Are there any research studies done using the SFMA model?

That’s a question that needs changing of wording.
The SFMA is an assessment. To arrive at a list of movements that hurt, or are demonstrating limited movement as measured by clinical observation, then measured by goniometer, or movements that clear minimum range of motion passively but that are not able to be done passively.

So the research question would be what?
Does the SFMA reveal movements that hurt?
Does the SFMA reveal movements that are limited? 
Does the SFMA reveal movements that are not limited passively but that are limited actively?


Often, the question that is asked is, is there any research that shows the SFMA works? Which is a misunderstanding. The SFMA is not a treatment. It isn’t an intervention.
The intervention follows the assessment which tells you if something hurts - then you use clinical reasoning and further evaluation (local biomechanical testing) to understand more about the painful movement. 
The intervention follows the assessment which tells you if something is limited - then you use clinical reasoning and further evaluation (local biomechanical testing) to understand what is limiting the movement - is it tissue or joint? In which direction are the accessory motions limited?
And further, the intervention follows the assessment which tells you if something has above minimum mobility passively not the person can not demonstrate it actively (ie a motor control problem), which requires further evaluation of the motor control dysfunction, which starts with rolling.
Here is a publication that discusses the use of the SFMA to guide treatment of an athlete with low back pain:

1. Goshtigian, G. R., & Swanson, B. T. (2016). Using the Selective Functional Movement Assessment and Regional Interdependence Theory to Guide Treatment of an Athlete with Back Pain: A Case Report. Int J Sports Phys Ther, 11(4), 575-595. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/27525182


And in this book, in chapter 29, is a case study of the Functional Movement Systems approach (using SFMA) to manage a patient where physical therapy was used, over lumbar discectomy - https://www.elsevierhealth.com.au/health-professions/manual-therapy/clinical-reasoning-in-musculoskeletal-practice-9780702059766.html 

So, the question, does the SFMA model work, and has it been shown to work in research, assumes the SFMA is a treatment. It is not. It is an assessment.

Another question is - is the assessment reliable? There are two studies on this.
2. Glaws, K. R., Juneau, C. M., Becker, L. C., Di Stasi, S. L., & Hewett, T. E. (2014). Intra- and inter-rater reliability of the selective functional movement assessment (sfma). Int J Sports Phys Ther, 9(2), 195-207. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/24790781

3. Dolbeer, J., Mason, J., Morris, J., Crowell, M., & Goss, D. (2017). INTER-RATER RELIABILITY OF THE SELECTIVE FUNCTIONAL MOVEMENT ASSESSMENT (SFMA) BY SFMA CERTIFIED PHYSICAL THERAPISTS WITH SIMILAR CLINICAL AND RATING EXPERIENCE. Int J Sports Phys Ther, 12(5), 752-763. 


Another question is - is the assessment valid? The answer is that it is not a single assessment - it is a battery that contains, within it, tests that are widely used with their own protocols to ensure maximum reliability. For example, the Modified Thomas Test is reliable and valid if the tester sets up the flexed hip to the right range (too much hip flexion rotates the pelvis posteriorly and gives a potentially false positive for hip extension mobility dysfunction). For example, the C1/2 rotation test is a valid passive assessment of whether C1/2 has 40 degrees rotation as long as the tester maintains full cervical flexion. For example, the FABER(E) test is an accepted assessment of whether the individual has minimum acceptable combined hip abd/E/ER that, if not present, screens for possible hip/SI dysfunction that requires further evaluation. For example, the PSLR is a widely used test to reveal whether hip flexion is at least 80 degrees, as per minimal mobility standards, before the opposite hip moves.

Recent research has looked at the reliability and criterion validity of the ankle dorsiflexion screen used in the FMS and similarly in the SFMA:
See this link - https://www.researchgate.net/profile/Kyle_Matsel 


Here is a research paper that included the SFMA and reported on overuse symptoms in baseball, prospectively:
4. Busch, A. M., Clifton, D. R., Onate, J. A., Ramsey, V. K., & Cromartie, F. (2017). RELATIONSHIP OF PRESEASON MOVEMENT SCREENS WITH OVERUSE SYMPTOMS IN COLLEGIATE BASEBALL PLAYERS. Int J Sports Phys Ther, 12(6), 960-966. 


So, the SFMA is a model, using batteries of evaluations, to provide a diagnoses of movement patterns that are functional as defined by the criteria, dysfunctional, painful or not. Thus, it is a model to diagnose. IF the research question is whether the diagnosis is correct…… let’s break it down - how do you validate whether someone answering yes to “does it hurt?” Is valid or not? That’s subjective and it leads us to further evaluation. IF you said that answering yes to “does it hurt?” Is not valid, then you might argue that pain is subjective, which is correct. Which leads us to further evaluation, which is what the SFMA does.
IF the research question is whether the diagnosis of mobility dysfunction is correct, we only have to look at whether an inclinometer, goniometer, or clinical observation (using strict criteria) is reliable and valid, and that research is done independently of anything related to SFMA - which is why we borrow what is already acceptable. Research would say that there is some reliability between raters for clinical observation (such as Craigs test) and clinical use of goniometers and clinical use of inclinometers.
IF the research question is whether the diagnosis of altered motor control is correct, then we only have to rely on whether the passive mobility is present (see research question before about whether mobility is present according to clinical observation/goniometer/inclinometer) and whether the person can demonstrate smooth and easy control of that movement - the criteria in teaching SFMA is that if there is jerking during the movement, it is not smooth and thus not controlled. IF there is breath holding, it is not easy and it is thus not controlled since a high-threshold strategy for that movement of a body part is considered inappropriate.

So, the question about whether the SFMA model has been researched is not quite the right question - since it is a process to arrive at 3 possible diagnoses using previously accepted reliable/valid tests (where present). Since it is not an intervention, research about effect of intervention is not applicable.

The SFMA model, in level 2, does continue to teach local biomechanical testing, and provides a reminder of that LBT that is taught outside the SFMA 2 course. It is not a manual therapy treatment course, or ART, or dry needling, or IASTM etc. course. It is not a MET course, or visceral manipulation course, or craniosacral or biodynamics course. But it does put the clinician into the space to apply whichever intervention they are trained in, trust through research-guided understanding, and clinical experience. What the level 2 does introduce, is cueing, feedback and motor control training - of which there is a very large body of research to support such training. The intervention of rolling, based on a SMCD diagnosis, has been researched and a reference is provided:

Clark, N., Voight, M., Campbell, A., Pierce, S., Sells, P., Cook, R., Henley, C., Schiller, L. . (2017). The relationship  between segmental  rolling ability and lumbar multifidus activation time. Int J Sports Phys Ther, 12(6), 921-930. Retrieved from https://spts.org/docs/default-source/v12n6/ijspt-12_6-08-clark_abs.pdf?sfvrsn=2


As always, happy to continue to supply more research articles on other specific questions.

For ongoing posts of articles and case studies of the SFMA, check out this link -  www.functionalmovement.com/articles