When "This Session Is Going Well" Is Exactly Wrong

When "This Session Is Going Well" Is Exactly Wrong

You finish a session feeling the work is on track — the alliance is warm, the client is engaged, the energy in the room is good. Then they don't show up next week. When you reach out, you get a short reply: "I think I'm doing okay now." Something was missed. The hard part is that you can't see where.

The research of Lambert and colleagues (2001) reframes this experience. It isn't a personal lapse in clinical skill. Clinician self-assessment and actual client outcomes diverge in systematic, predictable ways. The remedy isn't a sharper gut — it's a structural measurement loop: track client progress with a brief instrument every session, let the therapist see the result immediately, and adjust the next session accordingly. That loop is the intervention Lambert demonstrated. This article walks through the evidence base for feedback-informed treatment — routine outcome monitoring (ROM) — and how to fold it into everyday practice.

Why Clinical Intuition Misses: The Problem of Systematic Bias

Clinicians read verbal content, facial expression, and the felt energy of the session, then form a judgment: this is going well, this isn't. The literature shows that judgment carries a systematic bias.

On average, clinicians rate client outcomes more positively than the outcomes actually are. Intuition is especially unreliable for the one task that matters most for safety: catching a deteriorating client early. A client can say "I'm fine" while quietly getting worse — and the clinician, reading the room rather than the trajectory, misses the signal.

This is not a competence problem. The human cognitive system has real limits when it comes to integrating complex, longitudinal process information accurately and in real time. A structured measure doesn't replace clinical judgment; it compensates for a known blind spot in it.

The Core Evidence: Lambert et al. (2001) and Lambert & Shimokawa (2011)

The Lambert et al. (2001) design. Six hundred and nine outpatients were assigned to three conditions: measurement with no feedback; measurement plus immediate feedback to the therapist; and measurement plus feedback plus clinical support tools.

The crucial result appeared in the cases that were trending downward. Clients who were deteriorating under the no-feedback condition did significantly better when their therapist received feedback. The feedback prompted clinicians to intervene earlier — before a quiet decline became a lost case.

Lambert & Shimokawa (2011) synthesized work on ROM systems such as PCOMS and the OQ-System, and drew a sharp conclusion: the payoff is in the cases at risk of failure, not in the average case. ROM doesn't make a session that's already going well go better. Its value is in surfacing the case that is quietly coming apart, early enough to do something about it.

ORS and SRS: One Minute of Measurement Per Session

The OQ-45 used in Lambert's research is on the long side for use at every single session. The practical alternative, developed by Miller and Duncan, is the pair of scales most associated with feedback-informed treatment: the Outcome Rating Scale (ORS) and the Session Rating Scale (SRS).

Together that's eight items, completed in under a minute. The ORS tracks the trajectory of overall functioning; the SRS measures the alliance and whether the session fit the client.

Five Steps to Make ROM a Clinical Routine

1. Introduce the tools in the first session

Try framing it plainly: "At the start of each session I'll ask you for a quick check-in — just a few marks on a scale to show how things have been going. It takes less than a minute."

This introduction makes measurement part of the structure of therapy. Frame it as the opening of a conversation, not a test.

2. Use the ORS as a start-of-session check-in

Plot the four ORS dimensions (individual, interpersonal, social, overall) cumulatively, and the client's trajectory becomes visible at a glance.

If the score fails to improve — or declines — across three consecutive sessions, treat that as an early-warning sign. Even when your read of the room says "this is going well," the scores may be signaling first.

3. Use the SRS as an end-of-session rupture detector

The SRS is completed as the session closes. A low mark on any item — relationship, goals, approach, or overall — is something to explore on the spot:

"It seems like something in today's session didn't quite land for you. Can you tell me what that was?"

A low SRS is a signal of alliance rupture — and an opportunity to repair it inside the same session, while it's still live.

4. Respond to deterioration signals immediately

When the ORS trends downward, don't press on with the technique you had planned. Find out what's happening first:

"Looking at your recent scores, it seems like things have been harder lately. What's been going on outside our sessions?"

That exploration is the concrete intervention that prevents dropout.

5. Bring the data to supervision

ORS/SRS data make supervision sharper. "This client is at session eight and the ORS has been flat the whole time" is far more actionable than "something feels off with this case." Data-grounded supervision beats impression-based supervision every time.

How ROM Reduces Dropout

ROM's effect is most pronounced in dropout prevention. In the Lambert & Shimokawa (2011) synthesis, at-risk cases treated without feedback dropped out at significantly higher rates than those in the feedback conditions.

The mechanism is simple. When a client is quietly deteriorating, a therapist with no feedback takes no corrective action — there's nothing visible to act on. When the ORS shows a downward trend, the therapist pauses the planned technique and explores the client's state first. That single change of direction is what prevents the dropout.

ROM isn't a tool for making good cases better. It's a safety net for the case that is quietly falling apart.

The Loop Signals Before Your Intuition Does

The central message of Lambert et al. (2001) for clinicians: measurement is not the point. The loop is the point — seeing the result immediately and adjusting the next session because of it.

To catch the quietly deteriorating client early, make one minute of measurement per session a routine. The client's scores will signal a session or two ahead of your intuition. Integrating the ORS and SRS into the structure of a session is one of the simplest ways to use data and clinical judgment together rather than choosing between them. A security-first AI partner like Modalia AI can support this routine by handling session documentation and tracking outcome data over time, so the feedback loop runs without adding to your administrative load.