15.1 Evaluating the Progress and Effectiveness of Interventions
Key Takeaways
- Routine Outcome Monitoring (ROM) and Feedback-Informed Treatment (FIT) using session-by-session measures (OQ-45, ORS, SRS) reliably improve outcomes and reduce deterioration, with effect sizes comparable to adding a new treatment modality
- Single-case designs (AB, ABAB, multiple baseline) are the social worker's tool for rigorously evaluating whether an intervention is working for an individual client, not just for research
- Goal-Attainment Scaling (GAS) converts individualized, idiographic client goals into a 5-point scored scale (-2 to +2) that can be aggregated across clients while preserving individualized targets
- When progress data show no improvement, the expected sequence is: review the data, consult the client, modify the intervention, seek consultation, or refer — not simply continue the same plan
- Documentation of progress must be ongoing, measurable, tied to the treatment plan's goals and objectives, and sufficient to justify continuation, modification, or termination of services to a third-party reviewer
Intervention without evaluation is practice by guess. The ASWB Clinical exam expects you to treat evaluation as part of intervention, not a separate phase that comes after. Within the 32% Intervention and Practice domain, the competency on evaluating progress and effectiveness tests whether you can close the loop: collect data, interpret it with the client, decide whether the plan is working, and act on that decision. This section covers the methods the exam expects you to recognize and apply.
Why Evaluation Matters for the Exam
Vignettes on the exam commonly describe a client several sessions into treatment whose symptoms have or have not improved. You will be asked what the social worker should do next. The correct answer almost always involves reviewing measurable progress data with the client before deciding to continue, modify, or terminate — not simply pressing on with the original plan because it was clinically sound at intake. The exam also tests your familiarity with named measurement tools and single-case designs.
Routine Outcome Monitoring (ROM) and Feedback-Informed Treatment (FIT)
Routine Outcome Monitoring (ROM) is the practice of administering a brief standardized measure at every session (or at regular intervals) and using the score to track the client's trajectory. Feedback-Informed Treatment (FIT) is a specific form of ROM in which the client's scores are compared against an expected treatment trajectory, and deviations trigger a clinical conversation. The evidence base — including the work of Lambert, Miller, Duncan, and Sparks — shows that FIT reliably improves outcomes, reduces no-shows and dropout, and cuts deterioration rates by identifying non-responding clients early.
Three measures appear most often:
| Measure | Type | Length | What it captures |
|---|---|---|---|
| OQ-45 (Outcome Questionnaire) | Clinician- or client-administered | 45 items | Symptom distress, interpersonal role, social role — broad functioning |
| ORS (Outcome Rating Scale) | Client self-report | 4 visual analog items | Individual, interpersonal, social, overall well-being |
| SRS (Session Rating Scale) | Client self-report | 4 visual analog items | Therapeutic alliance: relationship, goals, approach, overall |
The ORS tracks outcome session by session; the SRS tracks the alliance in the same visit. Together they form the core of the PCOMS (Partners for Change Outcome Management System), listed on SAMHSA's National Registry of Evidence-based Programs and Practices. A client whose ORS is not rising by the third session, or whose SRS drops, is flagged for a conversation about what is and is not working — not silent continuation.
Single-Case Designs for Evaluating Individual Treatment
Single-case (sometimes called single-subject or N=1) designs let a clinician evaluate whether a specific intervention caused change for one client. They are a bridge between research and practice because they are rigorous yet idiographic. The exam expects you to recognize the basic phases.
| Design | Phases | What it demonstrates |
|---|---|---|
| AB | Baseline (A), then intervention (B) | Change coincides with intervention; weak internal validity because history is not controlled |
| ABA | Baseline, intervention, then withdraw intervention | Reversal strengthens causal inference but raises ethical concerns about withdrawing effective treatment |
| ABAB | Two cycles of baseline and intervention | Reversal design; stronger internal validity, same ethical concern |
| Multiple baseline | Same intervention started at different times across behaviors, settings, or people | Demonstrates effect without ethically problematic withdrawal; the strongest single-case option for clinical practice |
In clinical practice, you most often use a multiple-baseline logic: start the same intervention at different times across two target symptoms (e.g., panic frequency, then avoidance behavior) and observe whether change follows the introduction of the intervention each time.
Goal-Attainment Scaling (GAS)
Goal-Attainment Scaling (GAS) is an idiographic outcome method that fits cases where standardized measures do not capture the client's unique goals. The social worker and client write specific goals and define five levels of outcome for each:
- -2: much less than expected (worse)
- -1: less than expected
- 0: expected level of outcome
- +1: more than expected
- +2: much more than expected
A GAS score is computed by weighting each goal's outcome and converting to a standardized T-score with a mean of 50. GAS lets you aggregate across heterogeneous clients while preserving individualized, client-centered targets.
Revising the Treatment Plan Based on Data
The treatment plan is a living document. When session-by-session data, single-case data, or GAS scores show no improvement (or deterioration), the exam expects a defined sequence:
- Review the data with the client. Share the trajectory, not just the latest score. Ask the client's view of what is and is not working.
- Examine the alliance. A dropping SRS points to a relational problem; fix the relationship before changing technique.
- Re-examine the diagnosis and formulation. Is the working hypothesis correct? Are there unrecognized comorbidities, stressors, or cultural factors?
- Modify the intervention, consult, or refer. Switch technique within the same evidence base, seek formal consultation, or refer to a specialist.
- Document the change and the rationale. The revised plan should be dated, signed, and tied to the data.
Clinical judgment and data are not adversaries. The competent social worker integrates both: data trigger the conversation; clinical judgment and the client's values drive the decision.
When the Intervention Is Not Working
Continuing an ineffective intervention is an ethical problem, not just a clinical one. The NASW Code of Ethics (1.04) requires competence; if the plan is not producing progress and you lack the expertise to modify it, you must consult or refer. Documentation of the decision to modify, consult, or refer — and the data on which it was based — is what protects the client and the clinician if the case is later reviewed by a payer, a licensing board, or a court.
Documentation of Progress
Progress notes should be tied to the specific goals and objectives in the treatment plan, should report measurable change (scores, frequency counts, GAS levels), and should be sufficient to justify continuation, modification, or termination. Third-party payers and accreditation surveys expect to see that the plan was reviewed at a defined interval (often every 30, 60, or 90 days depending on setting), that the client participated in the review, and that any change was documented and signed.
A clinical social worker has been seeing a 38-year-old client with major depressive disorder for eight weekly sessions of CBT. The client's OQ-45 score has not declined from baseline, and the worker's SRS scores have hovered in the average range. Which is the MOST appropriate next step consistent with feedback-informed treatment?
A school-based social worker wants to evaluate whether a self-management intervention is reducing a 10-year-old's classroom outbursts. To avoid the ethical problem of withdrawing an effective intervention, which single-case design is MOST appropriate?
Clinical Judgment Plus Data
A common exam trap is the option that treats data as the sole driver of decisions or the option that treats clinical judgment as sufficient on its own. The correct answer almost always integrates the two: the OQ-45 score tells you the client is not improving, but clinical judgment — informed by the alliance, the client's context, and cultural formulation — determines whether the next step is a technique change, a consultation, or a referral. Evidence-based practice, per the Sackett definition adapted for social work, integrates the best research evidence, the clinician's expertise, and the client's values and preferences. The same triad applies at the level of progress evaluation.
A client and social worker set a goal of "client will manage anxiety in workplace meetings using a coping script." Using Goal-Attainment Scaling, which outcome level corresponds to the client using the script effectively in most meetings and reporting meaningfully reduced anxiety?