Dan Bilsker

Evidence-Based Psychiatry

Objective To apply the evidence-based medicine paradigm to the domain of psychiatric practice and to bring out the consequences for psychiatry of this approach in order to foster the emergence of an evidence-based psychiatry. Method The basic assumptions of traditional and evidence-based paradigms, as delineated by the Evidence-Based Working Group, are used to structure an exploration of the evidence-based approach to psychiatry. Theoretical and practical issues are considered and an example of evidence-based decision making is given. Results An evidence-based approach to psychiatry is described as one that emphasizes the importance of systematic observation and the use of rules of evidence in hypothesis testing. It is suggested that psychiatrists using this approach will be in a position to provide superior patient care. Discussion The application of scientific method to psychiatric problems is discussed as the essence of an evidence-based approach. The common error of “scientism” is described. The authors identify advantages and limitations of an evidence-based approach to psychiatric practice and advocate a decision-making process that balances individualized clinical acumen (phronesis) and information derived from empirical study of groups of patients (techne).

Routine outcome measurement by mental health-care providers: is it worth doing?

In the USA, many health maintenance organisations and state governments require specific outcome measures to be used routinely by mental health care providers. 6,7 However, reliance on treatment providers as the source of data raises a serious methodological concern. Quite apart from pragmatic difficulties that arise if we impose the burden of outcome measurement on this community (lack of motivation for accurate data collection, 8 reallocation of scarce clinical time, cost of training, etc), the provider-based approach violates a basic precept of outcome measurement design. The precept can be stated thus: the practitioner who is responsible for delivery of a clinical intervention, and thus has a stake in the outcome of the intervention, is not in a position to objectively assess the outcome. The practitioner should be regarded as a biased observer of client status. Bias (systematic error) affects the validity of any datacollection process and much of scientific method is devoted to its reduction. Three categories of bias are especially relevant to provider based outcome measurement: selection bias (the provider puts more effort into obtaining baseline data for patients likely to achieve a positive outcome), attrition bias (the provider puts more effort into obtaining follow-up data for clients with positive outcomes), and detection bias (unrealistically positive ratings are assigned to clients who have completed treatment). 9 Researchers in other domains have shown that bias effects are subtle and often unconscious. 10 A treatment effectiveness study in which the treatment provider was also the outcome rater would almost certainly be discarded from systematic review as being compromised. It could be argued that, routine outcome measurement is not a controlled research method and thus cannot be expected to meet stringent methodological standards. But, the basis for such standards is logical: data obtained from potentially biased raters might

Health service patterns indicate potential benefit of supported self-management for depression in primary care

Objective: To examine health service delivery in a Canadian province (British Columbia) to consider how Canadian health care services might be developed to best address the large number of individuals with mildly to moderately severe depressive illnesses. Method: We used provincial administrative data to describe patterns of medical services provided to individuals suffering from depression during 3 different time periods (1991–1992, 1995–1996, and 2000–2001) and to determine the frequency with which depression patients receive treatment from primary care physicians and psychiatrists. We then used these findings to consider the feasibility and potential applicability of the various approaches that have been described to decrease the burden of disease related to depression. Results: In the fiscal year 1991–1992, the “treated prevalence” rate was 7.7%; in 1995–1996, it was 8.7%; and in 2000–2001, it was 9.5%. In each cohort over the 10-year period, the proportion of individuals who received a diagnosis of depression and who were then treated by primary care physicians alone (no psychiatric services were provided) remained constant at 92%. Conclusions: Supported self-management is identified as a promising intervention that could be integrated into primary health care within the context of the Canadian health care system. It constitutes a feasible and practical approach to enhance the role of family physicians in the delivery of services to individuals with milder forms of depression and promotes the active engagement of individuals in their recovery and in prevention of future episodes.

Managing Depression-Related Occupational Disability: A Pragmatic Approach

Objective: To identify the crucial issues that arise for psychiatrists and other physicians when dealing with occupational disability in their patients with depression and to suggest practical strategies for responding more effectively to the challenges of this aspect of patient functioning. Method: We identify fundamental concepts in the occupational disability domain and draw crucial distinctions. The wider context for occupational disability is articulated, involving the workplace environment and the disability insurance industry. Research with direct relevance to clinical decision making in this area is highlighted. We make pragmatic suggestions for effective management of occupational disability in patients with depression. Results: To successfully manage issues of occupational disability, psychiatrists and other physicians must understand the distinction between impairment and disability. To make this decision fairly and accurately, the adjudicator requires particular types of information from the physician, with requirements varying across short-term or long-term disability claims; failing to provide relevant information may cause substantial stress or financial harm to the patient. Balanced and collaborative decision making regarding whether and for how long to take work absence will greatly help to maintain occupational function in the long term. Realistic expectations and support of the patients sense of personal competence foster recovery of occupational function. Conclusion: Management of depression-related disability is challenging. Thoughtful evaluation of the patients functional status, careful response to the requirements of disability determination, and a focus on functional recovery yield substantial benefits.

Teaching Evidence-Based Practice in Mental Health

This article is reprinted from the journal Evidence-Based Mental Health. It presents the core virtues of the evidence-based paradigm (questioning of unfounded beliefs, rigorous scrutiny of methodology, critical appraisal of proposed treatments), and the authors’ experiences in teaching evidence-based practice to psychiatric residents. The aim of such training is to produce informed consumers of evidence, not researchers or methodologists. The lessons learned in psychiatric training have obvious applications to social work education.

Supported self-management: a simple, effective way to improve depression care.

