Cornwall Counseling · Curiosity Essays
What the Score Can’t Tell Me
Clinical judgment, psychological testing, and the DSM
By Michael Cornwall
I am unsettled when a person fills out a questionnaire and the score seems to decide what diagnosis they have and what we will do about it. There is no exact science for understanding another person’s psychology. We can measure certain things with reasonable consistency and still be wrong about what they mean in that person’s life.
A number can look precise. It might tell me someone scored 14 last month and 10 today. The arithmetic is exact. Whether that drop means the person is doing better takes more work to understand. My view is that a seasoned clinician who has listened over time can often give a more accurate account of how treatment is going than a test score alone. I can explain what makes me think the person is improving, where they are still struggling, and how certain I am. My judgment remains open to what I learn next.
What kind of question are we asking?
A quiz on my website invites you to learn something and reflect on it. A validated clinical screener serves a different purpose. It looks for signs that deserve a closer look, and its usefulness depends in part on who is taking it and what it was designed to detect. A positive screen can occur in someone who does not meet diagnostic criteria, and a negative screen can miss someone who does. Research on the PHQ-9, a depression screener, shows both kinds of error even at a commonly used cutoff (Levis et al., 2019).
Formal psychological testing can answer more specific questions. If someone has a long history of struggling to read, a qualified evaluator can use standardized tests to examine particular reading skills and consider the results alongside educational history and other information. The test can contribute evidence a conversation alone would have difficulty producing. The evaluator still has to decide whether the test suits the question and how factors such as language, culture, and testing conditions affect what the results mean (American Psychological Association, 2020).
The DSM-5-TR provides shared criteria for identifying mental disorders (American Psychiatric Association, 2022). The clinician gathers the history, considers other possible explanations, and decides whether the applicable criteria fit. The American Psychiatric Association (n.d.) advises clinicians to use its DSM assessment measures to support clinical decision-making and to draw on other evidence when making a diagnosis.
How we answer questions about ourselves
When I ask someone how they have felt over the past two weeks, I am asking them to remember those days, decide what counts as a symptom, and fit the answer into the choices I gave them. Yesterday may loom larger than last Tuesday. Someone who has lived with exhaustion for years may call it normal. Someone else may deliberately leave something out because they worry what will happen if they disclose it. Self-report can be influenced by recall and by the answers people feel comfortable giving (Althubaiti, 2016).
Those same influences can show up in a conversation. What time gives me is a chance to ask for examples, revisit an answer, and give a person room to tell me something they were reluctant to say at first. If someone checks “sleeping well” and later describes waking every two hours, I want to know how they decided they were sleeping well. They may mean they slept better than they did a month ago. I would have missed that distinction if I had treated the checkbox as the whole answer.
What I would ask next
Imagine two people who both mark that they sleep poorly, have little energy, and struggle to concentrate. One says these difficulties began after a move to overnight shifts. The other says they have lost interest in things they once enjoyed and are beginning to miss work. The same score gives me a reason to ask both people more questions. It gives me very little reason to ask those questions in the same order.
“How were you sleeping before your schedule changed?”
“When you finally have a day off, how do you spend it?”
“You said you’ve stopped answering friends’ calls. What happens when the phone rings?”
I would want to understand the timeline, daily functioning, relevant health or medication changes, and any safety concerns that arise. A shift change could be part of the picture, and depression could still be present. With either person, I may need more information before I can reach a diagnosis. I can say so and explain what I am still trying to understand.
What time and experience add
Years in practice can help me recognize when a familiar explanation fits poorly or ask a better question when the answers seem to conflict. Time with this particular person gives me something different. I can hear how they described a difficulty at the first visit, what happened when they put a new skill to use, and what they still find hard. A clinician can also become attached to an early impression. A review of clinical judgment studies found only a small average gain in accuracy associated with experience (Spengler & Pilipis, 2015). Another review found that formal rules for combining information generally matched or outperformed unaided judgment on the prediction tasks studied (Grove et al., 2000). Those findings give me reason to check my impressions even when I have seen a similar situation many times.
Diagnosis is one part of the work
Even when two people meet the same diagnostic criteria, I may hear different answers when I ask, “How would you know our work together was helping?” One might want enough energy to get through a workday. Another might want to sit through dinner with family again. Their answers help us choose what to work on and give us something concrete to revisit at the next visit.
Suppose the second person’s symptom score barely moves. They still feel tense at dinner, but they are now staying at the table and taking part in the conversation. The distress deserves attention, and staying at the table matters because it was a goal we agreed on. Conversely, a worsening score can bring a difficulty to my attention that I missed in conversation. Reviewing measures together can help us see changes and decide whether to adjust the plan (Lewis et al., 2019).
What changed over time?
- Read the score
The symptom score barely moved.
- Ask about daily life
“How did dinner with your family go?”
- Compare with the goal
“I still felt tense, but I stayed and joined in.”
- Revisit the plan
Recognize the progress and keep working on the tension.
When I assess treatment, I want to be able to say what I have heard, what I have observed, and what the person says is different in daily life. If those pieces point in different directions, I can say where I am uncertain and ask what we should look at again. A diagnosis gives me a place to begin. Understanding how the work is going takes time with the person.
References
- Althubaiti, A. (2016). Information bias in health research: Definition, pitfalls, and adjustment methods. Journal of Multidisciplinary Healthcare, 9, 211–217. https://doi.org/10.2147/JMDH.S104807
- American Psychiatric Association. (n.d.). DSM-5-TR online assessment measures. https://www.psychiatry.org/psychiatrists/practice/dsm/educational-resources/assessment-measures
- American Psychiatric Association. (2022). Diagnostic and statistical manual of mental disorders (5th ed., text rev.). https://doi.org/10.1176/appi.books.9780890425787
- American Psychological Association. (2020). APA guidelines for psychological assessment and evaluation. https://www.apa.org/about/policy/guidelines-psychological-assessment-evaluation.pdf
- Grove, W. M., Zald, D. H., Lebow, B. S., Snitz, B. E., & Nelson, C. (2000). Clinical versus mechanical prediction: A meta-analysis. Psychological Assessment, 12(1), 19–30. https://doi.org/10.1037/1040-3590.12.1.19
- Levis, B., Benedetti, A., & Thombs, B. D. (2019). Accuracy of Patient Health Questionnaire-9 (PHQ-9) for screening to detect major depression: Individual participant data meta-analysis. BMJ, 365, Article l1476. https://doi.org/10.1136/bmj.l1476
- Lewis, C. C., Boyd, M., Puspitasari, A., Navarro, E., Howard, J., Kassab, H., Hoffman, M., Scott, K., Lyon, A., Douglas, S., Simon, G., & Kroenke, K. (2019). Implementing measurement-based care in behavioral health: A review. JAMA Psychiatry, 76(3), 324–335. https://doi.org/10.1001/jamapsychiatry.2018.3329
- Spengler, P. M., & Pilipis, L. A. (2015). A comprehensive meta-reanalysis of the robustness of the experience-accuracy effect in clinical judgment. Journal of Counseling Psychology, 62(3), 360–378. https://doi.org/10.1037/cou0000065