Evidence-based assessment/Vignettes/Arlene

Click Here for Landing Page

HGAPS New for Fall 2022: HGAPS and Psychology Conferences

Click Here for Landing Page

HGAPS is finding new ways to make psychological science conferences more accessible!

Here are examples from APA 2022 and the JCCAP Future Directions Forum. Coming soon... ABCT!
~ More at HGAPS.org ~

---

Arlene

Just Being a Teenager?

Clinical description

Arlene is a 16 year old Hispanic female who lives with her mother, father, and three siblings (a sister who is four years older and now in college, and a brother 1 year younger and sister two years younger). She is in regular education classes, and had been an A & B student until this year. Her self esteem has seemed more low this year, and her last marking period she had her lowest grades ever. She becomes angry when pressed about her grades, screaming and storming into the bedroom and locking everyone out for hours at a time. Her parents do not want her dating. However, she has a boyfriend and has been sneaking out of the house late at night to spend time with him. Her father caught her, and there was a massive argument. That weekend, she took a bottle full of approximately 200 aspirin, but then she made herself throw up. She has not talked with her parents about this, but did call her older sister, who encouraged her to talk with the school counselor that had been supportive of the sister.

Extended content
History of presenting problem Conceptualization Initial treatment plan

Assessment findings

Checklist scores

Arlene, her mother, and her social studies teacher all completed the Achenbach System of Empirically Based Assessment (ASEBA) checklists. Here are the results, reported as T scores (M = 50, standard deviation (SD) = 10, compared to other women between 11 and 18 years of age).

ASEBA Scores

Scale	Mom	Arlene	Teacher
Externalizing	67	73	68
Internalizing	59	68	46
Anxious/Depressed	57	62	50
Withdrawn	66	68	52
Somatic Complaints	50	70	50
Attention Problems	62	68	58
Social Problems	61	63	62
Thought Problems	55	64	50
Delinquent/Rule-Breaking	62	70	65
Aggressive Behavior	69	72	68

Substance use screening

The ASEBA includes three items (#2, #99, #105) that assess substance use, asking about tobacco, alcohol, and other drug use on the age 6-18 Child Behavior Checklists (CBCL) and Teacher Report Form (TRF), and the 11-18 Youth Self Report (YSR). These do not get a separate score in the scoring software or output. They are all embedded in the Rule-Breaking scale. In the paper version, there is a space for the respondent to write in a brief description. We are supposed to check these items manually, and read the description and probe during the interview.

If we combine the items, they work as a screening tool for substance misuse. Summing (or averaging) three substance-related items on the Achenbach Child Behavior Checklist or Youth Self Report is a quick way of checking whether more evaluation is needed about clinically concerning substance use. Low scores effectively rule out substance concerns in outpatient mental health settings; high scores increase the odds of misuse and warrant further assessment with more specialized methods.

Based on her mother’s report, Arlene’s average score on the substance use items of the CBCL was a 0.35 (raw score of 1); based on Arlene’s own report on the YSR, her average score on the substance use items was a 0.90.

Using a probability nomogram, we can integrate the above intake information (see Figure). We begin with a published benchmark of 4.5% for adolescent SUD (See Table 7.46B, Substance Abuse and Mental Health Services Administration, 2020).[1] Next, we incorporate the information from Arlene’s mother’s CBCL report. The DLR associated with an average CBCL substance use score of 0.35 is 4.03 (see Table 5). Drawing a straight line from the base rate through DLR 4.03 gives us a posterior probability of SUD at about 16%. From here, we can integrate Arlene’s YSR scores. First, we plot the above post-probability as the pre-test probability in the second nomogram (16%, see Figure 3b). Similarly, we incorporate the DLR of 9.01 associated with Arlene’s average score of 0.90 on the YSR substance use items. Extending a straight line from the new pre-test probability through the DLR 9.01 (from Table 5) gives us a final posterior probability of about 63% for Arlene to meet criteria for a SUD. Additional informational scores could be added to the nomogram to refine the posterior probability of a SUD diagnosis, assuming the DLR(s) associated with the score has been calculated.