Objectives: To introduce supported self-management (SSM) for depression, examine it through the use of a quality assessment framework, and show its potential for enhancing the Canadian health care system. Method: SSM is examined in terms of quality criteria: relevance, effectiveness, appropriateness, efficiency, safety, acceptability, and sustainability. Critical research is highlighted, and a case study is presented to illustrate the use of SSM with depressed patients. Results: SSM is defined by access to a self-management guide (workbook or website) plus encouragement and coaching by health care provider, family member, or other supporter. It has high relevance to depression care in Canada, high cost-effectiveness, high appropriateness for most people with depression, and high safety. Acceptability of this intervention is more problematic: many providers remain doubtful of its acceptability to their poorly motivated patients. Sustainability of SSM as a component of mental health care will require ongoing knowledge exchange among policy-makers, health care providers, and researchers. Conclusion: The introduction of SSM represents a unique opportunity to enhance the delivery of depression care in Canada. Actively engaging the distressed individual in changing depressive patterns can improve outcomes without mobilizing substantial new resources. Over time, we will learn more about making SSM compatible with constraints on provider time, increasing access to self-management tools, and evaluating the benefit to everyday clinical work.

Feasibility and impact of brief interventions for frequent cannabis users in Canada

Cannabis use is prevalent among young people, and frequent users are at an elevated risk for health problems. Availability and effectiveness of conventional treatment are limited, and brief interventions (BIs) may present viable alternatives. One hundred thirty-four young high-frequency cannabis users from among university students were randomized to either an oral (C-O; n = 25) or a written experimental cannabis BI (C-W; n = 47) intervention group, or to either an oral (H-O; n = 25) or written health BI (H-W; n = 37) control group. Three-month follow-up assessments based on repeated measures analysis of variance techniques found a decrease in the mean number of cannabis use days in the total sample (p = 0.024), reduced deep inhalation/breathholding use in the C-O group (p = 0.003), reduced driving after cannabis use in the C-W group (p = 0.02), and a significant reduction in deep inhalation/breathholding in the C-O group (p = 0.011) compared with controls. Feasibility and short-term impact of the BIs were demonstrated, yet more research is needed.

The Evolving Understanding of Major Depression Epidemiology : Implications for Practice and Policy

Objectives: Epidemiologic studies have confirmed that major depression (MD) is an extremely common condition, but also one that is associated with an unexpectedly broad spectrum of morbidity. It is no longer a tenable position to regard MD as being a simple indicator of treatment need, nor is a one-size-fits-all approach to treatment likely to be an effective guide to health care delivery. The objective of this commentary is to explore the implications of these new epidemiologic findings for policy and practice in Canada. Method: This paper is a selective review and commentary. Results: Whereas the acute and long-term treatment needs of a subset of individuals with MD have received much attention in the literature, the needs of other groups have not. A sizable proportion of individuals with episodes meeting the Diagnostic and Statistical Manual of Mental Disorders-fourth edition definition in community populations may not need the intensive treatment emphasized by current Canadian practice guidelines. The strategy of watchful waiting may have a role in primary care. On the policy front, guided and perhaps self-guided management strategies deserve greater emphasis than they have received. Stepped-care strategies are an appealing option, but how best to effectively implement these in the Canadian context is unclear. Conclusions: The spectrum of morbidity among individuals with MD in community populations is much wider than has been previously appreciated. The health system should respond with an appropriate spectrum of services, but many questions remain about how to facilitate this.

Factors in Delays in Discharge from Acute-Care Psychiatry

Objective: To determine if there are any potential opportunities for patients to be discharged earlier and to determine what factors are responsible for delays in discharge. Method: A survey was completed by clinical staff of all patients on the wards of 12 adult psychiatry units in the Greater Vancouver Regional Hospital District (GVRD) for a 1-day period. The survey included a modified Brief Psychiatric Rating Scale (BPRS) and the Discharge Readiness Inventory (DRI). A 1-month follow-up measured discharge and nondischarge outcomes. Results: Of the 327 patients surveyed, 42% were ready for discharge at the time of the assessment, and 37% of those who were ready were not discharged within 2 weeks. Delayed patients had significantly higher scores for disorientation, hallucinations, conceptual disorganization, and manifest psychopathology and significantly lower scores for Community Adjustment Potential (CAP) (P < 0.05). The most frequent reasons given for delays were ongoing medication adjustment, behaviour stabilization, and discharge planning. Patients who were delayed were more likely to need services, to need or be waitlisted for a residential placement, to be a client of the community-based mental health team that provides ongoing support to clients living in the community, to have a diagnosis of schizophrenia, and to have had no previous psychiatric hospital admission. Conclusion: The removal of all barriers to delays would reduce lengths of stay by 11% for this sample of patients. This would require a shift of about 42 out of 327 beds to an alternate level of care. These “earlier discharge patients” will need ongoing medication and behaviour monitoring through a variety of community services.

From evidence to conclusions in psychiatric research

Objective: To identify problems in the derivation of conclusions from evidence in psychiatry research. Method: The scientific model of falsificationism is described as determining the logical requirements for proving conclusions from research evidence. Common types of problematic conclusions are identified, and examples from the current research literature are given. Results: Poorly formed conclusions are based on inadequate hypotheses, ambiguously phrased, blind to negative findings, fallacious in logic, or neglectful of alternative explanations. Conclusions: The risks of accepting poorly formed conclusions are premature closure of scientific investigations, poor basis for decisions in psychiatric practice, and provision of misinformation to patients. It is recommended that practitioners be attentive to this aspect of critically appraising research.