Practice Innovations, Vol 11(2), Jun 2026, 97-108; doi:10.1037/pri0000326
It is well-established that therapist empathy (i.e., understanding the patient’s internal state) correlates with patient improvement. However, existing research has treated empathy as a static construct, ignoring that it can fluctuate dynamically throughout treatment. Moreover, most studies assess empathy from one perspective only (i.e., a patient or therapist’s rating), leaving it unknown whether each dyad member’s perception uniquely predicts symptom change. Addressing these gaps, the present study examined the relative associations with next-session symptom reduction of within-person increases in therapist-offered and patient-felt empathy. In addition, we explored whether treatment condition (cognitive behavioral therapy [CBT] alone or CBT integrated with person-centered principles) moderated these associations. Adults with generalized anxiety disorder (GAD) were randomly assigned to either CBT alone (n = 42) or CBT integrated with motivational interviewing principles (MI-CBT; n = 42), which were specifically used to address any emergent patient resistance (Westra et al., 2016). Patients and therapists rated empathy after odd-numbered sessions, and patients rated their worry after every session. Multilevel structural equation modeling demonstrated that when controlling for prior improvement, only greater-than-usual patient-felt empathy was associated with next-session worry reduction. Treatment condition did not moderate either empathy–outcome association. Results support that dynamic fluctuations in patients’ experience of empathy may be an important change correlate across standard or integrative CBT. Thus, more than trying to offer empathy at higher levels, therapists should regularly assess their patients’ experience of received empathy and work to strategically enhance it when it is low or wanes. (PsycInfo Database Record (c) 2026 APA, all rights reserved)