Choosing an intervention that actually follows your theory

START HERE

The essential idea

Build a defensible connection between case evidence, a maintaining process, an intervention, and the observation that will guide the next decision.

BUILD THE REASONING · 01

Choose a target before choosing a technique

An intervention follows a theory when its intended function addresses the mechanism in the formulation. A familiar worksheet, metaphor, or reflection is not automatically a good next step because it belongs to a respected school. Begin with a client-defined outcome, a recent episode, and a provisional account of what maintains the difficulty. Specify the change you hope the intervention will produce. For example: “Repeated checking prevents the client from learning whether an adequately reviewed email is acceptable; a bounded experiment can test that prediction.” This is more useful than “Use CBT for anxiety.”

Distinguish several reasons the same behavior might occur. A worker may delay sending an email because they predict humiliation from a minor error, because the approval process is unclear, because writing in a second language takes more time, because they are ambivalent about the request, or because they have decided that anxiety must disappear before action. These possibilities overlap, but they imply different next moves. Ask what happens before, during, and after the delay, what the person expects, what they have tried, and what occurs when the obstacle is absent.

Select a target at the level of detail needed for a decision. “Low self-esteem” is too broad to identify whether to clarify a workplace rule, test a prediction, practice willingness, or explore ambivalence. A specific target might be “checking the same already-reviewed sentence because the client predicts one awkward phrase will cause professional rejection.” Another might be “waiting for certainty before acting on a chosen value.” A formulation should also identify external barriers and relevant assessment needs. Technique selection begins after enough inquiry to avoid treating a practical obstacle as a psychological deficit.

BUILD THE REASONING · 02

Three plausible interventions can serve different purposes

In CBT, a behavioral experiment tests a specific prediction through an agreed action and observation. State the prediction, its expected probability or meaning where useful, the action, and what outcome would count as evidence. Reduce only the checking or other safety behavior necessary to make the test informative and appropriate; do not create unnecessary professional risk. Compare prediction with observation afterward and consider alternative explanations. A thought record may help identify and examine an appraisal, but another record is not always the best next step when the maintaining process is repeated reassurance without new learning.

In ACT, a willingness or defusion exercise aims to change how the person relates to thoughts and feelings so that valued action becomes more possible. The target is not proving the thought false or eliminating discomfort before action. A client might notice “I am having the thought that this must be perfect,” make room for manageable anxiety, and send an appropriately reviewed message because clear communication matters to them. Evaluate flexibility and participation alongside distress. If the exercise becomes another ritual used to guarantee calm, revisit its function rather than congratulate successful completion of the technique.

In motivational interviewing, the next task depends on the conversation. If the client is unsure whether sending the request serves their priorities, further planning can bypass unresolved ambivalence. Explore their reasons, concerns, and preferred direction. If the direction is clear but confidence is low, evoke past successes, strengths, and feasible steps; offer information with permission when useful. If commitment and ability are sufficient, planning can be appropriate. An importance ruler is not a universal response to every difficulty. A person who values the action highly but lacks access or confidence needs a different inquiry from someone who has not chosen it.

BUILD THE REASONING · 03

Preserve function while adapting delivery

Adapt an intervention to language, literacy, disability access, culture, setting, and the client’s preferences. A spoken prediction recorded briefly by the counselor can serve the same experimental function as a written form. A values conversation can use the client’s own relational language rather than an individualistic ideal of independence. A planning task can fit available time and resources. State what is essential to the intervention’s function and what is merely its usual format. Changing a worksheet format can improve access; removing the prediction and review from an experiment can change what the task actually does.

Make the proposed action and rationale understandable enough for the client to choose. Explain what you think may be happening, what the task would test or practice, and what alternatives exist. Invite correction. A client may reveal that an apparently excessive review process is required by their workplace or that a prior error had serious consequences. That information changes the risk and formulation. Do not use the language of evidence-based practice to shut down disagreement. The APA framework integrates research evidence, expertise, and client characteristics, culture, and preferences; it does not reduce choice to selecting a branded technique.

