Reichian "Emotional Armoring", Good Science Or Hokum?

[dasmoocher]

Some tidbits from the previously linked paper (Somatic experiencing treatment with social service workers following Hurricanes Katrina and Rita.):

However, the trauma field is now seeing the arrival of body-focused interventions such as the one used in this study. Somatic Experiencing®/Trauma Resiliency ModeTM (SE/TRM), in which the primary emphasis is on traumatic symptoms as patterns of dysregulation in the nervous system rather than on cognitions and emotions.
…
Participants in the present study were 142 staff from a social services agency who volunteered to attend the SE/TRM psychoeducation groups in the Baton Rouge and New Orleans offices and field sites. … Of the 272 staff who participated in the group sessions, 110 (40 percent) chose to participate in one to two individual SE/TRM treatment sessions. Of these, 19 participants had missing data on pretreatment variables and thus were removed from the sample, leaving a total of 91 participants in the treatment group, 51 of whom were selected for the comparison sample.
…
Because each social service staff member chose whether to receive treatment, assignment to the treatment and no-treatment conditions was nonrandom.
…
To correct for potential sample selection bias due to nonrandom assignment and to obtain unbiased estimates of the treatment effect, we used propensity score matching to create matched treatment and comparison groups. Propensity score matching is designed to find the best multivariate match for every treatment case from the available pool of comparison cases. For thi study, each person who chose treatment was matched with a person who did not choose treatment (that is, a person who had received only the psychoeducation group sessions) on the basis of a propensity score calculated from nine variables (gender, ethnicity, education level, city, coping, physical symptoms, psychological symptoms, PTSD symptoms, and resiliency).
…
The individual sessions used SE/TRM, a protocol that provides specific interventions that focus primarily on self-regulation (that is, restoring equilibrium to the nervous system) and secondarily on working with associated emotions and cognitions. SE/TRM teaches participants concrete skills to reduce their hyperarousal and dysregulation through tracking shifts in the nervous system by observing breath (rapid, shallow, panting), heart rate (increase, decrease), muscle tension, sifts in posture, changes in skin color,and involuntary body movements (eyes, head, neck, shoulders, hands, legs); resource use (internal and external);grounding techniques; pendulation (moving between states of relative organization and disorganization within the nervous system); and titration (the process of gradually accessing somatic activation, body sensations, feelings, and thoughts associated with the traumatic experience so that the nervous system can adjust to each increment without becoming overwhelmed) (Levine, 1997).
…
Levine, P. (2005). Somatic experiencing: Resilience, regulation, and self. Unpublished paper.

[my emphasis]

The fact that they’re looking at people who volunteered for the treatment is a red flag in my opinion; these may be people who believe in this stuff in the first place.

I don’t think I’ve ever seen an unpublished citation in the reference section of a paper before. You see things like manuscript in publication or personal communication, but to give a year and title to something no one can read?

Does anyone know if propensity score matching is the correct statistical analysis for self selecting groups? I’m a little confused on how to read this: “Of these, 19 participants had missing data on pretreatment variables and thus were removed from the sample, leaving a total of 91 participants in the treatment group, 51 of whom were selected for the comparison sample.” The comparison group is 51 of the 91 that got treatment?