Nomogram with worked vignette using the CBCL substance use scores

Note. Arlene is a 16-year-old Hispanic female who self-referred to a school counselor after an aborted suicide attempt that she engaged in following a conflict with her parents. Arlene, her mother, and her social studies teacher all completed the Achenbach System of Empirically Based Assessment (ASEBA) checklists. From her mother’s report, Arlene’s score on the substance use items of the CBCL was a 1; based on Arlene’s own report on the YSR, her score on the substance use items was a 3.          

(1)  Select the pre-test probability and mark on the first vertical line from the left. In this example, we use published benchmarks 4.5% as the base rate of SUDs in adolescents between age 12 and 17.

(2)  Identify the diagnostic likelihood ratio (DLR) associated with the risk factor and plot on the second vertical line. In this example, an average CBCL substance use subscale score of .35 is associated with a DLR of 4.03 (from Table 5 in this paper).

(3)  Connect the dots from (1) and (2) and extend the line to find a post-test probability of 16%.

(4)  To incorporate another information, repeat the process by using information from step (3) as the new starting point (shown via dotted lines).

(5)  An average score of .90 on the YSR substance use items has a DLR of 9.01.

(6)  Connect the dots from (4) and (5) to find a revised post-test probability of 63%.

Extended content
Select more specialized scales to refine probabilities Updating probabilities Critical items

Diagnostic interview findings

Diagnoses are based on a LEAD (Longitudinal expert evaluation of all data) consensus meeting following a Kiddie Schedule for Affective Disorders and Schizophrenia (KSADS) interview, using DSM-IV criteria. The same interviewer met with Arlene and then her mother, then discussed any differences of opinion with them as needed to use clinical judgment. KSADS results were reviewed with a licensed clinical psychologist to arrive at a final decision.

• Major Depressive Disorder (85% confidence post interview)
• Dysthymic Disorder (85% confidence)
• ADHD Inattentive (75% confidence)
• ODD (70% confidence)
• Anxiety NOS (65% confidence)
• Rule out: panic without agoraphobia (45% confidence)

Cognitive and achievement testing

Not done as part of the evaluation.

Prediction phase

Let's see how we would apply the EBA principles to Arlene:

Shortlist of probable hypotheses

Based on Arlene's age and the common clinical issues, here are the possible issues:

• A mood disorder definitely is a leading hypothesis. The suicide attempt also suggests evaluating mood disorder (although not everyone who attempts suicide has a mood disorder). Within the "mood" category, the hypotheses should consider major depression, dysthymia, and bipolar spectrum disorders, as well as other medical issues that could lead to mood symptoms.
• Substance misuse should be another hypothesis, based again on its prevalence in her age group.
• Anxiety disorders would be a third hypothesis.
• Conduct problems would be a fourth -- they are not immediately suggested by the description of the presenting problem, but they are common in the age group, and they also can be a risk factor for self harm.
• Attention problems are worth evaluating based on prevalence, though her prior academic performance does not suggest any additional reason for concern.
• The family conflict is also important to assess, as well as potential cultural issues (and differences of opinion between Arlene and her parents that might be influenced by differing degrees of acculturation).

Risk and protective factors and moderators

Arlene's gender and age increase the probability of a mood disorder, and may reduce the chances of conduct disorder. Her solid academic performance previously suggests potential resilience.

Her conflict with her father, and her keeping things secret from her parents, would be considerations before doing family therapy, and they may complicate consent for treatment (Arlene is still a minor). Additionally, family relations can impact the development and maintenance of depression. There is a strong link between childhood depression and dysfunction in the family. This would put Arlene at a higher risk for depressive symptoms to maintain if the family dysfunction remains.

Arlene seems to find support from her oldest sister. She could be a good resource for Arlene and could be considered in family therapy.

It would be helpful to know if either of Arlene's parents have a mood disorder. If they do, that would put her at much higher risk for having a mood disorder herself.

Some data suggest that Interpersonal Psychotherapy (IPT) may be particularly effective with Hispanic teens, perhaps moreso than Cognitive Behavioral Therapy (CBT), because of the greater emphasis on family (and familism). IPT would have an advantage of not requiring active participation of the father (unlike family therapy), since IPT is designed as an individual therapy.

