was reviewed by Health Canada to ensure consistency with the study objectives and obtain additional input from the various provincial and territorial representatives on the Health Canada Working Group; and
• preparation of final recommendations.

It is important to be clear that the guidelines emanating from this project are not based on:

• a formal rating of the strength of the research evidence (however, highest priority was given to research from experimental or quasi-experimental studies). Based on this research evidence we also distinguish, where possible, between the 'best' and the 'most promising' interventions for a given combination of concurrent disorders; or
• a consensus among the members of the expert panel (however, areas of strong agreement as well as dissenting opinion have been noted in an anonymous fashion).

These methodological decisions are based largely on budgetary constraints as well as the current state of the literature in the area of concurrent disorders. Future attempts to update these best practice guidelines may have a larger literature to draw upon in some areas, and therefore will be more suited to these methodological improvements over our work.