AAPB Minimum Quality Standards
AAPB's meetings are open to anyone interested in applied psychophysiology and biofeedback. Many of the attendees have little background in assessing the credibility or quality of presentations. Thus, it is incumbent upon the presenter to follow minimum standards of quality when presenting at AAPB. The cardinal rules for presented in clinical research are depicted on the table below. Presentations which do not meeting these standards will not be accepted for presentation and should neither be submitted nor presented.
| Cardinal Rules for Establishing Credibility When You: | ||
| Prepare a Clinical Presentation / Article | Listen to / Read a Clinical Presentation / Article | |
| Be certain to: 1. Title your presentation / article appropriately so it doesn't promise more than it can deliver or ascribe changes to one aspect of a multifaceted intervention. 2. Begin with a brief summary of what you did / found. 3. Describe the general characteristics of the group you worked with and define your inclusion and exclusion criteria. 4. Present how your patients were diagnosed. Don't fall into the trap of trusting diagnoses by others if such diagnoses are known to be frequently incorrect (e.g. physicians are terrible about correctly diagnosing headaches). Use recognized criteria so your audience will believe that your patients had the disorder you claim to be treating. 5. Use correct assessment techniques for the disorder (e.g. the MMPI is not valid for establishing the psychological components of low back pain). 6. Define your assessment so you establish the basis for saying that people learned the tasks you were teaching during your treatment. E.g., if you are teaching people to change their muscle tension, show the baseline status then show that those people who improved changed in the desired direction. This establishes the relationship between the intervention and changes in symptoms. 7. Use the correct outcome measurers and use them correctly. Review the literature so you are up to date. For example, 0 - 10 analog pain scales must define 10 to have an objective limit such as "would faint if had to bear the pain for one more second" rather than "most pain can imagine". 8. Establish pre and post treatment baselines of sufficient duration to establish symptom variability. E.g. headache baselines need to be between two weeks and a month. This is how you demonstrate effectiveness. 9. Use the correct design for the level of work done on the intervention already. E.g. a new idea needs only a baseline - intervention - baseline design while a test of an idea which has been shown to produce changes needs to incorporate a control group to show that changes are not due to non-specific effects. 10. Include sufficient subjects so your results are likely to be due to the intervention rather than chance variability. 11. Clearly explain what your intervention was and have some way to know that there was sufficient intensity to have a chance of causing a change. E.g. one relaxation session isn't likely to cure anything. 12. Present your results clearly with graphics rather than just tables. Show sufficient descriptive statistics so people decide what happened. 13. Never ascribe symptom changes to one aspect of a multifaceted intervention when you have no way to tease out the effect of that aspect. E.g. if you gave relaxation training and biofeedback, don't say that the changes were due to biofeedback. 14. Not worry about the need to prove an underlying mechanism for the technique you used. All you need to do in a clinical presentation is demonstrate that a change did take place. Other types of research demonstrate how & why. |
Is / are there: 1. Adequate diagnosis and assessment of the subjects? 2. Adequate pre treatment baseline to establish symptom variability? 3. Objective outcome measures relevant to the disorder? Were they used correctly? 4. Intensity of the intervention sufficient to produce an effect? 5. Way to check whether the intervention was successful (drug taken properly, behavioral technique successfully learned and then used). 6. Sufficient patient-subjects so result is credible? 7. Appropriate design for the question (e.g. single group, controls, believable placebo, etc.? 8. Sufficient descriptive statistics so results are clear? 9. Long enough follow-up so duration of results can be established? 10. In a multifaceted intervention, were any changes in symptoms ascribed to one element of the intervention when there is no way to differentiate the effects of each part? |
|
| From the inside front cover of "Clinical Research: Skills clinicians need to maintain effective practices" 2003, Behavioral Medicine Research & Training Foundation. By Richard A. Sherman | ||