Updating probability of diagnoses

Below is a worksheet with the DLRs left blank to be filled in. Answers are below.

Arlene	Common Dx Hypotheses (A)	Starting Prob. (B)	Broad Measure (D)				Cross-informant (E)						Confirmation (G)	Treatment Phase (I), (J), (K)
	Base Rate from Merikangas et al. (2010) NCS-A		Scale & Score	DLR	Revised Prob.	EAY Check	Next Test score	DLR	Revised Prob.	Next Test score	DLR	Revised Prob.	K-SADS Interview
Any Anxiety	Specific Phobia	0.19											Anxiety NOS (65%)
	PTSD	0.05
	GAD	0.02	CBCL T Internalizing 59	0.98	0.02	0.02	Other measures are better than Achenbach
	Panic Disorder	0.02
	Social Phobia	0.09
	Separation Anxiety	0.08
Any Impulse Control Disorder	ODD	0.13	CBCL T Aggressive 69	4.18	0.38	0.38	No data about TRF scales for aggressive						ODD (70%)
	CD	0.07	CBCL T Aggressive 69	4.18	0.24	0.24
	ADHD	0.09	CBCL T Attention 62	6.92	0.41	0.41	TRF T Attention 58	0.73	0.34				ADHD inattentive (75%)
Any Mood Disorder	MDD	0.12	CBCL T Anx/Dep 57	0.39	0.05	0.05	Haven't found data about TRF for internalizing						MDD (85%)
	BP	0.03	CBCL T Externalizing 67	1.26	0.04	0.04	YSR T Externalizing 73	2.32	0.09	TRF T Externalizing 68	2.03	0.17
	Dysthymia	Included above											Dysthymia (85%)
Any Substance Abuse Disorder		0.11	CBCL: Read and sum items #2, 99, 105

Answers

Arlene Common Dx Hypotheses (A) Starting Prob. (B) Broad Measure (D) Cross-informant (E) Confirmation (G) Treatment Phase (I), (J), (K) Base Rate from Merikangas et al. (2010) NCS-A Scale & Score DLR Revised Prob. EAY Check Next Test score DLR Revised Prob. Next Test score DLR Revised Prob. K-SADS Interview Any Anxiety Specific Phobia 0.19 Anxiety NOS (65%) PTSD 0.05 GAD 0.02 CBCL T Internalizing 59 0.98 0.02 0.02 Other measures are better than Achenbach Panic Disorder 0.02 Social Phobia 0.09 Separation Anxiety 0.08 Any Impulse Control Disorder ODD 0.13 CBCL T Aggressive 69 4.18 0.38 0.38 No data about TRF scales for aggressive ODD (70%) CD 0.07 CBCL T Aggressive 69 4.18 0.24 0.24 ADHD 0.09 CBCL T Attention 62 6.92 0.41 0.41 TRF T Attention 58 0.73 0.34 ADHD inattentive (75%) Any Mood Disorder MDD 0.12 CBCL T Anx/Dep 57 0.39 0.05 0.05 Haven't found data about TRF for internalizing MDD (85%) BP 0.03 CBCL T Externalizing 67 1.26 0.04 0.04 YSR T Externalizing 73 2.32 0.09 TRF T Externalizing 68 2.03 0.17 Bipolar I (85%) Dysthymia Included above Dysthymia (85%) Any Substance Abuse Disorder 0.11 CBCL #2 CBCL #99 CBCL #105

Cross-informant perspectives

Mention that these have DLRs. Also unpack the implications of agreement and disagreement for the client (and add a section about treatment implications of disagreement on the Conceptual Model Pages)

Prescription phase

Genogram and family functioning

Here is a genogram of Arlene's family:

Treatment selection

The diagnostic interview suggests a combination of a major depressive episode and a prior dysthymia, sometimes referred to as a "double depression." This suggests that Arlene's stress and mood problems have persisted for a long time, and may be more difficult to treat. The mood disorders clearly are associated with impairment and should be a major focus of treatment.

