Evidence-Based Psychiatry in clinical care and community mental health programs and policies - Misconceptions, achievements and future directions*

Milos JENICEK

MD, PhD, FRCPC, Professor (McMaster), Professor Emeritus (Montr�al) and Adjunct Professor (McGill). 
E-mail: jenicekm@mcmaster.ca
Postal address: 5922, Fifth Line, RR1, Rockwood, Ontario, N0B 2K0, Canada. Tel/Fax: (519) 856-1324.

Summary:

Since the appearance of Evidence-Based Medicine (EBM) ten years ago, Evidence-Based Psychiatry (EBP) has kept pace with this movement, facing its critics and searching for the best direction to take in the future. Some misconceptions about EBP stem from a rather narrow view of EBM and limited applications in mental health given the challenging specifics of mental disease.

However, clinical EBP shares the same clinical-epidemiological rules as other specialties and requires multiple adaptations of soft data, interface of qualitative and quantitative methodologies and decision-making in situations where patient involvement is more passive. Community health programs and health policies now have more stringent and expanded epidemiological (and other) criteria and requirements to which EBP must adhere despite the additional challenges inherent to the mental health field.

Better retrieval of evidence, expanded views and evaluation criteria in the assessment of the quality of evidence beyond the cause-effect relationship, uses of evidence, and evaluation of the effectiveness of EBP practice itself are all domains where practices are still uneven in extent and content and where a better balance should be achieved in the nearest of futures.

EBP does not dehumanize psychiatry. On the contrary, EBP makes it more pragmatic and expands its practices and research within the Hippocratic oath obligations and ideals. Reciprocally, EBM itself should benefit from the EBP experience in the exciting, but very challenging area of mental disease and health.��

There will never be enough clinical experience.

There will always be more than enough patients in distress

and a need for effective help.

In fact, is that not the reason

why we take the Hippocratic oath?

One of the exploratory models in psychiatry, the theme of this Congress, is undoubtedly Evidence-Based Psychiatry (EBP) and its translation into psychiatric research, clinical practice and community mental health programs and policies. Psychiatry within the Evidence-Based Medicine (EBM) movement, some basic misconceptions about Evidence-Based Psychiatry, its current major achievements, and directions for the future merit the following comments.

1.�������� Evidence-Based Medicine, Public Health and Psychiatry

Medicine has always been evidence-based. Only the meaning and paradigm of evidence has changed over time from a simple belief and conviction to empirical experience. The Salerno school of medicine was perhaps the first one of its kind to be founded on empiricism. Following evidence was the understanding of biological mechanisms of disease, health and cure. Ultimately, evidence focused on the nature of scientific practice and on experience-generated information and ensuing clinical choices and decisions. This type of EBM is about ten years young. Praise and reservations about it equally abound^1-6. In the spirit of EBM and Evidence-Based Public Health (EBPH)⁷, EBP is the integration of best research evidence with clinical expertise and patient and community values in the clinical practice of psychiatry, research and community health programs as well as in mental health policies. Problem definition, evidence retrieval, evidence evaluation, its uses, and evaluation of such uses are its fundamental steps in all the above-mentioned domains of interest. In Canada, EBP is a current endpoint of the past 50 years of experience and transition from inferential psychiatry⁸.

Throughout the past decade, teaching and practice of EBM acquired its basic methodology^9-11, and managed to integrate in its process not only clinical medicine, epidemiology and biostatistics, but also qualitative research, research integration, decision-making, health economics and most recently logic and critical thinking in health sciences as a distinct methodological domain¹². EBM is now applied to an increasing number of clinical specialties such as public health^7,13, family medicine¹⁴, pediatrics¹⁵, cardiology¹⁶, critical care¹⁷, or health promotion¹⁸ to name just a few from an ever increasing number. Psychiatry has its anchor in the online periodical Evidence-Based Mental Health¹⁹, companion to its Evidence-Based Medicine²⁰ counterpart. Academic departments teach EBP in increasing numbers; McMaster University and University of British Columbia are perhaps among the most active in Canada in this field. Evidence-based recommendations for programs in mental health appear in Canadian and US preventive and clinical guides as well^21-23.

2.�������� Some misconceptions and points to ponder in EBP

Of all medical specialties, psychiatry is probably the most challenging in which to practice Evidence-Based Medicine. Data are soft. Patient participation in the entire medical process is limited. Many third parties (such as family or civic bodies) are involved in decision-making and care. The over-reliance on specific clinical experience (phronesis) often prevails over techne (production of and uses of evidence from observational, experimental and other scientific sources²⁴). Some of the ensuing misconceptions and points to ponder are worthy of mention.

