We are often asked about how to develop and test a treatment protocol, or how to test EMDR's effectiveness with a new clinical population. Different types of research are appropriate at different stages of development of a new protocol. Here are the stages most commonly used to develop a protocol:
1. You get an idea and decide to try it out. You have good luck with a case, and wonder if you could do it again. You may have used standard EMDR in a new type of treatment problem or with a new population (new for EMDR at least), or you may have adapted EMDR in some way; or you may have combined EMDR with other interventions in a unique way.
2. You try more cases with the same treatment problem, using some of the same approaches you used in the first one, and experimenting with other things as well. This will help you to figure out whether the first success was just a fluke or whether you might have discovered an innovation. At this stage, it's helpful to begin keeping a log that tracks key characteristics of the client and provides a place for brief notes on treatment response and variations to your approach that you've had to make.
3. You discern a pattern from your experience with #2, and develop a protocol. For example, if there are 5 steps and most people require most of the steps, you may simply include all 5 steps in the protocol to make sure that you're always covered. If most people only need two of the steps, you figure out what the difference is between those that need only two and those that need all five, and you identify the decision rules (this is where the log can be quite helpful).
4. You test the protocol, and your theory underlying the protocol, in a single system design (SSD) series with outcome measures appropriate to the treatment targets. SSDs are systematic case studies that employ repeated measurement over time so trends (and changes in trends) in behavior can be examined. At this point it's important to write down the protocol with as much specificity as possible, so that others who use your "treatment manual" can do what you did. In addition to creating a narrative that describes your protocol, a checklist (that covers each step) will make it much more likely that others will be able to implement your protocol correctly.
5. Generally, if you've changed the EMDR protocol in some way, you'll also need to present some single system design cases that also show how why a change in the standard EMDR protocol was necessary. One good method for doing this is the A-B-C single system design, where A is a baseline (pre-intervention) phase, B is the standard EMDR protocol; and C is the adapted protocol that you've created. The only time that this might not be necessary is when you're targeting something that is contraindicated for standard EMDR (such as behaviors to stabilize clients to increase readiness for trauma processing). The general rule is: if there is an established protocol already for what you're doing, you'll need to show how what you're doing is an improvement.
6. If the case series results are encouraging, try teaching your protocol to some other practitioners. Ideally, obtain their feedback, questions, & concerns on implementing the protocol (through a log, or through consultation). This step will identify steps in the protocol you haven't articulated well (usually because they were intuitive steps for you, but not intuitive for others!), and will help to refine your treatment manual for the next stage of this process.
7. You conduct a controlled study comparing your treatment to a wait-list control or to a standard-care-only control. Note that "standard care" at your research site may not be the same as "state of the art," so be prepared to provide some description of it.
8. If the controlled study results are encouraging, you conduct an additional controlled study comparing your protocol to the other leading approach to this treatment problem.
9. If the results of the controlled, comparative treatment study are positive, then step 8 should ideally be repeated by some other research team.
For steps 4-9, it can be especially helpful to consult researchers prior to beginning that particular step, so that you ensure that you've designed the methodologically strongest study for that stage. All of the steps above are such that at each phase of development, you are investing the minimum resources required to determine whether going to the next level is warranted. Sometimes the findings are not positive, and you have to decide whether your study was conducted poorly, whether you can modify your protocol to improve effectiveness, or whether it's not really worthwhile after all. If you choose to persist, you will probably want to repeat the "failed" step successfully before going further.
This model can also be used by those who wish to conduct a research project in a specific area of interest. By reviewing the literature you can determine the current stage of development for EMDR for your treatment problem of interest. With careful analysis you can decide whether the studies to date have been conducted properly, paving the way for the next phase, or whether something needs to be redone in a better way. Using this model can help researchers to avoid conducting case studies in areas where controlled studies have already been completed, as well as to avoid conducting controlled studies when case studies have not yet demonstrated that a larger project is warranted. With this model, researchers can use their resources to best contribute to the literature by accomplishing the natural "next step."
Bloom, M., Ficher, J., & Orme, J.G. (1999). Evaluating practice (3rd ed.). Boston: Allyn & Bacon.
Rothman, J., & Thomas, E.J. (Eds.)(1994). Intervention research: Design and development for human service. Binghamton, NY: Haworth.
Rubin, A. (July, 1997). Empirically validating EMDR with single case designs: A step-by-step guide for EMDR therapists. Workshop presented at the EMDR International Association Conference, San Francisco, CA.
Thomas, E.J. (1984). Designing interventions for the helping professions. Beverly Hills, CA: Sage