Cognitive Behavioral Therapy (CBT) is a well-supported treatment option. Interpersonal Psychotherapy for Adolescent Depression (IPT-A) is a good option as well. IPT may be the best option for Arlene, because studies have shown it to be especially effective with Hispanic youth. It places a large emphasis on the adolescent's independence, and focuses on the reciprocal relationship between mood and relationships.

Interpersonal Psychotherapy

Interpersonal Psychotherapy (IPT) is a time-limited treatment that focuses on the relationships between the patient and other people. IPT focuses on relationships because they are critical for the well-being and psychological adjustment of the patient. The original purpose of IPT was to reduce symptoms of Depression. However, it is now utilized to improve/enhance the quality of the patient's relationships with others. Research since its development demonstrates that IPT is effective when adapted for mood, eating, and anxiety disorders.

IPT has three defined phases (beginning, middle, end) that require strategies and tasks for both the therapist and patient. These phases aim at resolving four main social areas: grief, interpersonal role disputes, role transitions, and interpersonal deficits. The first phase typically lasts 1-3 sessions. During this part of treatment, the therapist conducts a diagnostic assessment and reviews the patient's psychiatric history, which establishes the context for the treatment. The middle phase of treatment focuses on specific strategies to deal with potential problem areas: complicated bereavement, role disputes, role transitions, interpersonal deficits. A role transition is when a person goes from one role in a relationship to another role (e.g., the end of a relationship with divorce or the beginning of a new relationship with marriage). When the end of treatment is near, the clinician/therapist reminds the patient that this is a role transition.

Outcomes of treatment through IPT are influenced by the relations between the patient and significant others. Since there is a time limit on treatment, the patient is pressured to take action as an overall task.

Why is IPT a plausible treatment option for Arlene?

When studying how IPT could be adapted for Hispanic populations, group meetings were excluded, concluding IPT, which focuses on the individual, is better for Hispanic populations.

This centrality of family adaptation of IPT includes the impact of family and the patient's role in it on their changes in mood or emotions while having deep ties of self in the family. The improvement of the individual role develops a more cohesive family dynamic. This adaptation best suits Arlene with the current familial conflicts she is experiencing, from her rebellion against her parents with her current relationship (role transition) to trying to overdose after an extreme argument with her father.

Moderating factors

The double depression is a moderating factor suggesting worse prognosis, along with potential demoralization and early drop out from treatment.

Client preferences

Arlene was originally leaning towards an antidepressant medication, thinking that she could take it without telling her parents. After discussing the pros and cons of medication (including the effect size in youths, the potential side effects, and the fact that her parents would find out as part of the consent process), as well as the pros and cons of different evidence based therapies, she elected to try IPT. She wanted to revisit the possibility of a stimulant helping with her inattention, but she opted to wait and see if that improved along with her mood if the IPT helped.

Process phase

Clinically significant change

Reliable change index

Pick a treatment target and specify what the RCI would be for it. Discuss how you would explain to Arlene

Nomothetic benchmarks

A, B, Cs of Jacobson definitions. General stuff about limitations would go on the main concept page. Here it is focused on the client -- what are the benchmarks they will focus on? How explained to them?

Minimum important difference (MID)

Minimum important different (MID) is another way of looking at whether treatment is starting to help Arlene. Streiner, Norman, & Cairney (2015) offer a rule of thumb that changes of a half standard deviation (Cohen's d ~ .5) are usually large enough that patients notice and agree that they are showing improvement. If we picked one of the Achenbach scales with T-scores as an outcome measure, then we would be looking to see whether Arlene's score dropped by at least 5 points after some treatment. In general, MID values are often smaller than the RCI thresholds for the same measure, meaning that it will be easier to show improvement using them.

Client goals & tracking

These would be personal goals and idiographic measurement -- YTOPS, etc.

Process measures

This would be traces such as coming to sessions, doing homework assignments. (Not sure of other specifics involved in current IPT protocols?)

Progress measures

YTOPS again and goal setting.

Termination planning and maintenance

Revisit Jacobson benchmarks. Is there much chance of relapse? What things would the client need to pay attention to if they were going to nip that in the bud?