The impact of interpersonal patient and therapist behavior on outcome in cognitive-behavior therapy. A review of empirical studies.

Hagopian et al. (2000) read every paper they could find on how the therapist acts during CBT. They looked for studies that linked warmth, empathy, and realness to client outcomes.

The team pulled data from mixed clinical samples. They did not run new experiments. They simply mapped what was already known about Rogerian traits inside a directive therapy.

Across the pile of studies, warmth, empathy, positive regard, and a solid alliance showed a steady, middle-sized impact on CBT results. These so-called "nonspecific" factors matter even when therapy is highly scripted.

In plain words, the therapist’s human side helps the manual work better.

Volpato et al. (2026) extends the same idea. They tracked clients for a year and saw that early CBT tasks actually built the alliance, which then fed back into symptom drops. The review said alliance matters; the new study shows it can grow through the homework you already assign.

Rojahn et al. (1994) is a predecessor case that previewed the point. One depressed adult got better when the therapist shaped warmer in-session responses on the spot. The review later bundled many studies saying the same thing.

Lappalainen et al. (2007) looks like a contradiction at first: novice therapists got better outcomes with ACT than with CBT. But their trial kept no alliance measures. The older review predicts that if the trainees had shown more empathy in the CBT arm, the gap might shrink or flip.

You can follow the manual and still be human. Smile, use the client’s name, ask how the last exercise felt, and mean it. These micro-moments are not fluff; they are active ingredients. Start sessions with a quick empathy check: "What part of last week’s homework felt hardest?" Then shape their answer with warmth and praise. Five extra minutes of real conversation can boost the whole protocol.