Plan how to detect a poor fit. A task may be too burdensome, target the wrong process, become a new ritual, or strain the relationship. Agree what will be reviewed and what would lead to adjustment. A client who completes an experiment but feels coerced supplies both outcome data and relationship data. A client who cannot complete it because the necessary opportunity never occurred has not provided a clean test of the prediction. Examine implementation before concluding that the model failed or the client resisted. Fidelity means preserving a meaningful mechanism, not insisting on an unchanged form despite contradictory facts.

BUILD THE REASONING · 04

Use a decision record and a revision rule

A brief decision record can make theoretical reasoning visible: client goal; evidence for the maintaining process; chosen intervention and function; relevant evidence and fit considerations; agreed task; expected observation; and revision rule. This is not paperwork for its own sake. It helps a supervisor or future reader understand why this intervention came next and what would count as useful feedback. For a prediction test, the review should compare what was expected and observed. For an ACT exercise, it should examine the relationship to inner experience and valued action. For MI, it should attend to the client’s language and chosen direction.

Do not make symptom reduction the only marker for every approach. Anxiety can remain during a successful willingness exercise; a behavioral experiment can reveal that a feared event sometimes occurs while also changing an exaggerated conclusion about its consequences. Conversely, lower anxiety does not establish that the desired mechanism changed: reassurance can briefly calm a client while maintaining reliance on checking. Use the model-specific process and the client’s broader outcome together. Neither a theory-consistent process measure nor a symptom score alone proves that the treatment is effective for this person.

Revise with calibrated conclusions. If the workplace truly requires additional approval, problem-solving the process may take priority over reducing checking. If the client wants the action but cannot tolerate uncertainty, a flexibility-focused task may fit. If the client does not want the action once its costs are explored, the original goal may need revision. Combining methods can be appropriate, but explain the order and function instead of stacking techniques until something helps. Evidence-based selection remains an ongoing reasoning process: propose, collaborate, observe, and adjust within competence and the client’s actual circumstances.

FROM CONCEPT TO DECISION

The unsent project update

1. 1

   Identify the maintaining hypothesis

   The evidence suggests a threat prediction protected by repeated checking and requested reassurance. Importance is already high; another importance discussion would not address the clearest current process.

   “After the required check, what do you expect would happen if you sent the message without asking me to approve it?”

2. 2

   Agree an informative, proportionate task

   Choose a routine low-stakes update that still meets workplace requirements. Specify the predicted response and reduce the extra approval step; do not ask Aiden to send inaccurate or sensitive material.

   “Would you be willing to test one ordinary update after the required check, and compare the response with your prediction?”

3. 3

   Observe and interpret

   Suppose a colleague asks a normal clarifying question. Examine whether that is the predicted loss of trust, an ordinary communication event, or evidence requiring a revised prediction. Avoid reassuring away every ambiguity.

4. 4

   Revise the next intervention if function changes

   If Aiden understands the evidence but now waits for anxiety to vanish before sending, discuss whether a willingness exercise linked to timely communication would address that new target. Explain the change in function rather than rebrand the same task.

MAKE THE REASONING YOURS

Work through a new situation

READ WITH A PURPOSE

Best references

These readings deepen the specific models, evidence, and decisions in this lesson.

1. 01
   APA task force (2006): Evidence-based practice in psychology

   Connect research evidence, clinical expertise, and client characteristics, culture, and preferences when choosing and revising an intervention.

2. 02
   Beck Institute: Understanding CBT

   Check the cognitive formulation, guided discovery, behavioral work, cultural adaptation, and collaborative review principles.

3. 03
   ACBS: About Acceptance and Commitment Therapy

   Study the functional account of fusion, avoidance, flexibility, and valued action; this model account is distinct from evidence for a specific protocol.

4. 04
   Miller & Rollnick: Motivational Interviewing, fourth edition

   Use the originators’ engaging, focusing, evoking, and planning framework to distinguish ambivalence, confidence work, and planning; full book access may be required.