Unclear questions raised

Unclear research questions undermine their value as evidence. �The role of tricyclic and clinically similar compounds in the treatment of mood disorders� is certainly less operational as evidence than the same question when focused on an �intervention-outcomes-population setting and condition of interest� package: �Does the bupropion therapy diminish the yearly frequency and severity of clinically important depression episodes in older patients suffering from a bipolar affective disorder?� It is obvious that we need a greater number of such operational questions and problem formulations.

Evidence itself

The domain of evidence is much wider than a cause-effect relationship between beneficial factors (treatment) and their effect (patient improvement). Criteria and grading of evidence were most fully developed in the area of treatment where they have also been applied most often. In fact, criteria and treatment somehow overshadowed other elements such as risk, diagnosis or prognosis.

Today, we define evidence in medicine as any data and ensuing information useful either for the understanding of the health problem or for clinical and public health decisions about it ¹².� Such evidence may be very bad or extremely good within a considerably wide range. Grading of evidence appears now not only in treatment, but also in diagnosis and other areas of clinical interest.

A good clinical case report is a valuable piece of evidence for about twenty good reasons²⁵. Case reports and case series, with the exception of n-of one clinical trials, are not considered good evidence in terms of causation. Their value lies elsewhere. We should nevertheless prepare them lege artis.

In psychiatry, one may object that good evidence is not available because of the frequent absence of golden standards. The correct response to this contention is innovation and finding something that might replace the golden standards. (For example, a well-founded and structured consensus based as much on evidence like the recent American^26.27, or Canadian²⁸ consensus on the recognition, assessment and management of dementing disorders^26-28, depression^29,30, and other psychiatric topics³¹ might be pertinent.)

Disproportionate attention to hard and soft data in psychiatry

Some methodologists prefer hard data (measurable and well-defined such as blood count) to soft data (harder to quantify and measure such as mood). A hardening of soft data may be appropriate or other methodological ways of study such as qualitative research might be worthy of consideration.

Qualitative research is based on words rather than on numbers and uses a wide spectrum of descriptive terms, empathic retrieval of information from patients. It focuses on a distinctive context rather than on the representativeness of the problem under study³². In these terms, any psychiatric patient assessment might be seen as a piece of qualitative research, if it is well done! Qualitative research now has a rapidly improving and structured methodology from which psychiatry will increasingly benefit. Let us perform qualitative research as well as its quantitative counterpart.

Equilibrium between various steps of the evidence-based process

Current efforts and experience are disproportionately invested in various steps of the EBP practice.

• In evidence-retrieval, the methodology is becoming increasingly refined³³, but evidence-retrieval is not a main goal of EBP. It is one of its essential tools.
• In evidence-evaluation^9-12, the current experience is the richest one as reflected in any issue of Evidence-Based Mental Health.
• In applications to a particular patient and specific clinical setting, more than a psychiatrist�s clinical experience and patient values and preferences (often limited in psychiatry) are needed such as patient compatibility in characteristics, eligibility and settings of clinical trials and other studies from which usable evidence emerges.� Also, the EBM methodology of pondering risks and benefits in general and for each particular patient in psychiatry was worked up³⁴.
• Evidence implementation^35,36 and uses in practice must be integrated within local medical and societal structure, culture and preferences. Typically, evidence in psychiatry �does not travel well�.
• The evaluation of the evidence-based psychiatric care itself remains, for the moment, the weakest link in the EBP chain. The practice of EBP is a clinical intervention itself that should prove to be more beneficial for the patient than any of its alternatives like claim-based, authority-based, or unsubstantiated humanism and belief-based cares.

Cause-effect link oversimplifications

Causation in medicine is more than statistical significance. The brilliance of biostatisticians and other specialists of quantitative methodology often lets us forget logic and critical thinking based on knowledge and clinical experience. Small p does not necessarily mean an important causal factor. And because it is difficult to properly analyze �associations�, titles like �war and post-traumatic stress syndrome� or �electro-convulsive therapy and depression� are cheap substitutes for �a war experience is a cause (or is not) of post-traumatic stress syndrome� or �electro-convulsive therapy improves (or does not) depression in (well defined) patients�. Even editorial boards of some journals unfortunately follow such practices.

Lack of attention to logic and critical thinking in psychiatry

Informal logic and critical thinking are as vital in psychiatry as clinical knowledge and experience and mastery of quantitative and other research. h methodology. Discussing patients at clinical rounds or writing a coherent and meaningful �discussion and conclusions� section of a psychiatric paper requires the knowledge and mastery of logic and critical thinking. Current research methodology is vital for �problem formulation, material and methods and results� sections. Logic and critical thinking^37,38 is a learned experience as much as the theory of measurement or biostatistical analytical methods.

