4.1. Putative mechanisms of action
Interestingly, controlling for session attendance did not alter the magnitude of the comparative and within-group effects examined. Given that session attendance did not account for the differential efficacy of L-CBSMversus T-CBSM, what other factors might explain the relatively greater effects from L-CBSM? One potential explanation is the opportunity for visual transactions among live groups and/or via a more intensive social support experience. While telephone-delivered individual therapies have demonstrated efficacy comparable to face-to-face individual therapy in a variety of patient populations [27,43,44], including ME/CFS [45], past research on the efficacy of telephone-delivered group psychotherapy has been mixed. Using an individual format in past telephone-delivered therapies may have maximized participants' communication with their therapist and uptake of CBT-related skills. In this study of T-CBSM, participants may have been less engaged during sessions than in L-CBSM, and the study therapist could easily miss this lack of engagement in the absence of visual feedback. If, for instance, a patient was not engaged in a relaxation demonstration, she or he may not have learned the skill or benefited from practicing relaxation, and the T-CBSM therapists were not able to see the patient to provide corrective feedback or further instruction. It is also possible that a significant portion of the relative L-CBSM versus TCBSM intervention benefit comes from having weekly social support, which may have been more salient for group members seeing each other face-to-face (i.e., in L-CBSM).