Such lack of attention to logic and critical thinking in psychiatry may be found even at the very core of clinical psychiatric practice. For example, the assessment of a patient�s thought content and structure is an integral part of a psychiatric interview and largely determining for the diagnosis of psychosis and other problems. However, clearer directions exist for thought content like delusions than for thought process. For example, do we have and do we need clearer inclusion and exclusion criteria to allow us to conclude that a patient�s thought process is tangential, that his ideas are taking off in several directions at once? Does psychiatry need better guides to tell us how to assess a patient�s argumentation? Does a psychiatrist always know what is a good or bad argument?

Computer-based methodology is extremely comforting to its users. Laptops, however, are less useful when it comes to discussion. Critical thinking counts much more³⁹. Forthcoming initiatives⁴⁰ should provide an even better balance between thinking and computing.

Lack of necessary ingredients for psychiatric community mental health program prioritization, implementation and evaluation

Community mental health programs are torn between good will and belief on one hand and the frequent lack of evidence on the other. Several ingredients are needed to determine priorities in the context of a mental health program as well as tools for the evaluation of program effectiveness. In order to say that this or that disease is a priority for an intervention, occurrence of disease, disease clinical importance, its controllability in proportional terms and operational considerations of health programs (target population) must be known. These measures vary between primary prevention (incidence control), secondary prevention (disease prevalence and duration control) and tertiary prevention (control of disease spectrum and gradient). If disease occurrence indices, its course and other clinical-epidemiological characteristics, attributable fractions as measures of potential effectiveness and accessibility to the target population are not known, mental health community programs and policies remain an act of good and/or political will and belief rather than evidence-grounded decisions. It is the psychiatrists� responsibility to find and use relevant evidence for all the above-mentioned components and no one else can do it for them.

Sackett⁴¹ recognizes three risks of preventive medicine if its programs are not evidence-based. They apply particularly to psychiatry. Preventive programs may be assertive if they indiscriminately target all individuals regardless of their health and risk characteristics (wearing helmets or seatbelts by law), presumptuous in their often blind belief that the program must do more good than harm (legalizing or fighting addictive substances) and overbearing (attacking those who question their value without counter-evidence). Programs in primary mental health prevention should be as rigorously justified and evaluated as are already secondary and tertiary prevention programs.

3.�������� Achievements and future directions

For an outsider, it is absolutely astonishing to see one of the most difficult medical specialties achieve so much in so little time. The diagnostic rules are well embedded now (and in very operational terms) in the Diagnostic and Statistical Manual of the American Psychiatric Association⁴².� EBM itself⁴³ as well as clinical and fundamental epidemiology in psychiatry are doing equally well⁴⁴. What is needed at this point is more high quality evidence that can be used to test the effectiveness of Evidence-based Psychiatry.

As for the future, we must still correct, improve, and complete the previously mentioned misconceptions, missing evidence and its proper uses. In other words, the following might be considered:

In medicine and community health as a whole, many highly qualified individuals, be they internists, epidemiologists or community medicine specialists as well as various academic groups around the globe, continuously contribute to the development of evidence-based medicine for uses in all medical specialties. Psychiatrists should not replicate such initiatives. It would be a waste of time, energy and human and material resources. Psychiatry and its related health sciences such as psychology, social services, or health services administration should focus on problems that other specialists cannot study as well: Obtaining valuable information about soft data, innovating research in ethically difficult situations, dealing with politically and economically challenging situations and problems, decision-making involving not only the patient and his or her psychiatrist, but the family, working environment, community and its administrative and political bodies, as well as obtaining evidence from these sources and using it to the best benefit of the patient, his or her family and his or her community.

Good fundamental and clinical epidemiology or rigorous quantitative research like the research provided by randomized double-blind controlled clinical trials is essential, but no longer sufficient. The evidence needed comes from a much wider spectrum of research domains and practical experience. Both must be integrated in an equilibrium, which has still not been attained. As already mentioned, EBM and EBP are about ten years young. We should not expect its past experience to provide everything. Instead, we should fulfill such expectations through our own innovative and relevant contributions in the years to come.

Let us paraphrase for our purposes what W Churchill said about democracy, (speech in the House of Commons, November 1947):� �It has been said that Evidence-based Psychiatry is the worst form of approach to mental health problem-solving except all those other forms that have been tried from time to time.� Empiricism, psychoanalysis, and the like: is there any other best practice in psychiatry and why?

EBP should remain an open system like democracy in EM Forster�s 1951 Two cheers for democracy: �Two cheers for Evidence-Based Psychiatry: one because it admits variety and two because it permits criticism.� If it doesn�t for some people right now, it should in its further evolution.

Just as EBM has already enriched psychiatry, EBP enriches the EBM experience across all medical specialties. EBP does not mean dehumanized psychiatry. It only drives our fulfillment of Hippocratic ideals in line with our best knowledge and experience to the benefit of the patients we serve.